CN118316029A - A distribution box intelligent power regulation method and system based on artificial intelligence - Google Patents

Abstract

The application relates to the technical field of artificial intelligence, in particular to an intelligent power adjustment method and system for a distribution box based on artificial intelligence. According to the application, the electricity consumption data and the auxiliary data of the distribution box are acquired, and the electricity consumption data of the distribution box is acquired based on the electricity consumption data and the auxiliary data; analyzing power consumption information based on a preset deep learning model, and judging real-time power requirements of the distribution box; inquiring a historical operation database of the power distribution system to obtain historical operation data of the distribution box; based on the historical operation data, adjusting the real-time power demand to obtain an optimized power adjustment strategy; the power of the distribution box is regulated according to the optimized power regulation strategy, the load fluctuation characteristic can be adaptively matched, and the time lag and the error of power regulation are reduced, so that the operation efficiency and the power supply reliability of the distribution box are effectively improved, and the intelligent management and the optimal control of the distribution network are realized.

Description

A distribution box intelligent power regulation method and system based on artificial intelligence

Technical Field

The present application relates to the technical field of artificial intelligence, and in particular to an artificial intelligence-based intelligent power regulation method and system for a distribution box.

Background technique

With the rapid development of social economy and the continuous growth of electricity demand, the distribution network is facing severe challenges in power supply reliability, power quality and operation efficiency. In order to adapt to the development needs of the distribution network under the new situation, it is urgent to introduce advanced technical means to realize the intelligent management and optimization control of distribution equipment and improve the safety, reliability and economy of the distribution network.

Traditional power distribution control methods are usually designed based on historical experience and typical scenarios, and are difficult to adapt to complex and changeable actual operating conditions. Secondly, fixed model parameters are difficult to reflect the dynamic changes of power distribution equipment and loads in real time, resulting in insufficient timeliness and accuracy of control decisions, which needs to be further improved.

Summary of the invention

In order to solve the problems of poor adaptability, lack of timeliness and accuracy of existing power distribution control methods, the present application provides an intelligent power regulation method and system for a distribution box based on artificial intelligence, which adopts the following technical solutions:

In a first aspect, the present application provides an artificial intelligence-based intelligent power regulation method for a distribution box, comprising the following steps:

Collecting power consumption data and auxiliary data of the distribution box, and obtaining power consumption information of the distribution box based on the power consumption data and auxiliary data;

Analyze the power consumption information based on a preset deep learning model to determine the real-time power demand of the distribution box;

Query the historical operation database of the power distribution system and obtain the historical operation data of the distribution box;

Based on the historical operation data, the real-time power demand is adjusted to obtain an optimized power regulation strategy;

The power of the distribution box is adjusted according to the optimized power adjustment strategy.

By adopting the above technical solution, the existing power regulation method of the distribution box mainly relies on preset control rules and fixed optimization models, which is difficult to adapt to the complexity and variability of actual power distribution scenarios. For example, in industrial parks, due to the frequent start and stop of production equipment, the fixed optimization model cannot reflect the dynamic changes of the equipment status in real time, resulting in the power regulation lag of the distribution box, which increases line loss and equipment loss. In response to the above problems, the intelligent power regulation method of the distribution box based on artificial intelligence proposed in this application analyzes the real-time power consumption information of the distribution box through a preset deep learning model, judges the power demand, and dynamically optimizes the regulation strategy based on historical operation data. It can adaptively match the load fluctuation characteristics and reduce the time lag and error of power regulation, thereby effectively improving the operating efficiency and power supply reliability of the distribution box, and realizing the intelligent management and optimization control of the distribution network.

Optionally, the auxiliary data includes temperature and humidity data, meteorological data, holiday information and electrical equipment operation information.

By adopting the above technical solution, the existing power regulation method only considers the real-time power consumption data of the distribution box, ignoring the external factors that affect the power load, resulting in insufficient adaptability and robustness of power regulation. For example, when an industrial park is affected by external factors such as bad weather and extreme temperatures, these factors may change the power consumption characteristics of the equipment, causing deviations in the power regulation decisions of the distribution box, and increasing the risk of equipment damage and safety accidents. In response to the above problems, the present application comprehensively utilizes auxiliary data such as temperature and humidity data, meteorological data, holiday information, and power equipment operation information when the distribution box is intelligently adjusted for power. Through correlation analysis of multi-source data, it dynamically evaluates the impact of external factors on equipment power consumption, and adaptively adjusts the power output of the distribution box, ensuring that the adjustment decision of the distribution box is accurately matched with the actual operation requirements of the equipment, thereby improving the energy utilization efficiency and production safety of the industrial park.

Optionally, analyzing the power consumption information based on a preset deep learning model to determine the real-time power demand of the distribution box specifically includes the following steps:

Using a convolutional neural network to extract features from the power consumption information, and obtaining load level depth features and load fluctuation depth features of the distribution box;

The real-time power demand of the distribution box is generated according to the load level depth characteristics and the load fluctuation degree depth characteristics.

By adopting the above-mentioned technical solution, the present application uses a convolutional neural network to perform deep feature extraction on the power consumption information of the distribution box. The extracted deep features of load level and fluctuation degree can reflect the real-time load status of the distribution box, significantly improve the accuracy of demand forecasting, and provide a reliable decision-making basis for the intelligent adjustment of the distribution box, effectively reduce power adjustment errors, and improve the robustness of the distribution system.

Optionally, querying the historical operation database of the power distribution system to obtain the historical operation data of the distribution box specifically includes the following steps:

Query the historical operation database and mark the historical power demand evaluation index of the distribution box in the previous time period;

Based on the mark, historical operation data corresponding to the historical power demand assessment index is obtained.

By adopting the above-mentioned technical solution, the existing power regulation method of the distribution box usually adopts a simple data query and screening method when using historical operation data. Due to the differences in production tasks and equipment operating conditions in different periods, the power demand of the distribution box will show dynamic changes. The traditional historical data query method based on time windows or load thresholds is difficult to accurately match the current load state. The present application introduces a historical power demand evaluation index, which constructs a mapping relationship between historical operation data and power demand status by quantitatively evaluating and marking the power demand status of the distribution box in different historical periods. When performing historical data queries, by matching the current power demand evaluation index, historical operation data similar to the current status can be quickly and accurately retrieved, which improves the efficiency and accuracy of data queries, provides high-quality historical sample support for the optimization of regulation strategies, and improves the real-time and effectiveness of strategy optimization.

Optionally, adjusting the real-time power demand based on the historical operation data to obtain an optimized power regulation strategy specifically includes the following steps:

Based on the historical operation data, obtaining a historical power demand evaluation index and historical adjustment measures of the distribution box;

Compare the current real-time power demand with the historical power demand evaluation index, and mark the situations with the same comparison results as the same load situation;

Based on the historical regulation measures of the same load conditions, adjust the output power curve of the corresponding distribution box;

Based on the adjusted output power curve, the real-time power demand corresponding to the same load situation is adjusted to obtain an optimized power regulation strategy.

By adopting the above technical solution, in the wind farm, due to the variability of natural conditions such as wind speed and wind direction, the power demand of the distribution box shows significant fluctuations and uncertainties. It is difficult to adapt to the changes in actual operating conditions by using a fixed optimization model, and it is easy to have the problem of deviation between the optimization results and the actual demand, which affects the grid-connected efficiency and stability of the wind farm. When optimizing the strategy, this application fully taps the distribution box adjustment experience and knowledge contained in the historical operation data. By comparing the real-time power demand with the historical demand assessment index, similar load conditions similar to the current state are identified, and the historical adjustment measures in this case are used to adaptively adjust the output power curve of the distribution box, so that the optimized power adjustment strategy can inherit the successful experience of history and improve the environmental adaptability of the strategy. At the same time, the adjusted output power curve has a higher matching degree with the real-time power demand, which effectively reduces the deviation between the optimization results and the actual demand, and enhances the robustness and reliability of the power regulation of the distribution box.

Optionally, the method further comprises the following steps:

Based on the output power curve in the optimized power regulation strategy and the preset power association rule, the main transformer, energy storage system and transmission line of the distribution box are associated to obtain associated power equipment;

Based on the preset maintenance cycle of each power device, generate the estimated maintenance time of the corresponding associated power device;

Based on the estimated maintenance time of the associated power equipment, adjusting the actual maintenance time of the corresponding associated power equipment;

Based on the adjusted actual maintenance time of each power equipment, each component of the distribution box is regularly inspected and maintained to achieve intelligent management.

By adopting the above technical solution, the existing distribution equipment maintenance management method usually adopts fixed time intervals or fault repair strategies, which makes it difficult to fully consider the actual operating status of the equipment and related impacts, resulting in poor maintenance efficiency and effect. This application is based on the optimized output power curve of the distribution box, comprehensively considers the operating status of related power equipment, associates the equipment through power association rules, selects time periods with relatively low loads to maintain some related equipment, and avoids peak load periods. At the same time, concentrating related equipment in the same time period for power outage maintenance can reduce the number of maintenance times, improve maintenance efficiency, and avoid the impact of frequent power outages on users.

Optionally, the method further comprises the following steps:

Based on the power consumption information, establish a power consumption pattern profile of the distribution box;

Obtain energy consumption data and operating status information of other intelligent systems related to the area where the distribution box is located;

Correlate and analyze the power consumption pattern portrait, energy consumption data of other intelligent systems, and operating status information to identify the comprehensive energy consumption pattern in the area where the distribution box is located;

The optimized power regulation strategy is adjusted based on the comprehensive energy consumption mode, and the adjustment information is fed back to the other intelligent systems.

By adopting the above technical solutions, the existing power regulation method of the distribution box focuses on the operating status and energy consumption data of the distribution equipment itself, lacks information interaction and collaborative optimization with other intelligent systems in the area, resulting in the adjustment strategy of the distribution box being out of touch with the overall energy consumption pattern of the region, making it difficult to achieve optimal management of regional energy efficiency. For example, in a smart park, in addition to the distribution system, multiple intelligent systems such as building automation and lighting control are deployed, and their energy consumption behavior and operating status have a significant impact on the regional power consumption characteristics. This application first establishes a refined power consumption pattern portrait based on the power consumption information of the distribution box itself, and then actively obtains the energy consumption and status data of other intelligent systems in the region, identifies the regional comprehensive energy consumption pattern through correlation analysis, and uses it to guide the dynamic optimization of the distribution box adjustment strategy. At the same time, the optimized adjustment information is fed back to the relevant intelligent system, realizing the two-way interaction between the distribution box and the external system, and constructing a regional energy efficiency closed-loop optimization mechanism. While improving the operating efficiency of the distribution box, it promotes the collaborative optimization of the overall energy consumption pattern of the region and achieves a regional level improvement in energy utilization efficiency.

In a second aspect, the present application provides an artificial intelligence-based intelligent power regulation system for a distribution box, comprising:

A data collection module, used to collect power consumption data and auxiliary data of the distribution box, and obtain power consumption information of the distribution box based on the power consumption data and auxiliary data;

A real-time power demand judgment module, used to judge the real-time power demand of the distribution box based on the power consumption information;

The historical operation data query module is used to query the historical operation database of the power distribution system and obtain the historical operation data of the distribution box;

A power regulation strategy generating module, used to adjust the real-time power demand based on the historical operation data to obtain an optimized power regulation strategy;

The power regulation module is used to regulate the power of the distribution box according to the optimized power regulation strategy.

In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned artificial intelligence-based intelligent power regulation method for a distribution box when executing the computer program.

In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned artificial intelligence-based intelligent power regulation method for a distribution box.

In summary, the present application includes at least one of the following beneficial technical effects:

1. The intelligent power regulation method for distribution boxes based on artificial intelligence proposed in this application analyzes the real-time power consumption information of the distribution box through a preset deep learning model, determines the power demand, and dynamically optimizes the regulation strategy in combination with historical operation data. It can adaptively match the load fluctuation characteristics and reduce the time lag and error of power regulation, thereby effectively improving the operating efficiency and power supply reliability of the distribution box, and realizing intelligent management and optimization control of the distribution network;

2. This application introduces a historical power demand assessment index, and constructs a mapping relationship between historical operation data and power demand status by quantitatively evaluating and marking the power demand status of the distribution box in different historical periods; when querying historical data, by matching the current power demand assessment index, historical operation data similar to the current status can be quickly and accurately retrieved, which improves the efficiency and accuracy of data query, provides high-quality historical sample support for the optimization of regulation strategies, and improves the real-time and effectiveness of strategy optimization;

3. This application identifies similar load conditions that are similar to the current state by comparing the real-time power demand with the historical demand assessment index, and uses the historical adjustment measures in this case to adaptively adjust the output power curve of the distribution box, so that the optimized power adjustment strategy can inherit historical successful experience and improve the environmental adaptability of the strategy; at the same time, the adjusted output power curve has a higher degree of match with the real-time power demand, which effectively reduces the deviation between the optimization results and the actual demand, and enhances the robustness and reliability of the power adjustment of the distribution box.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is an exemplary flow chart of an intelligent power regulation method for a distribution box based on artificial intelligence according to an embodiment of the present application;

FIG2 is a flow chart of step S20 in the method for intelligent power regulation of a distribution box based on artificial intelligence according to an embodiment of the present application;

Figure 3 is a flow chart of step S30 of the distribution box intelligent power adjustment method based on artificial intelligence in the present application embodiment;

FIG4 is a flow chart of step S40 in the method for intelligent power regulation of a distribution box based on artificial intelligence according to an embodiment of the present application;

5 is a flow chart of intelligent management based on associated power equipment in the intelligent power regulation method for a distribution box based on artificial intelligence according to an embodiment of the present application;

6 is a flow chart of power adjustment based on comprehensive energy consumption in a region in an intelligent power adjustment method for a distribution box based on artificial intelligence according to an embodiment of the present application;

FIG7 is a schematic diagram of a module of an intelligent power regulation system for a distribution box based on artificial intelligence according to an embodiment of the present application;

FIG. 8 is a diagram showing the internal structure of an electronic device according to an embodiment of the present application.

Detailed ways

The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and/or" used in the present application refers to any or all possible combinations comprising one or more listed items.

In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

The present application provides an artificial intelligence-based intelligent power regulation method and system for a distribution box, which analyzes the real-time power consumption information of the distribution box through a preset deep learning model, determines the power demand, and dynamically optimizes the regulation strategy in combination with historical operating data. It can adaptively match the load fluctuation characteristics and reduce the time lag and error of power regulation, thereby effectively improving the operating efficiency and power supply reliability of the distribution box, and realizing intelligent management and optimized control of the distribution network.

The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

The embodiment of the present application provides a method, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. In this embodiment, the terminal device is an electronic device, but is not limited to this, and can also be a smart tablet, a computer, etc. The terminal device and the server can be directly or indirectly connected through wired or wireless communication, and the embodiment of the present application does not limit this.

Referring to FIG1 , an intelligent power regulation method for a distribution box based on artificial intelligence comprises the following steps:

S10: Collect power consumption data and auxiliary data of the distribution box, and obtain power consumption information of the distribution box based on the power consumption data and the auxiliary data.

In this embodiment, the distribution box refers to an electrical device used to distribute electric energy to multiple electrical equipment or areas, and is usually installed at the power consumption site or in a building. Electricity consumption data refers to data reflecting the operating status of the distribution box and the use of electric energy, which is collected and obtained through various sensors and instruments, including voltage data, current data, power data, energy consumption data, harmonic data, event data, etc. Auxiliary data refers to data related to the operating environment and load characteristics of the distribution box. In this embodiment, the auxiliary data includes temperature and humidity data, meteorological data, holiday information, and electrical equipment operation information.

Specifically, voltage, current, power and other sensors are installed at the inlet and outlet lines, switches and other locations of the distribution box to collect relevant electrical quantities in real time. Temperature and humidity sensors are arranged near the distribution box to collect environmental parameters. The data collected by each sensor is then uploaded to the data concentrator or cloud platform through the communication gateway, and auxiliary data such as meteorological data, holiday information and power equipment operation information are obtained from the meteorological department, calendar, and power consumption information system to form a power consumption information data set.

S20. Analyze the power consumption information based on the preset deep learning model to determine the real-time power demand of the distribution box.

In this embodiment, the preset deep learning model may include a convolutional neural network, a recurrent neural network, a long short-term memory network, etc.

Specifically, the present application divides the historical electricity consumption data into multiple samples at fixed time intervals, and each sample contains various electricity consumption data and auxiliary data at the current moment, as well as the actual electricity consumption in the future. Then, a deep learning model suitable for processing time series data, such as a long short-term memory network, is selected, and the model is trained and optimized using the training set data. After the training is completed, the prediction performance of the model is evaluated using the test set data to ensure that it meets the requirements of actual applications. Finally, the currently collected electricity consumption data and auxiliary data are input into the trained model to obtain the power demand forecast value of the distribution box in the future, providing a basis for subsequent power optimization and adjustment.

S30: query the historical operation database of the power distribution system to obtain the historical operation data of the distribution box.

In this embodiment, the historical operation database of the distribution system refers to a large database that stores the long-term operation records of the distribution network, including the operating status, electrical parameters, fault events, and related ambient temperature, adjustment instructions, and other information of equipment such as distribution boxes, substations and transmission lines over a period of time in the past.

Specifically, by querying the historical operation database of the distribution system, the operating data of the target distribution box in different historical time periods can be obtained, such as voltage, current, power, power factor, etc., as well as the related ambient temperature, load type, and adjustment instructions. These historical data can reflect the load characteristics and adjustment rules of the distribution box in actual operation, and serve as an important reference for optimizing the current power adjustment strategy.

S40: Based on the historical operation data, adjust the real-time power demand to obtain an optimized power regulation strategy.

In this embodiment, the optimized power regulation strategy means that under the premise of ensuring power supply reliability and power quality, the operating parameters and control strategy of the distribution box are reasonably adjusted to make its output power curve match the actual load demand as much as possible, thereby reducing power waste and equipment loss and improving energy utilization efficiency.

Specifically, the historical load data is analyzed, similar load curves are classified into one category, and several typical load patterns are formed. The current real-time power demand of the distribution box is compared with the typical load pattern in the historical operation data to identify the historical period that is closest to the current load situation. Then, the operating status and adjustment measures of the distribution box in the historical period are analyzed, and the optimization model is established in combination with the current environmental conditions, equipment status and other factors. The goal is to minimize the error between the real-time power demand and the optimized output power curve, and the historical operating status and adjustment measures are used as constraints to solve the optimized output power curve. Finally, according to the optimized output power curve, the corresponding power control strategy is formulated to guide the actual operation of the distribution box, so as to achieve efficient and energy-saving operation of the distribution network while dynamically adapting to load changes.

S50: Regulate the power of the distribution box according to the optimized power regulation strategy.

Specifically, the actual output power of the distribution box is dynamically adjusted according to the optimized power regulation strategy. Specifically, the distribution automation system converts the power regulation strategy into a series of specific control instructions, such as adjusting the transformer tap position, switching the reactive compensation capacitor group, adjusting the start sequence of the motor, etc., and executes these instructions through the intelligent control unit of the distribution box, such as the local intelligent terminal or the local controller. At the same time, the distribution automation system collects the operating data of the distribution box in real time, calculates the deviation between the actual output power and the optimized power curve, and automatically adjusts the actual output power so that the actual output power tracks the optimized power curve as much as possible, ensuring that the distribution box reduces network losses and improves power supply efficiency while meeting the user's power demand.

In one embodiment, as shown in FIG. 2 , in step S20, the power consumption information is analyzed based on a preset deep learning model to determine the real-time power demand of the distribution box, which specifically includes the following steps:

S21. Use a convolutional neural network to extract features from electricity consumption information to obtain deep features of the load level and load fluctuation degree of the distribution box.

In this embodiment, the load level deep features refer to a set of numerical indicators extracted by convolutional neural networks that reflect the load level of the distribution box, and characterize the statistical characteristics of the distribution box on different time scales, such as the average load rate, maximum load rate, and minimum load rate. The load fluctuation degree deep features refer to a set of numerical indicators extracted by convolutional neural networks that reflect the load fluctuation of the distribution box. These indicators can measure the dynamic characteristics of the load curve, such as the smoothness, peak-to-valley difference rate, and fluctuation frequency.

Specifically, the present application converts the power consumption information of the distribution box into a two-dimensional matrix or image, and inputs it into a pre-trained convolutional neural network model. Through the filtering operation of the convolution layer and the downsampling operation of the pooling layer, the deep features reflecting the load level and load fluctuation degree of the distribution box are extracted, such as the average load rate, the peak-to-valley difference rate, the fluctuation of the load curve, etc., so as to more comprehensively and accurately characterize the real-time load status of the distribution box and provide effective input information for subsequent power demand prediction.

S22. Generate the real-time power demand of the distribution box according to the load level depth characteristics and the load fluctuation degree depth characteristics.

Specifically, the extracted load level deep features and load fluctuation degree deep features are input into a pre-trained power demand forecasting model. The model can be a multi-layer perceptron neural network or a long short-term memory neural network, which converts the load features into power demand forecast values for a period of time in the future.

In one embodiment, as shown in FIG3 , in step S30, querying the historical operation database of the power distribution system to obtain the historical operation data of the distribution box specifically includes the following steps:

S31. Query the historical operation database and mark the historical power demand evaluation index of the distribution box in the previous time period.

In this embodiment, the historical power demand evaluation index refers to a quantitative indicator for comprehensively evaluating the rationality of the actual power demand of the distribution box in a specific historical period, which is calculated by weighted summation or principal component analysis and other methods in combination with the statistical data of the load level, load fluctuation, power factor, power quality and other aspects of the distribution box in the period. The higher the index value, the closer the actual power demand of the distribution box in the period is to the optimal state, that is, the load curve is stable, the power factor is high, the power quality is good, etc.; the lower the index value, the more problems such as low power consumption, power waste, and poor power quality exist in the period, which need to be optimized and adjusted. By marking the power demand evaluation index of each period in the historical database, it can provide a basis for judgment for subsequent searches for historical periods similar to the current load status, and then refer to the optimization and adjustment measures of the period to formulate targeted power optimization strategies.

S32. Based on the tag, obtain historical operation data corresponding to the historical power demand assessment index.

In this embodiment, based on the real-time power demand characteristics of the current distribution box, such as load level, load fluctuation degree, etc., similar historical time periods are searched in the historical operation database. Specifically, the Euclidean distance or cosine similarity between the current power demand feature vector and the feature vector of each historical time period can be calculated, and several time periods with the smallest distance or the largest similarity can be selected as candidates. Then, among the candidate time periods, time periods with higher historical power demand evaluation indexes are preferentially selected, because the actual electricity demand corresponding to these time periods is more reasonable, and their adjustment measures and operating status can provide useful references for the optimization of the current time period. Finally, based on the detailed operating data of the distribution box in these preferred time periods, including the actual load curve, control instruction sequence and equipment parameters, it is used as the input for optimizing the current power regulation strategy.

In one embodiment, as shown in FIG. 4 , in step S40, based on the historical operation data, the real-time power demand is adjusted to obtain an optimized power regulation strategy, which specifically includes the following steps:

S41. Based on historical operation data, obtain the historical power demand evaluation index and historical adjustment measures of the distribution box.

Specifically, the historical power demand evaluation index of the distribution box in the preferred time period is extracted as a reference standard for measuring the rationality of the actual power demand in the period. Then the historical adjustment measures of the distribution box in the period are analyzed, including operation records such as transformer tap adjustment, reactive compensation device switching, load control, and the corresponding changes in electrical parameters such as voltage, current, and power.

S42: Compare the current real-time power demand with the historical power demand evaluation index, and mark situations with the same comparison results as similar load situations.

Specifically, the real-time power demand of the current distribution box is compared with the power demand of the selected preferred historical period. The similarity between the current power demand feature vector and the power demand feature vector of the historical period is calculated. If the similarity exceeds a preset threshold, the two are marked as the same load situation. This means that the current power demand pattern of the distribution box is highly similar to that of the historical period, and they show similar characteristics in terms of load level, load fluctuation, power factor, etc. By identifying the historical period closest to the current load state, we can infer the power supply problems and optimization space that may be faced in the current period, and refer to the optimization and adjustment measures of the historical period to quickly formulate a power optimization strategy.

S43. Based on historical regulation measures for similar load conditions, adjust the output power curve of the corresponding distribution box.

Specifically, we extract historical adjustment measures under similar load conditions, including operation records such as transformer tap adjustment, reactive compensation device switching, load control, and corresponding changes in electrical parameters such as voltage, current, and power. Then, we analyze the impact of these adjustment measures on the operating status of the distribution box and evaluate their applicability and effectiveness under the current load conditions. For adjustment measures that have been proven to effectively improve the operating performance of the distribution box, we apply them to the output power curve of the current distribution box. By adjusting the transformer tap position, switching capacitor banks, controlling controllable loads, etc., we make the actual output power of the distribution box as close to the optimal power curve as possible.

S44. Based on the adjusted output power curve, the real-time power demand corresponding to the same load condition is adjusted to obtain an optimized power regulation strategy.

Specifically, the adjusted output power curve is compared with the original real-time power demand curve, and the power difference between the two curves at each time point is calculated. These power differences are then used as corrections and superimposed on the original real-time power demand curve to obtain an optimized real-time power demand curve. This curve comprehensively considers the actual output capacity and operating status of the distribution box, and can more accurately reflect the actual power demand of the distribution box under the current load conditions.

In one embodiment, as shown in FIG5 , after step S50, the method further includes the following steps:

S51. Based on the output power curve in the optimized power regulation strategy and the preset power association rules, the main transformer, energy storage system and transmission line of the distribution box are associated to obtain associated power equipment.

In this embodiment, the preset power association rule is used to describe the power balance and constraint relationship between different power devices inside the distribution box.

Specifically, assume that the distribution box contains a main transformer, an energy storage system and three transmission lines, where the rated capacity of the main transformer is 1000kVA, the maximum charge and discharge power of the energy storage system is 200kW, and the rated currents of the three transmission lines are 200A, 300A and 250A respectively. By analyzing the output power curve in the optimized power regulation strategy, we find that in the next hour, the total power load of the distribution box will show a trend of first rising and then falling, and the maximum load will appear in the 45th minute, reaching 850kW. According to the preset power association rules, we calculate that at this load level, the load rate of the main transformer will reach 85%, the energy storage system needs to provide 100kW of discharge power, and the currents of the three transmission lines are 180A, 270A and 225A respectively. Since the operating status of the main transformer and the energy storage system is closely related to the output power, and the currents of the two transmission lines are close to the rated values, we mark them as associated power equipment and focus on monitoring and optimization in subsequent dispatching control.

S52: Based on the preset maintenance cycle of each electric power device, generate an estimated maintenance time of the corresponding associated electric power device.

In this embodiment, we pre-set the corresponding maintenance cycle for each device according to the type and operation characteristics of the associated power equipment. Specifically, for the main transformer, considering its long continuous operation time and frequent load changes, we set its maintenance cycle to perform routine maintenance every 5,000 hours of operation and a comprehensive overhaul every 10,000 hours of operation. For the energy storage system, due to its high number of charge and discharge times and complex working conditions, we set its maintenance cycle to perform routine maintenance every 1,000 charge and discharge cycles and a comprehensive overhaul every 5,000 cycles. For the transmission line, considering its large environmental impact and relatively high failure rate, we set its maintenance cycle to perform routine inspections every year of operation and a comprehensive overhaul every 5 years of operation. Based on these preset maintenance cycles, combined with the cumulative operation time or number of the current equipment, we can calculate the expected maintenance time of each associated power equipment. For example, for a main transformer that has been in operation for 3,000 hours, the expected time for the next routine maintenance is 2,000 hours later, and the expected time for the next comprehensive overhaul is 7,000 hours later. By generating an estimated maintenance schedule for the equipment, we can make maintenance plans in advance, reasonably arrange equipment shutdown and spare parts replacement, and ensure the reliability and stability of the distribution box.

S53: Based on the estimated maintenance time of the associated power equipment, adjust the actual maintenance time of the corresponding associated power equipment.

In this embodiment, after generating the estimated maintenance schedule of the associated power equipment, the maintenance time of the equipment is dynamically adjusted in combination with the actual operation requirements and site conditions of the distribution box. Specifically, the load forecast and grid dispatch plan of the distribution box in the future are first analyzed to identify the time window with low load and sufficient grid backup as the preferred equipment maintenance time. Then the health status and remaining life of the associated power equipment are evaluated. For equipment with good operating status and sufficient remaining life, the maintenance cycle is extended to reduce unnecessary shutdowns and manual intervention; for equipment with abnormal signs and high risk levels, its maintenance cycle is shortened, and preventive maintenance and status assessment are carried out in advance. Further, combined with on-site factors such as weather conditions, operation time, and manpower arrangements, the actual maintenance time of the equipment is fine-tuned to make it conform to the optimal maintenance window and operation specifications as much as possible. For example, for a main transformer that is expected to undergo routine maintenance within the next 1000 hours, if its operating status is normal and the grid load is low in the next week, its actual maintenance time is postponed to the next low period to minimize the impact on the operation of the distribution box. By dynamically adjusting and optimizing the maintenance time, the resource utilization efficiency and operating economy of the distribution box can be maximized while ensuring the reliability of the equipment. At the same time, concentrating the related equipment in the same time period for power outage maintenance can reduce the number of maintenance times, improve maintenance efficiency, and avoid the impact of frequent power outages on users.

S54. Based on the adjusted actual maintenance time of each power equipment, each component of the distribution box is regularly inspected and maintained to achieve intelligent management.

In this embodiment, we perform regular inspection and maintenance on the various components of the distribution box according to the adjusted actual maintenance schedule of the associated power equipment.

In one embodiment, as shown in FIG6 , the method further includes the following steps:

S61. Based on the electricity consumption information, a power consumption pattern profile of the distribution box is established.

In this embodiment, the power consumption pattern portrait refers to the extraction of a series of key indicators and patterns that reflect the characteristics and laws of power consumption behavior by analyzing and mining the historical power consumption data of the distribution box.

Specifically, taking a residential distribution box as an example, we first collect the power consumption information of the distribution box in the past year through smart meters and online monitoring systems, including time series data such as voltage, current, power, electricity, and power factor. Then, we preprocess and extract features from these data to obtain key indicators that reflect the power consumption pattern, such as load curves, peak power consumption periods, seasonal changes, etc. On this basis, the power consumption pattern of the distribution box is divided into several typical categories, such as weekday mode, weekend mode, summer air-conditioning mode, etc., combined with external information such as weather and holidays, to generate corresponding power consumption pattern portraits. For example, for the summer air-conditioning mode, its power load usually reaches its peak in the afternoon, and shows a significant positive correlation with the maximum daily temperature.

S62. Obtain energy consumption data and operating status information of other intelligent systems related to the area where the distribution box is located.

Specifically, by establishing data interfaces with other intelligent systems in the area, such as building automation, lighting control and other intelligent systems, the energy consumption data of other energy-consuming equipment in the area can be collected in real time. At the same time, the operating status parameters of these devices can be obtained, such as power on and off time, process parameters, fault alarms, etc.

S63, correlate and analyze the power consumption pattern portrait, energy consumption data of other intelligent systems, and operation status information to identify the comprehensive energy consumption pattern in the area where the distribution box is located.

Specifically, the data of each system is aligned and synchronized according to the timestamp to form a multi-dimensional time series data set. Then, algorithms such as multivariate statistical analysis and time series mining are applied to explore the correlation and causal relationship between different energy consumption indicators, identify the key factors affecting the comprehensive energy efficiency of the park, and comprehensively characterize the energy consumption and influencing factors in the area where the distribution box is located by aggregating the data of various intelligent systems in the area.

S64. Adjust the optimized power regulation strategy based on the comprehensive energy consumption mode, and feed back the adjustment information to other intelligent systems.

In this embodiment, based on the identified comprehensive energy consumption pattern, the power regulation strategy of the distribution box is further optimized and adjusted, and the adjustment information is fed back to other intelligent systems in the park. Specifically, the comprehensive energy consumption pattern is first embedded in the intelligent scheduling algorithm of the distribution box, and it is used as an important decision-making basis and constraint condition. For example, during the peak period of regional electricity consumption, we can predict the load change trend in the future period through the comprehensive energy consumption pattern, and adjust the power supply strategy of the distribution box in advance, such as enabling the energy storage system, adjusting the transformer tap, etc., to ensure the safety of the power grid and the reliability of power supply. Through the two-way interaction between the distribution box and the external system, a regional energy efficiency closed-loop optimization mechanism is constructed. While improving the operating efficiency of the distribution box, it promotes the coordinated optimization of the overall energy consumption pattern of the region and achieves a regional-level improvement in energy utilization efficiency.

The implementation principle of an intelligent power regulation method for a distribution box based on artificial intelligence in an embodiment of the present application is as follows: by analyzing the real-time power consumption information of the distribution box through a preset deep learning model, the power demand is judged, and the regulation strategy is dynamically optimized in combination with historical operating data. This can adaptively match the load fluctuation characteristics, reduce the time lag and error of power regulation, thereby effectively improving the operating efficiency and power supply reliability of the distribution box, and realizing intelligent management and optimized control of the distribution network.

In the second aspect, the present application provides an intelligent power regulation system for a distribution box based on artificial intelligence. The intelligent power regulation system for a distribution box based on artificial intelligence of the present application is described below in combination with the above-mentioned intelligent power regulation method for a distribution box based on artificial intelligence. Please refer to Figure 7, which is a module schematic diagram of the intelligent power regulation system for a distribution box based on artificial intelligence in an embodiment of the present application.

An intelligent power regulation system for a distribution box based on artificial intelligence, comprising:

A data acquisition module is used to collect power consumption data and auxiliary data of the distribution box, and obtain power consumption information of the distribution box based on the power consumption data and auxiliary data;

A real-time power demand judgment module is used to judge the real-time power demand of the distribution box based on power consumption information;

The historical operation data query module is used to query the historical operation database of the power distribution system and obtain the historical operation data of the distribution box;

A power regulation strategy generation module is used to adjust the real-time power demand based on historical operation data to obtain an optimized power regulation strategy;

The power regulation module is used to regulate the power of the distribution box according to the optimized power regulation strategy.

Optionally, the real-time power demand judgment module includes:

A feature acquisition unit is used to extract features of power consumption information using a convolutional neural network to obtain load level depth features and load fluctuation depth features of the distribution box;

The real-time power demand generation unit is used to generate the real-time power demand of the distribution box according to the load level depth characteristics and the load fluctuation degree depth characteristics.

Optionally, the historical operation data query module includes:

A historical power demand evaluation index marking unit is used to query the historical operation database and mark the historical power demand evaluation index of the distribution box in the previous time period;

The historical operation data acquisition unit is used to acquire the historical operation data corresponding to the historical power demand evaluation index based on the mark.

Optionally, the power regulation strategy generation module includes:

An index and measure acquisition unit, used for acquiring a historical power demand evaluation index and historical adjustment measures of a distribution box based on historical operation data;

A comparison unit, used to compare the current real-time power demand with the historical power demand evaluation index, and mark the situations with the same comparison results as the same load situation;

The curve adjustment unit is used to adjust the output power curve of the corresponding distribution box based on historical adjustment measures of similar load conditions;

The regulation strategy generating unit is used to adjust the real-time power demand corresponding to the same load situation based on the adjusted output power curve to obtain an optimized power regulation strategy.

Optionally, the system further includes:

An associated power equipment acquisition module is used to associate the main transformer, energy storage system and transmission line of the distribution box based on the output power curve in the optimized power regulation strategy and the preset power association rules to obtain associated power equipment;

An estimated maintenance time generation module, used to generate an estimated maintenance time of a corresponding associated power device based on a preset maintenance cycle of each power device;

An actual maintenance time adjustment module, used to adjust the actual maintenance time of the corresponding associated power equipment based on the estimated maintenance time of the associated power equipment;

The intelligent management module is used to perform regular inspection and maintenance of the components of the distribution box based on the adjusted actual maintenance time of each power equipment to achieve intelligent management.

Optionally, the system further includes:

A power consumption pattern portrait building module, used to build a power consumption pattern portrait of the distribution box based on power consumption information;

An information acquisition module is used to obtain energy consumption data and operating status information of other intelligent systems related to the area where the distribution box is located;

The comprehensive energy consumption pattern recognition module is used to correlate and analyze the power consumption pattern portrait, energy consumption data of other intelligent systems, and operating status information to identify the comprehensive energy consumption pattern in the area where the distribution box is located;

The feedback module is used to adjust the optimized power regulation strategy based on the comprehensive energy consumption mode and feed back the adjustment information to other intelligent systems.

In one embodiment, the present application provides an electronic device, which may be a server, and its internal structure diagram may be as shown in Figure 8. The electronic device includes a processor, a memory, and a network interface connected via a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the electronic device is used to store data. The network interface of the electronic device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an intelligent power regulation method for a distribution box based on artificial intelligence is implemented.

Those skilled in the art will understand that the structure shown in FIG. 8 is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

In one embodiment, an electronic device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the above-mentioned computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims (10)

1. An intelligent power regulation method for a distribution box based on artificial intelligence, characterized in that the method comprises: Collecting power consumption data and auxiliary data of the distribution box, and obtaining power consumption information of the distribution box based on the power consumption data and auxiliary data; Analyze the power consumption information based on a preset deep learning model to determine the real-time power demand of the distribution box; Query the historical operation database of the power distribution system and obtain the historical operation data of the distribution box; Based on the historical operation data, the real-time power demand is adjusted to obtain an optimized power regulation strategy; The power of the distribution box is adjusted according to the optimized power adjustment strategy. 2. According to the artificial intelligence-based intelligent power regulation method for distribution box according to claim 1, it is characterized in that the auxiliary data includes temperature and humidity data, meteorological data, holiday information and power equipment operation information. 3. According to the method for intelligent power regulation of a distribution box based on artificial intelligence in claim 1, it is characterized in that the power consumption information is analyzed based on a preset deep learning model to determine the real-time power demand of the distribution box, which specifically includes the following steps: Using a convolutional neural network to extract features from the power consumption information, and obtaining load level depth features and load fluctuation depth features of the distribution box; The real-time power demand of the distribution box is generated according to the load level depth characteristics and the load fluctuation degree depth characteristics. 4. The method for intelligent power regulation of a distribution box based on artificial intelligence according to claim 1 is characterized in that the querying of the historical operation database of the power distribution system to obtain the historical operation data of the distribution box specifically comprises the following steps: Query the historical operation database and mark the historical power demand evaluation index of the distribution box in the previous time period; Based on the mark, historical operation data corresponding to the historical power demand assessment index is obtained. 5. The method for intelligent power regulation of a distribution box based on artificial intelligence according to claim 1 is characterized in that adjusting the real-time power demand based on the historical operation data to obtain an optimized power regulation strategy specifically comprises the following steps: Based on the historical operation data, obtaining a historical power demand evaluation index and historical adjustment measures of the distribution box; Compare the current real-time power demand with the historical power demand evaluation index, and mark the situations with the same comparison results as the same load situation; Based on the historical regulation measures of the same load conditions, adjust the output power curve of the corresponding distribution box; Based on the adjusted output power curve, the real-time power demand corresponding to the same load situation is adjusted to obtain an optimized power regulation strategy. 6. The method for intelligent power regulation of a distribution box based on artificial intelligence according to claim 1, characterized in that the method further comprises the following steps: Based on the output power curve in the optimized power regulation strategy and the preset power association rule, the main transformer, energy storage system and transmission line of the distribution box are associated to obtain associated power equipment; Based on the preset maintenance cycle of each power device, generate the estimated maintenance time of the corresponding associated power device; Based on the estimated maintenance time of the associated power equipment, adjusting the actual maintenance time of the corresponding associated power equipment; Based on the adjusted actual maintenance time of each power equipment, each component of the distribution box is regularly inspected and maintained to achieve intelligent management. 7. The method for intelligent power regulation of a distribution box based on artificial intelligence according to claim 1, characterized in that the method further comprises the following steps: Based on the power consumption information, establish a power consumption pattern profile of the distribution box; Obtain energy consumption data and operating status information of other intelligent systems related to the area where the distribution box is located; Correlate and analyze the power consumption pattern portrait, energy consumption data of other intelligent systems, and operating status information to identify the comprehensive energy consumption pattern in the area where the distribution box is located; The optimized power regulation strategy is adjusted based on the comprehensive energy consumption mode, and the adjustment information is fed back to the other intelligent systems. 8. An intelligent power regulation system for a distribution box based on artificial intelligence, characterized by comprising: A data collection module, used to collect power consumption data and auxiliary data of the distribution box, and obtain power consumption information of the distribution box based on the power consumption data and auxiliary data; A real-time power demand judgment module, used to judge the real-time power demand of the distribution box based on the power consumption information; The historical operation data query module is used to query the historical operation database of the power distribution system and obtain the historical operation data of the distribution box; A power regulation strategy generating module, used to adjust the real-time power demand based on the historical operation data to obtain an optimized power regulation strategy; The power regulation module is used to regulate the power of the distribution box according to the optimized power regulation strategy. 9. An electronic device, characterized in that it comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the intelligent power regulation method for a distribution box based on artificial intelligence described in any one of claims 1 to 7 are implemented. 10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the distribution box intelligent power regulation method based on artificial intelligence described in any one of claims 1 to 7 are implemented.