CN118625143A - A method, system, electronic device and medium for predicting SOH and RUL of energy storage battery - Google Patents

Abstract

A method, system, electronic device and medium for predicting the SOH and RUL of an energy storage battery, relating to the technical field of energy storage battery detection. The method comprises: obtaining the battery parameters of the energy storage battery and the driving parameters of the electric vehicle corresponding to the energy storage battery; determining the driving behavior characteristics of the electric vehicle based on the driving parameters, and determining the battery aging characteristics of the energy storage battery based on the battery parameters; aligning the timestamps of the driving behavior characteristics and the battery aging characteristics, and constructing a battery health feature vector of the energy storage battery within a preset period; inputting the battery health feature vector into a training model to obtain a SOH prediction model of the energy storage battery, and predicting the current battery health value of the energy storage battery based on the SOH prediction model; predicting the current remaining battery life of the energy storage battery based on the current battery health value, accumulated usage time and standard battery life of the energy storage battery. The implementation of the technical solution provided in the present application achieves the effect of improving the accuracy of predicting the SOH and RUL of the energy storage battery.

Description

A method, system, electronic device and medium for predicting SOH and RUL of energy storage battery

Technical Field

The present application relates to the technical field of energy storage battery detection, and in particular to a method, system, electronic device and medium for predicting the SOH and RUL of an energy storage battery.

Background Art

With the transformation of the global energy structure and the improvement of environmental protection requirements, the market demand for electric vehicles as a clean energy means of transportation continues to grow. One of the core components of electric vehicles is the energy storage battery, whose performance directly affects the vehicle's mileage, safety and cost of use. Therefore, it is crucial to ensure that the battery maintains a good performance state during its service life. The state of health (SOH) and remaining useful life (RUL) of the battery are important indicators for evaluating battery performance and safety. SOH reflects the current health status of the battery relative to a new battery, while RUL predicts the battery's usable state in the future and its possible termination time. For electric vehicle manufacturers and users, accurately predicting the battery's SOH and RUL can not only optimize the use of the battery and extend its service life, but also reduce maintenance costs and improve vehicle reliability and user trust.

At present, the existing methods for predicting the SOH and RUL of energy storage batteries analyze the health status of energy storage batteries by detecting the data of energy storage batteries, so as to achieve the purpose of predicting the SOH and RUL of energy storage batteries. However, in actual applications, affected by the driving habits of users, the degree of aging of energy storage batteries of electric vehicles varies under different driving conditions. Predicting the health status of batteries only by detecting the data of energy storage batteries themselves is often different from the actual health status of energy storage batteries, resulting in low accuracy in predicting the SOH and RUL of energy storage batteries.

Summary of the invention

The present application provides a method, system, electronic device and medium for predicting the SOH and RUL of an energy storage battery, which has the effect of improving the accuracy of predicting the SOH and RUL of the energy storage battery.

In a first aspect, the present application provides a method for predicting SOH and RUL of an energy storage battery, comprising:

Acquiring battery parameters of an energy storage battery and driving parameters of an electric vehicle corresponding to the energy storage battery;

Based on the driving parameters, determining the driving behavior characteristics of the electric vehicle, and based on the battery parameters, determining the battery aging characteristics of the energy storage battery;

Aligning the timestamps of the driving behavior feature and the battery aging feature within a preset period, and constructing a battery health feature vector of the energy storage battery within the preset period according to the driving behavior feature and the battery aging feature;

Inputting the battery health feature vector into a training model to obtain a SOH prediction model of the energy storage battery, and predicting a current battery health value of the energy storage battery based on the SOH prediction model;

The current remaining battery life of the energy storage battery is predicted based on the current battery health value, accumulated usage time and standard battery life of the energy storage battery.

By adopting the above technical solution, the battery parameters and vehicle driving parameters are obtained, and then the driving behavior characteristics are determined based on the driving parameters, and the battery aging characteristics are determined based on the battery operation data. Then the driving behavior characteristics and battery aging characteristics are aligned in the time dimension to construct a comprehensive feature vector describing the battery health status. The feature vector is input into the training model for training to obtain the SOH prediction model, and based on this, the current SOH of the energy storage battery, that is, the battery health value, is predicted, and the current RUL, that is, the remaining battery life, is predicted in combination with the predicted SOH, the cumulative usage time and the standard battery life. The battery parameters of the battery itself that affect battery aging are integrated with the driving parameters of the external electric vehicle corresponding to the energy storage battery, and the impact of driving habits and other usage conditions on battery attenuation is considered in multiple dimensions, thereby improving the accuracy of SOH and RUL predictions.

Optionally, the current battery health value, accumulated usage time and standard battery life of the energy storage battery are substituted into a preset battery life prediction formula to obtain the current remaining battery life of the energy storage battery; wherein the preset battery life prediction formula is:

In the formula, R represents the current remaining battery life of the energy storage battery, T _life represents the standard life of the energy storage battery, SOH represents the current battery health value of the energy storage battery, f _fast represents the aging factor of the energy storage battery in the fast charging mode, p _fast represents the fast charging ratio of the energy storage battery, f _slow represents the aging factor of the energy storage battery in the slow charging mode, p _slow represents the slow charging ratio of the energy storage battery, and T _used represents the current accumulated usage time of the energy storage battery.

By adopting the above technical solution, when predicting the remaining battery life, a preset battery life prediction formula is used, and the summarized battery health value, cumulative usage time and standard life data are used as model inputs, and the remaining available time of the battery is automatically output based on the formula calculation. The standard formula algorithm can achieve efficient and automated prediction of the remaining life of large quantities of batteries, combining multiple influencing factors, and the calculation results are more accurate and reliable.

Optionally, based on the driving duration, braking frequency and driving speed in the driving parameters, multiple driving behaviors of the electric vehicle are determined, and the driving behaviors include emergency braking behavior, high-speed driving behavior and long-distance driving behavior; based on the sampling period of each driving behavior and the driving parameters, a driving behavior time series of the electric vehicle is generated, and the driving behavior time series is used as the driving behavior feature.

By adopting the above technical solution, a variety of specific driving behavior types, including emergency braking, high-speed driving, and long-distance driving, are determined based on driving parameters such as driving duration, braking frequency, and driving speed, which fully reflects all aspects of driving habits. And based on the sampling period, these driving behaviors generate time series as features. This time series feature of multiple types of driving behaviors can more comprehensively describe the battery usage environment and loss process. Based on these diverse driving behavior time series features, a battery health assessment model that is adaptable to various complex power usage environments can be trained to improve the accuracy of battery status prediction.

Optionally, if the driving duration is greater than or equal to a preset duration, the long-distance driving behavior is determined as the driving behavior of the electric vehicle; if the braking frequency is greater than or equal to a preset frequency, the emergency braking behavior is determined as the driving behavior of the electric vehicle; if the driving speed is greater than or equal to a preset speed, the high-speed driving behavior is determined as the driving behavior of the electric vehicle.

By adopting the above technical solution, preset thresholds for driving time, braking frequency and driving speed are set, and different types of driving behaviors are determined based on threshold judgment, including long-distance driving, emergency braking and high-speed driving. This threshold-based judgment method can clearly distinguish different driving characteristics, simplify the complex driving process into specific behavior categories, facilitate feature extraction and modeling, and make battery health assessment more accurate.

Optionally, determine the sampling time of the battery parameters; determine the voltage change rate of the energy storage battery based on the sampling time and the voltage; determine the internal resistance increase rate of the energy storage battery based on the sampling time and the internal resistance; determine the capacitance decay rate of the energy storage battery based on the sampling time and the capacitance; and use the voltage change rate, the internal resistance increase rate and the capacitance decay rate as battery aging characteristics of the energy storage battery.

By adopting the above technical solution, when extracting battery aging characteristics, the sampling time of the battery parameters is first determined, and then the voltage change rate, internal resistance increase rate and capacitance decay rate are calculated based on the sampling time and the three key parameters of voltage, internal resistance and capacitance, and these three indicators are used as comprehensive features to describe the degree of battery aging. The innovation of this feature extraction method lies in the introduction of the sampling time of the time dimension, which can more reasonably reflect the changing trend of battery parameters over time; at the same time, using the parameter change rate as a feature can eliminate the influence of the inherent differences of different batteries.

Optionally, the preset period is divided into multiple standard time periods; a feature matrix of each standard time period is generated based on driving behavior characteristics and battery aging characteristics within each standard time period; features of each row in each feature matrix are concatenated in chronological order to obtain a feature vector of each standard time period; and a battery health feature vector of the energy storage battery within the preset period is determined based on the feature vector of each standard time period.

By adopting the above technical solution, the time evolution law of the battery health state can be well captured through time segmentation. Combining driving behavior and battery aging characteristics in each time period can comprehensively describe the internal and external factors affecting battery degradation in this time period. The process of splicing time period feature vectors ensures that the feature vectors can reflect the temporal causal relationship and dynamic change trend. The resulting comprehensive battery health feature vector can accurately reflect the changes in the battery health state during the entire evaluation cycle, laying a data foundation for subsequent SOH and RUL predictions.

Optionally, the battery health feature vector is divided into training set data and validation set data according to a preset ratio, and the training model is trained based on the training set data and the validation set data until a preset iteration termination condition is reached, wherein the preset iteration termination condition is that the number of iterations reaches a preset threshold or the loss function of the training model converges; the training model that reaches the preset iteration termination condition is determined as the SOH prediction model.

By adopting the above technical solution, the constructed battery health feature vector is divided into training set data and validation set data according to a preset ratio. Then the prediction model is trained based on the training set data, and the validation set data is used to evaluate the generalization ability of the model on unseen data, and the training is stopped according to the preset iteration termination condition (which can be the number of iterations reaching a threshold or the model loss function convergence), and finally the model parameters when the termination condition is reached are determined as the final SOH prediction model. This alternating iterative training strategy based on training sets and validation sets can greatly improve the generalization performance of the model and prevent overfitting; the validation set plays a role in monitoring the performance of the model and provides a basis for stopping training early; using the convergence of the loss function as one of the termination conditions can ensure that the model achieves excellent convergence performance.

In a second aspect of the present application, a SOH and RUL prediction system for an energy storage battery is provided, the system comprising:

A parameter acquisition module, used to acquire battery parameters of the energy storage battery and driving parameters of the electric vehicle corresponding to the energy storage battery; a feature determination module, used to determine the driving behavior characteristics of the electric vehicle based on the driving parameters, and to determine the battery aging characteristics of the energy storage battery based on the battery parameters;

An SOH prediction module is used to align the timestamps of the driving behavior characteristics and the battery aging characteristics within a preset period, and construct a battery health feature vector of the energy storage battery within the preset period according to the driving behavior characteristics and the battery aging characteristics; input the battery health feature vector into a training model to obtain an SOH prediction model of the energy storage battery, and predict the current battery health value of the energy storage battery based on the SOH prediction model;

The RUL prediction module is used to predict the current remaining battery life of the energy storage battery according to the current battery health value, accumulated usage time and standard battery life of the energy storage battery.

In a third aspect of the present application, an electronic device is provided, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can implement a method for predicting the SOH and RUL of an energy storage battery when loaded and executed by the processor.

In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements a method for predicting SOH and RUL of an energy storage battery.

In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

By adopting the technical solution of the present application, the battery parameters and vehicle driving parameters are obtained, and then the driving behavior characteristics are determined based on the driving parameters, and the battery aging characteristics are determined based on the battery operation data. Then, the driving behavior characteristics and the battery aging characteristics are aligned in the time dimension to construct a comprehensive feature vector that describes the battery health status. The feature vector is input into the training model for training to obtain the SOH prediction model, and based on this, the current SOH of the energy storage battery, that is, the battery health value, is predicted, and the current RUL, that is, the remaining battery life, is predicted in combination with the predicted SOH, the cumulative usage time and the standard battery life. The battery parameters of the battery itself that affect battery aging are integrated with the driving parameters of the external electric vehicle corresponding to the energy storage battery, and the impact of driving habits and other usage conditions on battery attenuation is considered in multiple dimensions, thereby improving the accuracy of SOH and RUL predictions.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic flow chart of a method for predicting SOH and RUL of an energy storage battery provided in an embodiment of the present application;

FIG2 is a schematic diagram of the structure of a SOH and RUL prediction system for an energy storage battery disclosed in an embodiment of the present application;

FIG. 3 is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application.

Description of reference numerals: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory.

DETAILED DESCRIPTION

In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

An embodiment of the present application provides a method for predicting the SOH and RUL of an energy storage battery. In one embodiment, please refer to Figure 1, which is a flow chart of the method for predicting the SOH and RUL of an energy storage battery provided in an embodiment of the present application. The method can be implemented by a computer program, which can be integrated into an application or run as an independent tool application. The method can also be implemented by a single-chip microcomputer, or it can be run on a SOH and RUL prediction system for an energy storage battery based on a von Neumann system. Specifically, the method may include the following steps:

Step 101: Obtain battery parameters of the energy storage battery and driving parameters of the electric vehicle corresponding to the energy storage battery.

Among them, battery parameters refer to data that directly reflects the working status of the battery. In the embodiment of the present application, they can be understood as the physical and chemical properties of the battery such as voltage, internal resistance and capacitance, which are used to reflect the health status and electrochemical process of the battery.

Driving parameters refer to data reflecting the use and driving conditions of electric vehicles. In the embodiment of the present application, they can be understood as parameters such as driving distance, driving time, number of brakes, average speed, etc., which are used to reflect the impact of the user's driving habits on the battery.

Specifically, in order to accurately predict the state of health (SOH) and remaining useful life (RUL) of the energy storage battery, it is necessary to consider the impact of the user's actual driving conditions on the battery. Obtain the battery parameters of the energy storage battery, including voltage, current, temperature, etc. These parameters directly reflect the working state of the battery, and obtain the driving parameters of the corresponding electric vehicle, including driving distance, driving time, number of brakes, average speed and other parameters. Driving parameters can be collected by on-board sensors, which reflect the user's driving habits. The purpose of obtaining the parameters and driving parameters of the energy storage battery is to establish a battery health feature vector for the subsequent establishment, and to fully consider the impact of the battery's own factors and usage factors on the battery health status. For example, frequent high-speed acceleration will accelerate the degradation of the battery; long-term high temperature conditions will also reduce the battery capacity. Comprehensive analysis of the two types of parameters can make the prediction results more accurately reflect the actual health status of the battery.

Step 102: Determine the driving behavior characteristics of the electric vehicle based on the driving parameters, and determine the battery aging characteristics of the energy storage battery based on the battery parameters.

Among them, driving behavior characteristics refer to characteristic parameters that reflect the driver's driving habits. In the embodiment of the present application, they can be understood as emergency braking behavior, high-speed driving behavior, and long-distance driving behavior, etc., which are used to reflect the impact of the driver's driving style on the battery health status.

Battery aging characteristics refer to characteristic parameters that reflect the attenuation of battery performance. In the embodiments of the present application, they can be understood as voltage attenuation rate, internal resistance growth rate, capacitance attenuation rate, etc., which are used to quantitatively describe the decline in battery capacity and power performance.

Specifically, on the basis of obtaining the energy storage battery parameters and driving parameters, the battery health characteristics are further extracted. The purpose of this step is to obtain characteristics that can fully reflect the battery health status and lay the foundation for constructing the battery health feature vector. According to the driving time, braking frequency and average speed in the driving parameters, the driving behavior characteristics of the electric vehicle can be determined, mainly including: emergency braking behavior, high-speed driving behavior and long-distance driving behavior. The extraction method of these behavior characteristics can pre-set the corresponding judgment conditions and thresholds, and the corresponding behavior characteristics can be determined when the driving parameters meet the judgment conditions. For example, when the braking frequency exceeds a certain number of times/kilometer, it is determined as an emergency braking behavior, which can reflect the impact of the user's actual use on the battery. And according to the battery operation data of the energy storage battery, such as the time series data of voltage, internal resistance, and capacitance, the change rate of these parameters can be calculated to determine the battery aging characteristics, such as the decay rate of voltage, the growth rate of internal resistance, and the decay rate of capacitance, which can well quantify the health decay state of the battery.

Based on the above embodiment, as an optional embodiment, in step 102: determining the driving behavior characteristics of the electric vehicle based on the driving parameters, this step may also include the following steps:

Step 201: Determine multiple driving behaviors of the electric vehicle according to driving duration, braking frequency and driving speed in driving parameters, wherein the driving behaviors include emergency braking behavior, high-speed driving behavior and long-distance driving behavior.

Among them, driving time refers to the length of a single driving trip, which is used to judge the impact of long-term driving. Braking frequency refers to the number of braking times per unit of driving mileage, which is used to judge the impact of frequent starts and stops on urban roads. Driving speed refers to the average speed of the vehicle, which is used to judge the impact of high-speed driving.

Driving behavior refers to the way the driver operates during driving. In the embodiments of the present application, it can be understood as emergency braking behavior, high-speed driving behavior, and long-distance driving behavior, which is used to reflect the impact of the driver's driving habits on the battery health status. These driving behaviors can be obtained by judging driving parameters. For example, emergency braking behavior reflects the behavioral habit of frequent braking under urban road conditions. High-speed driving behavior reflects the behavioral habit of long-term high-speed driving. Long-distance driving behavior reflects the behavioral habit of long-distance driving.

Specifically, in order to fully consider the impact of the user's driving habits on battery health, it is necessary to determine the driving behavior characteristics based on the driving parameters. The purpose of this step is to extract behavioral characteristics that have an important impact on the battery status from the driving parameters. Different driving behaviors are judged based on the three driving parameters of driving time, braking frequency and average speed. If the driving time exceeds the preset threshold, such as more than 2 hours, it can be judged as long-distance driving behavior. Long-distance driving will consume more power and increase the burden on the battery. If the number of braking times per unit mileage exceeds the preset value, such as more than 20 times/kilometer, it can be judged as emergency braking behavior. Frequent braking reflects urban road conditions and also increases battery mechanical fatigue. If the average vehicle speed exceeds the preset speed threshold, such as more than 80 kilometers/hour, it can be judged as high-speed driving behavior. High-speed driving requires greater power and accelerates battery performance attenuation. According to these behavioral characteristics, the working state of the battery under different working conditions can be reflected.

Based on the above embodiment, as an optional embodiment, in step 201: determining multiple driving behaviors of the electric vehicle according to the driving time, braking frequency and driving speed in the driving parameters, this step may also include the following steps:

Step 211: If the driving time is greater than or equal to the preset time, the long-distance driving behavior is determined as the driving behavior of the electric vehicle.

Specifically, in order to accurately judge long-distance driving behavior, it is necessary to judge based on the driving time and the preset time threshold. The purpose of judging long-distance driving behavior is to identify the battery effect caused by long-term driving. Long-term driving consumes more power and increases the chemical degradation of the battery. Obtain the driving time data in the driving parameters, and then compare it with the preset time threshold. The preset time threshold can be set according to the actual situation, for example, it is set to 2 hours. If the duration of a detected driving time is greater than or equal to 2 hours, it can be determined that the driving process is a long-distance driving behavior. The final long-distance driving behavior identifier will be used as part of the driving behavior characteristics, reflecting the impact of long-term operation of the battery. Combined with the battery's own aging characteristics, it can evaluate the impact of such long-distance driving on the battery health status and improve the accuracy of SOH and RUL predictions.

Step 221: If the braking frequency is greater than or equal to the preset frequency, the emergency braking behavior is determined as the driving behavior of the electric vehicle.

Specifically, in order to accurately judge emergency braking behavior, it is necessary to judge based on the braking frequency and the preset frequency threshold. The purpose of judging emergency braking behavior is to identify the battery effect caused by frequent braking. Frequent braking increases the number of charge and discharge cycles of the battery, and also brings greater mechanical shock to the battery. Count the number of brakes within a unit mileage, calculate the braking frequency, and then compare it with the preset frequency threshold. The preset frequency threshold can be set according to actual conditions, for example, set to 20 brakes per kilometer. If the detected braking frequency is greater than or equal to 20 times/kilometer, it can be judged as an emergency braking behavior. The final emergency braking mark will be used as part of the driving behavior characteristics to reflect the impact of frequent braking on the battery.

Step 231: If the driving speed is greater than or equal to the preset speed, the high-speed driving behavior is determined as the driving behavior of the electric vehicle.

Specifically, in order to accurately judge high-speed driving behavior, it is necessary to make a judgment based on the driving speed and the preset speed threshold. The purpose of judging high-speed driving behavior is to identify the battery effect caused by long-term high-speed driving. High-speed driving requires the battery to continuously output a large power, which increases the burden on the battery and shortens the battery life. The average vehicle speed data in the driving parameters is obtained, and then compared with the preset speed threshold. The preset speed threshold can be set according to actual conditions, for example, set to 80 kilometers per hour. If the detected average speed is greater than or equal to 80 kilometers per hour, it can be judged that the driving process is a high-speed driving behavior. The high-speed driving behavior identifier finally obtained will be used as part of the driving behavior characteristics to reflect the impact of high-speed driving on the battery.

Step 202: Generate a driving behavior time series of the electric vehicle according to the sampling period of each driving behavior and driving parameter, and use the driving behavior time series as a driving behavior feature.

Among them, the sampling period refers to the time interval for collecting driving parameters. In the embodiment of the present application, it can be understood as a time period for collecting driving parameters determined according to a fixed time length (for example, every 5 minutes) or a fixed driving distance (for example, every 10 kilometers), which is used to generate a driving behavior time series to reflect the changes in driving behavior over time.

The driving behavior time series refers to a driving behavior data sequence arranged in chronological order. In the embodiment of the present application, it can be understood as a driving behavior sequence formed by connecting the driving behaviors identified according to the sampling period in chronological order, which is used to reflect the impact of the driver's long-term driving habits on the battery.

Specifically, after determining the driving behavior characteristics of electric vehicles, it is necessary to generate a time series of driving behaviors in order to subsequently construct a healthy feature vector for the battery. The purpose of generating a driving behavior time series is to record the changes of different driving behaviors over time and reflect the user's long-term driving habits. It is based on the different driving behaviors identified in the previous steps and the sampling period sequence of the driving parameters to generate the driving behaviors corresponding to each time period. For example, if high-speed driving behavior is detected in time period 1 and emergency braking behavior is detected in time period 2, the following driving behavior time series can be constructed: [Time period 1: high-speed driving; Time period 2: emergency braking; ...] Among them, the time period can be set to a fixed time interval, such as a time period every 5 minutes, or it can be set to a fixed mileage segment, such as a time period every 10 kilometers. By generating a driving behavior time series, the impact of the user's driving habits on the battery can be reflected from a long-term dimension, as an important part of the battery health feature vector.

Based on the above embodiment, as an optional embodiment, in step 102: determining the battery aging characteristics of the energy storage battery based on the battery parameters, this step may also include the following steps:

Step 203: Determine the sampling duration of the battery parameters.

Among them, the sampling duration of battery parameters refers to the time span for collecting battery operating parameter data. In the embodiment of the present application, it can be understood as a period of time covering the main service life cycle of the battery, such as 3 years, which is used to obtain time series data such as voltage, current, temperature, etc. that can fully reflect the battery aging process.

Specifically, after obtaining the operating parameters of the battery, it is necessary to determine a reasonable battery parameter sampling time to extract the aging characteristics of the battery. The purpose of determining the battery parameter sampling time is to obtain sufficient data that can fully reflect the battery aging process. The sampling time needs to be set neither too short nor too long. A relatively reasonable time range can be selected based on the expected service life of the battery. For example, if the expected life cycle of the battery is 5 years, the sampling time can be set to 3 years to ensure that the aging process of the battery in the middle and late stages is covered. The efficiency of data storage and computing processing also needs to be considered. Too long a time will generate a large amount of redundant data, and too short a time cannot observe the full attenuation of battery performance. For example, the sampling time of the battery parameters is set to 3 years. During this period, parameters such as voltage, current, temperature, and internal resistance are collected as basic data for extracting battery aging characteristics.

Step 204: Determine the voltage change rate of the energy storage battery according to the sampling time and voltage; determine the internal resistance increase rate of the energy storage battery according to the sampling time and internal resistance; determine the capacitance attenuation rate of the energy storage battery according to the sampling time and capacitance.

The voltage change rate refers to the quantitative ratio of the battery voltage change over time. In the embodiment of the present application, it can be understood as the average attenuation rate of the battery voltage value within the sampling period, which is used to quantitatively describe the degree of degradation of the battery power supply performance.

The internal resistance increase rate refers to the ratio of the battery's internal resistance to change over time. In the embodiment of the present application, it can be understood as the average growth rate of the battery's internal resistance value within the sampling period, which is used to quantitatively describe the degree of decrease in the battery's chemical activity.

The capacitance decay rate refers to the ratio of the change of battery capacitance over time. In the embodiment of the present application, it can be understood as the average decay rate of the battery capacitance value within the sampling period, which is used to quantitatively describe the degree of degradation of the battery energy storage performance.

Specifically, after obtaining the time series data of the battery parameters, it is necessary to determine the aging characteristics of the battery based on these data. The purpose of determining the voltage change rate, internal resistance increase rate and capacitance decay rate is to extract characteristic parameters that can quantitatively describe the degree of battery aging. The voltage, internal resistance and capacitance time series data of the battery are collected within the set battery parameter sampling time, and then the average voltage decay rate, the average internal resistance growth rate, and the average capacitance decay rate are calculated. The calculation method is: voltage change rate = (initial voltage-final voltage)/sampling time, internal resistance increase rate = (final internal resistance-initial internal resistance)/sampling time, capacitance decay rate = (initial capacitance-final capacitance)/sampling time. The change rate of these parameters can intuitively reflect the degree of attenuation of battery performance, so it is determined as the aging characteristics of the battery. The obtained aging characteristics will be combined with the user behavior characteristics to form a characteristic vector of the battery health state, which is used to evaluate the battery's SOH and remaining life, and to achieve accurate monitoring and prediction of battery health.

Step 205: The voltage change rate, the internal resistance increase rate, and the capacitance decay rate are used as battery aging characteristics of the energy storage battery.

Specifically, after obtaining the voltage change rate, internal resistance increase rate and capacitance decay rate, they are integrated into the battery aging characteristics. The purpose of determining these parameters as battery aging characteristics is to obtain quantitative indicators that can reflect the overall health status of the battery. The voltage change rate reflects the battery's power supply performance decay, the internal resistance increase rate reflects the battery's chemical activity decay, and the capacitance decay rate reflects the battery's energy storage performance decay. These three parameters can measure the battery's health from different aspects. Combining them into a feature vector can comprehensively reflect the battery's aging condition.

Step 103: aligning the timestamps of the driving behavior characteristics and the battery aging characteristics within a preset period, and constructing a battery health feature vector of the energy storage battery within the preset period based on the driving behavior characteristics and the battery aging characteristics.

Among them, the timestamp refers to an identifier that records the time when an event occurs. In the embodiment of the present application, it can be understood as indicating the specific time point when the driving behavior characteristics and the battery aging characteristics are detected, which is used to align the two within the same time period to construct a battery health feature vector.

The battery health feature vector refers to a comprehensive feature expression that includes the battery usage environment information and its own status information. In the embodiment of the present application, it can be understood as a combination of driving behavior characteristics and battery aging characteristics aligned in the same time period, which is used to comprehensively reflect the health status of the battery and serve as the input of the battery health assessment and life prediction model.

The preset period refers to the time range for extracting the battery health feature vector, which can be understood as a fixed time span in the embodiment of the present application, such as 1 month, which is used to collect and align driving behavior characteristics and battery aging characteristics within this period to construct a feature vector that reflects the health status of the battery at this stage.

Specifically, after obtaining the driving behavior characteristics and battery aging characteristics, the two characteristics need to be aligned in the same time period to construct the battery health feature vector. The purpose of time alignment is to make the driving behavior characteristics and battery aging characteristics correspond one to one, reflecting the battery usage environment and health status in the same time period. A fixed period can be set in advance, such as 1 month. Then, in this period, the driving behavior time series and the battery aging parameter time series are extracted respectively to ensure that the timestamps of the two are aligned. Taking January 1 to January 31 as a period, the driving behavior characteristics contain the characteristics of 31 time periods, and the battery aging characteristics also contain the characteristic values of each day of the month. In the same time period, the driving behavior characteristics and battery aging characteristics are combined to form the battery health feature vector in the period, for example: [January 1 behavior, January 1 aging characteristics; ...; January 31 behavior, January 31 aging characteristics]. The health feature vector contains information about the battery usage environment and its own status, which can fully reflect the health of the battery. This vector is used as input for SOH evaluation and RUL prediction models, which can improve the accuracy of the results.

Based on the above embodiment, as an optional embodiment, in step 103: constructing a battery health feature vector of the energy storage battery within a preset period according to the driving behavior characteristics and the battery aging characteristics, this step may also include the following steps:

Step 301: Divide a preset cycle into a plurality of standard time periods; generate a feature matrix for each standard time period according to driving behavior characteristics and battery aging characteristics within each standard time period.

Among them, the standard time period refers to the sub-time interval after the preset period is evenly divided. In the embodiment of the present application, it can be understood as a daily time period divided into equal intervals of a month, which is used to extract battery health features in each time period and construct a normalized feature matrix to provide structured samples for the training of the health assessment model.

The feature matrix refers to a matrix expression after the battery health characteristics are organized in a unified format. In the embodiment of the present application, it can be understood as a matrix formed by arranging the driving behavior characteristics and battery aging characteristics of each standard time period in a fixed order, which is used as a training sample for the health assessment model to facilitate the model to learn the changes in the health status of the battery in different time periods.

Specifically, after obtaining the battery health feature vector of the preset period, it is necessary to further construct a feature matrix of the standard time period. The purpose of dividing the preset period into standard time periods is to obtain a regular and comparable feature representation and provide structured samples for model training. The preset period of one month can be divided into multiple standard time periods at equal intervals, such as one time period per day. Then, in each standard time period, the driving behavior characteristics and battery aging characteristics of the time period are extracted and arranged in a fixed order to form a feature matrix. In this way, a regular matrix sample can be obtained. Each row represents the battery health characteristics of a time period, and a standardized feature matrix is obtained, which can be conveniently used as a training sample for the model to improve the training effect. At the same time, it is also convenient for comparative analysis of battery health characteristics in different time periods. The division of standard time periods and the generation of feature matrices provide effective samples for the training of health assessment and prediction models, which can improve the accuracy of battery status detection.

Step 302: In chronological order, the features of each row in each feature matrix are concatenated to obtain a feature vector for each standard time period.

Among them, the time sequence refers to the order in which events or features are arranged according to the time of occurrence. In the embodiment of the present application, it can be understood as the time sequence relationship between the rows in the standard time period feature matrix, which is used to splice the feature vectors according to this time sequence, so that the feature vectors retain dynamic time information and can reflect the process of battery health status changing over time.

The features of each row in the feature matrix refer to the combination of driving behavior features and battery aging features within the same standard time period. In the embodiment of the present application, it can be understood as a sequence of feature items representing a standard time period in the feature matrix, which is used to splice in chronological order to construct a long sequence feature vector containing the complete features of each time period.

A feature vector refers to a one-dimensional vector expression that integrates multiple features into a fixed order. In the embodiment of the present application, it can be understood as obtaining a one-dimensional feature sequence containing complete time period information by splicing each row of the feature matrix of the standard time period, which is used to provide a suitable sample input form for the battery health assessment and prediction model based on deep learning.

Specifically, after obtaining the feature matrix of the standard time period, it is necessary to further construct a feature vector containing complete time information. For this purpose, it is a reasonable method to splice each row of the matrix in time order. The purpose of this is to obtain a long sequence feature vector containing dynamic time information, which can reflect the trajectory of the battery health status changing over time. This vector not only retains the integrity of the features of each time period, but also reflects the order of time. It is a suitable sample form for the model to learn the battery aging law. According to the time labels of each row of the feature matrix, its time order is confirmed, and then the features are spliced row by row in time order. The driving behavior features and battery aging features of the same time period are combined to construct the feature expression of the time period. Each row is spliced in turn until the traversal is completed, and finally a long sequence vector containing the features of the entire time period is formed. In this way, through the time sequence splicing of the matrix row features, the feature vector obtained not only maintains the health status characteristics of the battery in each standard time period, but also reflects that battery aging is a time-dynamic process.

Step 303: Determine the battery health feature vector of the energy storage battery within a preset period according to the feature vector of each standard time period.

Specifically, after obtaining the feature vector of the standard time period, it is necessary to further determine the battery health feature vector that reflects the entire preset cycle. The purpose is to obtain a feature expression that contains complete information of the preset cycle, which can fully reflect the changes in the battery health status during this stage as the input of the battery health assessment. Traverse the feature vectors constructed for each standard time period, combine these feature vectors in chronological order, and construct a long sequence feature vector containing all standard time periods as the final battery health feature vector. This feature vector encompasses the health status and change trajectory of the battery at different time periods in the preset cycle, which can fully reflect the overall health characteristics of the battery in this cycle.

Step 104: Input the battery health feature vector into the training model to obtain the SOH prediction model of the energy storage battery, and predict the current battery health value of the energy storage battery based on the SOH prediction model.

Among them, the training model refers to a prediction model established through a machine learning algorithm. In the embodiment of the present application, it can be understood as a data-driven prediction model based on a battery health feature vector, which is used to learn the laws of changes in the battery health state and realize intelligent evaluation and prediction of the battery health indicator SOH.

The SOH prediction model refers to a machine learning model used to predict the battery state of health. In the embodiment of the present application, it can be understood as an LSTM network model trained based on the battery health feature vector, which is used to predict the SOH value of the battery in the future period and evaluate the battery health state.

The battery health value refers to an indicator for evaluating the overall health status of the battery. In the embodiment of the present application, it can be understood as the predicted battery SOH value, which is used to determine the health of the battery and the remaining life.

Specifically, after constructing a vector that fully reflects the battery health characteristics, it is necessary to further train the prediction model based on this vector to realize the evaluation of the battery health status. The purpose is to realize the intelligent prediction of the battery health value SOH through the machine learning model and provide a data-driven battery health management solution. The constructed battery health feature vector is used as a sample input and trained using sequential models such as LSTM. The model can learn the battery health evolution law through the feature vector. The trained model contains both the battery usage environment information and the timing pattern of battery aging. In actual prediction, the battery health feature vector of the latest cycle is monitored and input into the trained SOH prediction model. The model can give the battery health value of the battery at the current moment, that is, the SOH of the energy storage battery, reflecting the health status of the battery.

Based on the above embodiment, as an optional embodiment, in step 104: inputting the battery health feature vector into the training model to obtain the SOH prediction model of the energy storage battery, this step may also include the following steps:

Step 401: Divide the battery health feature vector into training set data and validation set data according to a preset ratio, and train the training model based on the training set data and the validation set data until a preset iteration termination condition is reached. The preset iteration termination condition is that the number of iterations reaches a preset threshold or the loss function of the training model converges; the training model that reaches the preset iteration termination condition is determined as the SOH prediction model.

Specifically, the loss function of the training model refers to an indicator that measures the fitting effect of the model on the training data. In the embodiment of the present application, it can be understood as the error calculated by the machine learning model for the training data during the iteration process.

The number of iterations refers to the number of times the training data is gradient updated and optimized during the model training process. In the embodiments of the present application, it can be understood as the number of times the model parameters are repeatedly adjusted using the training data set when training the model. It is used to control the number of rounds of model training to achieve a balance between training effect and computational cost.

Specifically, after obtaining the vector data reflecting the battery health characteristics, it is necessary to obtain a prediction model with strong generalization by dividing the training data and the verification data and iterative training. The purpose is to fit the law of battery health changes based on the training data, and to improve the model's prediction ability for new data through verification data, so as to obtain a battery health assessment model that has learned both characteristic patterns and generalization capabilities. The feature vector is divided into a training set and a verification set at a ratio of, for example, 8:2. Then load the machine learning model, iteratively train the model with the training data, and continuously optimize the battery health evolution pattern in the learning feature vector. At the same time, the verification set data is used to evaluate the prediction effect of the model on new data during the iteration process. When the loss function converges or reaches the preset number of iterations, the training is stopped. At this time, a final model that fits the training data well and has strong generalization capabilities for new data is obtained. In this way, through the iterative model optimization of the training verification method, a battery health assessment model that represents the training data and has generalization capabilities can be obtained.

Step 105: predict the current remaining battery life of the energy storage battery according to the current battery health value, accumulated usage time and standard battery life of the energy storage battery.

Among them, the cumulative usage time refers to the total usage time of the battery from leaving the factory to the current moment. In the embodiment of the present application, it can be understood as the total operating time accumulated in chronological order after the battery is put into use. It is used to reflect the battery usage history, judge the battery consumption level, and predict the remaining life of the battery.

The standard battery life refers to the total expected duration of a battery from the time it leaves the factory to the time it is scrapped under normal conditions of use. In the embodiments of the present application, it can be understood as the normal battery use cycle time specified in the battery product technical specifications.

The remaining battery life refers to the expected remaining working time of the battery from the current moment until it can no longer be used normally. In the embodiment of the present application, it can be understood as the remaining time that the battery can be used from the current moment based on the battery health status assessment and the calculation of the used time, that is, the current RUL of the energy storage battery.

Specifically, in order to achieve an accurate assessment of the remaining battery life, it is necessary to comprehensively consider multiple factors such as the current health status, usage history, and standard life for predictive analysis. The purpose is to accurately judge the remaining available time of the battery, which can better guide the scientific use and planned maintenance of the battery. Obtain the SOH health status value of the battery at the current moment, which can reflect the degree of battery wear; then count the cumulative usage time of the battery to date to determine the battery consumption history; refer to the industrial standard full life cycle of the battery, and combine the SOH value to determine the proportion of the standard life that has been used. Through the pre-set formula, comprehensively consider the impact of various factors on the remaining life, and obtain the specific time when the battery can be reused from the current moment.

Based on the above embodiment, as an optional embodiment, in step 105: predicting the current remaining battery life of the energy storage battery according to the current battery health value, accumulated usage time and standard battery life of the energy storage battery, this step may also include the following steps:

Step 501: Substitute the current battery health value, accumulated usage time and standard battery life of the energy storage battery into a preset battery life prediction formula to obtain the current remaining battery life of the energy storage battery; wherein the preset battery life prediction formula is:

In the formula, R represents the remaining battery life of the energy storage battery, T _life represents the standard life of the energy storage battery, SOH represents the current battery health value of the energy storage battery, f _fast represents the aging factor of the energy storage battery in fast charging mode, p _fast represents the fast charging ratio of the energy storage battery, f _slow represents the aging factor of the energy storage battery in slow charging mode, p _slow represents the slow charging ratio of the energy storage battery, and T _used represents the current cumulative usage time of the energy storage battery.

Among them, the preset battery life prediction formula refers to a mathematical formula used to calculate the current remaining battery life of the energy storage battery. In the embodiment of the present application, the preset battery life prediction formula can be understood as a multivariate analysis model that includes factors such as the current battery health value, cumulative usage time, and standard battery life of the energy storage battery. The preset battery life prediction formula is used to quantitatively calculate and evaluate the current remaining battery life of the energy storage battery. By substituting actual test data, the current remaining battery life of the energy storage battery can be calculated.

Specifically, in order to evaluate the remaining available time of the battery, it is necessary to perform calculations based on a standard life prediction formula. The purpose is to use a pre-established mathematical formula, input data from the actual battery situation, and intelligently predict the remaining battery life of the battery. Obtain the SOH health status value obtained by monitoring the battery at the current moment, and at the same time count the cumulative working hours that the battery has been used, and prepare the industrial standard full life cycle data of the battery product. Then substitute the above three factors into the preset life prediction formula in turn, execute the algorithm calculation, and finally automatically obtain the RUL value of the battery from the current moment, that is, the remaining battery life. Compared with manual experience estimation, the use of a standard formula algorithm can achieve efficient and automated prediction of a large amount of remaining battery life. The result reflects the true health and consumption status of the battery, making the subsequent use plan of the battery more accurate and reliable.

The formula consists of three parts. The first part represents the standard battery life of the energy storage battery and the impact function of the battery health value on the remaining battery life. Among them, T _life represents the standard life of the energy storage battery, which is the total usage time that the battery is expected to achieve under ideal conditions. This is an estimate provided by the manufacturer based on the battery performance degradation under standard test conditions. SOH represents the current battery health value of the energy storage battery, which is obtained based on a comprehensive consideration of the driving behavior of the electric vehicle and the battery status. The calculation result of this part represents the ideal life of the battery in theory under the current health state of the battery. It is an adjustment of the expected life when the battery health state has not suffered any loss. If the health state of the battery decreases, that is, the SOH battery health value decreases, this means that the effective life of the battery is also reduced accordingly.

The second part represents the impact function of the charging mode of the energy storage battery on the remaining battery life, including fast charging mode and slow charging mode. f _fast represents the aging factor of the energy storage battery in the fast charging mode, that is, the proportional factor of the fast charging mode on the accelerated attenuation of the battery life. This factor describes the additional damage to the battery health caused by the fast charging method compared with the normal charging conditions. Due to the large current and high temperature, fast charging will accelerate the chemical aging process of the battery, thereby shortening the battery life. p _fast represents the proportion of time that the battery adopts the fast charging mode during the entire use period. This high proportion means that the battery experiences fast charging more frequently and is more likely to be negatively affected by the fast charging mode. f _slow represents the proportional factor of the slow charging mode on the accelerated attenuation of the battery life. Generally, this factor is low because the slow charging has a relatively small impact on the battery. The slow charging mode is generally considered to be more gentle on the battery and has less impact on battery overheating or other rapid attenuation. p _slow represents the proportion of time that the battery adopts the slow charging mode during the entire use period. A higher slow charging ratio usually helps maintain battery health and delay the battery aging process. This section combines the types of battery charging modes, i.e. fast charging and slow charging, and their usage frequency, i.e. the ratio, as well as the impact of these modes on battery life, i.e. the accelerated aging factor. The product of each item represents the specific contribution of the charging mode to the overall health and life of the battery. Fast charging usually has a greater impact on the overall health of the battery due to its higher aging factor.

The third part represents the influence function of the current cumulative usage time of the energy storage battery and the standard battery life on the remaining battery life. This part calculates the proportion of the remaining battery life, which means the proportion of the service life minus the proportion of the total expected life to obtain the proportion of the remaining life. If T _used is close to T _life , this ratio will approach 0, indicating that the battery is approaching the end of its expected life. On the contrary, if T _used is much less than T _life , this ratio will approach 1, which means that the battery still has most of its life to be used.

In summary, the formula _ffast × _pfast + _fslow × _pslow represents the total impact of charging mode on battery life, and The product of these two gives the adjusted remaining life proportion of the battery due to usage and charging mode, and combined with T _life × SOH, it represents the theoretical maximum life of the battery in its current health state. This preset battery life prediction formula can more accurately predict the remaining battery life, helping to manage and optimize battery usage.

2 , a SOH and RUL prediction system for an energy storage battery provided in an embodiment of the present application includes: a parameter acquisition module, a feature determination module, a SOH prediction module, and a RUL prediction module, wherein:

A parameter acquisition module, used to acquire battery parameters of the energy storage battery and driving parameters of the electric vehicle corresponding to the energy storage battery;

A characteristic determination module, for determining the driving behavior characteristics of the electric vehicle based on the driving parameters, and determining the battery aging characteristics of the energy storage battery based on the battery parameters;

The SOH prediction module is used to align the timestamps of the driving behavior characteristics and the battery aging characteristics within a preset period, and construct a battery health feature vector of the energy storage battery within a preset period based on the driving behavior characteristics and the battery aging characteristics; the battery health feature vector is input into the training model to obtain the SOH prediction model of the energy storage battery, and based on the SOH prediction model, the current battery health value of the energy storage battery is predicted;

The RUL prediction module is used to predict the current remaining battery life of the energy storage battery based on the current battery health value, cumulative usage time and standard battery life of the energy storage battery.

On the basis of the above embodiment, the RUL prediction module is further used to substitute the current battery health value, cumulative usage time and standard battery life of the energy storage battery into a preset battery life prediction formula to obtain the current remaining battery life of the energy storage battery; wherein the preset battery life prediction formula is:

In the formula, R represents the remaining battery life of the energy storage battery, T _life represents the standard life of the energy storage battery, SOH represents the current battery health value of the energy storage battery, f _fast represents the aging factor of the energy storage battery in fast charging mode, p _fast represents the fast charging ratio of the energy storage battery, f _slow represents the aging factor of the energy storage battery in slow charging mode, p _slow represents the slow charging ratio of the energy storage battery, and T _used represents the current cumulative usage time of the energy storage battery.

On the basis of the above embodiment, the feature determination module is also used to determine multiple driving behaviors of the electric vehicle according to the driving duration, braking frequency and driving speed in the driving parameters, and the driving behaviors include emergency braking behavior, high-speed driving behavior and long-distance driving behavior; according to the sampling period of each driving behavior and driving parameter, a driving behavior time series of the electric vehicle is generated, and the driving behavior time series is used as the driving behavior feature.

Based on the above embodiment, the feature determination module is also used to determine the long-distance driving behavior as the driving behavior of the electric vehicle if the driving time is greater than or equal to the preset time; if the braking frequency is greater than or equal to the preset frequency, the emergency braking behavior is determined as the driving behavior of the electric vehicle; if the driving speed is greater than or equal to the preset speed, the high-speed driving behavior is determined as the driving behavior of the electric vehicle.

On the basis of the above embodiments, the feature determination module is also used to determine the sampling time of the battery parameters; determine the voltage change rate of the energy storage battery according to the sampling time and voltage; determine the internal resistance increase rate of the energy storage battery according to the sampling time and internal resistance; determine the capacitance decay rate of the energy storage battery according to the sampling time and capacitance; and use the voltage change rate, internal resistance increase rate and capacitance decay rate as battery aging characteristics of the energy storage battery.

On the basis of the above embodiment, the SOH prediction module is also used to divide the preset period into multiple standard time periods; generate a feature matrix for each standard time period according to the driving behavior characteristics and battery aging characteristics within each standard time period; concatenate the characteristics of each row in each feature matrix in chronological order to obtain a feature vector for each standard time period; and determine the battery health feature vector of the energy storage battery within the preset period according to the feature vector of each standard time period.

Based on the above embodiment, the SOH prediction module is also used to divide the battery health feature vector into training set data and validation set data according to a preset ratio, and train the training model according to the training set data and the validation set data until a preset iteration termination condition is reached. The preset iteration termination condition is that the number of iterations reaches a preset threshold or the loss function of the training model converges; the training model that reaches the preset iteration termination condition is determined as the SOH prediction model.

It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

The present application also discloses an electronic device. Referring to FIG3 , FIG3 is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application. The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 , and at least one communication bus 302 .

The communication bus 302 is used to realize the connection and communication between these components.

The user interface 303 may include a display interface and a camera interface. The optional user interface 303 may also include a standard wired interface and a wireless interface.

The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface diagrams and applications, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

Among them, the memory 305 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also be at least one storage device located away from the aforementioned processor 301. Referring to Figure 3, the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program for a method for predicting SOH and RUL of an energy storage battery.

In the electronic device 300 shown in FIG3 , the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 301 can be used to call the application program of a SOH and RUL prediction method of an energy storage battery stored in the memory 305. When executed by one or more processors 301, the electronic device 300 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

In the several implementation modes provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not recorded in the present disclosure. The description and examples are to be regarded as exemplary only.

Claims (10)

1. A method for predicting SOH and RUL of an energy storage battery, comprising: Acquiring battery parameters of an energy storage battery and driving parameters of an electric vehicle corresponding to the energy storage battery; Based on the driving parameters, determining the driving behavior characteristics of the electric vehicle, and based on the battery parameters, determining the battery aging characteristics of the energy storage battery; Aligning the timestamps of the driving behavior feature and the battery aging feature within a preset period, and constructing a battery health feature vector of the energy storage battery within the preset period according to the driving behavior feature and the battery aging feature; Inputting the battery health feature vector into a training model to obtain a SOH prediction model of the energy storage battery, and predicting a current battery health value of the energy storage battery based on the SOH prediction model; The current remaining battery life of the energy storage battery is predicted based on the current battery health value, accumulated usage time and standard battery life of the energy storage battery. 2. The SOH and RUL prediction method of the energy storage battery according to claim 1, characterized in that the current remaining battery life of the energy storage battery is predicted according to the current battery health value, accumulated usage time and standard battery life of the energy storage battery, comprising: Substituting the current battery health value, accumulated usage time and standard battery life of the energy storage battery into a preset battery life prediction formula to obtain the current remaining battery life of the energy storage battery; Wherein, the preset battery life prediction formula is: In the formula, R represents the current remaining battery life of the energy storage battery, T _life represents the standard life of the energy storage battery, SOH represents the current battery health value of the energy storage battery, f _fast represents the aging factor of the energy storage battery in the fast charging mode, p _fast represents the fast charging ratio of the energy storage battery, f _slow represents the aging factor of the energy storage battery in the slow charging mode, p _slow represents the slow charging ratio of the energy storage battery, and T _used represents the current accumulated usage time of the energy storage battery. 3. The SOH and RUL prediction method of the energy storage battery according to claim 1, characterized in that the determining the driving behavior characteristics of the electric vehicle based on the driving parameters comprises: Determining multiple driving behaviors of the electric vehicle according to the driving duration, braking frequency, and driving speed in the driving parameters, wherein the driving behaviors include emergency braking behavior, high-speed driving behavior, and long-distance driving behavior; A driving behavior time series of the electric vehicle is generated according to the sampling periods of the driving behaviors and the travel parameters, and the driving behavior time series is used as the driving behavior feature. 4. The SOH and RUL prediction method of the energy storage battery according to claim 3, characterized in that the multiple driving behaviors of the electric vehicle are determined according to the driving duration, braking frequency and driving speed in the driving parameters, including: if the driving duration is greater than or equal to a preset duration, the long-distance driving behavior is determined as the driving behavior of the electric vehicle; If the braking frequency is greater than or equal to a preset frequency, determining the emergency braking behavior as the driving behavior of the electric vehicle; If the driving speed is greater than or equal to a preset speed, the high-speed driving behavior is determined as the driving behavior of the electric vehicle. 5. The SOH and RUL prediction method of the energy storage battery according to claim 1, characterized in that the battery parameters include voltage, internal resistance and capacitance, and the determining the battery aging characteristics of the energy storage battery based on the battery parameters includes: determining the sampling time of the battery parameters; Determining a voltage change rate of the energy storage battery according to the sampling duration and the voltage; Determining an internal resistance increase rate of the energy storage battery according to the sampling duration and the internal resistance; Determining a capacitance decay rate of the energy storage battery according to the sampling duration and the capacitance; The voltage change rate, the internal resistance increase rate and the capacitance decay rate are used as battery aging characteristics of the energy storage battery. 6. The SOH and RUL prediction method of the energy storage battery according to claim 1, characterized in that the step of constructing a battery health feature vector of the energy storage battery within the preset period according to the driving behavior characteristics and the battery aging characteristics comprises: Dividing the preset period into a plurality of standard time periods; Generating a feature matrix for each standard time period according to the driving behavior characteristics and battery aging characteristics within each standard time period; The features of each row in each feature matrix are concatenated in chronological order to obtain a feature vector for each standard time period; and a battery health feature vector of the energy storage battery within the preset period is determined based on the feature vector for each standard time period. 7. The SOH and RUL prediction method of the energy storage battery according to claim 1, characterized in that the step of inputting the battery health feature vector into a training model to obtain the SOH prediction model of the energy storage battery comprises: Dividing the battery health feature vector into training set data and validation set data according to a preset ratio, and training the training model according to the training set data and the validation set data until a preset iteration termination condition is reached, wherein the preset iteration termination condition is that the number of iterations reaches a preset threshold or the loss function of the training model converges; The training model that reaches the preset iteration termination condition is determined as the SOH prediction model. 8. A SOH and RUL prediction system for an energy storage battery, characterized in that the system comprises: A parameter acquisition module, used to acquire battery parameters of the energy storage battery and driving parameters of the electric vehicle corresponding to the energy storage battery; a feature determination module, used to determine the driving behavior characteristics of the electric vehicle based on the driving parameters, and to determine the battery aging characteristics of the energy storage battery based on the battery parameters; An SOH prediction module is used to align the timestamps of the driving behavior characteristics and the battery aging characteristics within a preset period, and construct a battery health feature vector of the energy storage battery within the preset period according to the driving behavior characteristics and the battery aging characteristics; input the battery health feature vector into a training model to obtain an SOH prediction model of the energy storage battery, and predict the current battery health value of the energy storage battery based on the SOH prediction model; The RUL prediction module is used to predict the current remaining battery life of the energy storage battery according to the current battery health value, accumulated usage time and standard battery life of the energy storage battery. 9. An electronic device, characterized in that it includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the SOH and RUL prediction method of an energy storage battery as described in any one of claims 1-7. 10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores instructions, and when the instructions are executed, the SOH and RUL prediction method of the energy storage battery according to any one of claims 1 to 7 is executed.