CN119670023B - Pump unit fault monitoring method based on multidimensional sensing fusion recognition - Google Patents

Abstract

The present invention relates to the technical field of fault monitoring, and discloses a method for monitoring pump unit faults based on multi-dimensional perception fusion identification, the method comprising: obtaining operation stage information of a key monitoring area, and determining a fault monitoring threshold of a pump unit to be monitored based on the operation stage information; analyzing historical operation records, judging whether it is necessary to optimize the fault monitoring threshold based on the analysis result, and if so, setting an optimization coefficient corresponding to the fault monitoring threshold, and obtaining an optimized fault monitoring threshold; parsing abnormal operating condition data, judging whether it is necessary to compensate the optimized fault monitoring threshold based on the analysis result, and if so, calculating an abnormal operating condition impact index of the abnormal operating condition data, and setting a compensation coefficient of the optimized fault monitoring threshold based on the abnormal operating condition impact index, and obtaining a compensated fault monitoring threshold; determining a fault warning level of the pump unit to be monitored according to the compensated fault monitoring threshold. The present invention ensures the accuracy and effectiveness of monitoring.

Description

A pump unit fault monitoring method based on multi-dimensional perception fusion recognition

Technical Field

The present invention relates to the technical field of fault monitoring, and in particular to a pump unit fault monitoring method based on multi-dimensional perception fusion identification.

Background Art

With the rapid development of industrial technology, pump units, as key equipment in various industrial processes, have a direct impact on the efficiency and safety of the entire production line due to their operating status. However, due to the complex and changeable working environment of pump units, they are often affected by multiple factors such as fluid media, temperature, and pressure, resulting in frequent failures. Traditional fault monitoring methods often rely on single sensor data, which makes it difficult to fully and accurately reflect the operating status of pump units.

Therefore, it is necessary to design a pump unit fault monitoring method based on multi-dimensional perception fusion recognition to solve the problems existing in current technology.

Summary of the invention

In view of this, the present invention proposes a pump unit fault monitoring method based on multi-dimensional perception fusion recognition, aiming to solve the problem that traditional fault monitoring methods in current technology often rely on single sensor data and are difficult to fully and accurately reflect the operating status of the pump unit.

The present invention proposes a pump unit fault monitoring method based on multi-dimensional perception fusion recognition, comprising the following steps:

S100: Determine a pump unit to be monitored, divide the pump unit to be monitored into monitoring areas, and obtain key monitoring areas and common monitoring areas; obtain operation stage information of the key monitoring areas, and determine a fault monitoring threshold of the pump unit to be monitored based on the operation stage information;

S200: extracting corresponding historical operation records from a historical operation record library based on the key monitoring area, the common monitoring area and the fault monitoring threshold, analyzing the historical operation records, and judging whether it is necessary to optimize the fault monitoring threshold based on the analysis result, and if so, setting the optimization coefficient corresponding to the fault monitoring threshold, and obtaining the optimized fault monitoring threshold;

S300: collecting abnormal operating condition data of the common monitoring area, analyzing the abnormal operating condition data, judging whether the optimized fault monitoring threshold needs to be compensated based on the analysis result, and if so, calculating the abnormal operating condition impact index of the abnormal operating condition data, and setting the compensation coefficient of the optimized fault monitoring threshold based on the abnormal operating condition impact index, and obtaining the compensated fault monitoring threshold;

S400: Determine the fault warning level of the pump unit to be monitored according to the compensation fault monitoring threshold.

Furthermore, the historical operation records are analyzed, and based on the analysis results, it is determined whether the fault monitoring threshold needs to be optimized, including:

Extracting the historical operation records of the key monitoring areas respectively and recording them as key historical operation records; extracting the historical operation records of the common monitoring areas and recording them as common historical operation records;

Parsing the key historical operation records to obtain key historical normal operation records and key historical abnormal operation records;

According to the key historical normal operation records and the key historical abnormal operation records, the abnormal occurrence frequency of the key monitoring area is obtained, and recorded as the key abnormal frequency;

Parsing the common historical operation records to obtain common historical normal operation records and common historical abnormal operation records;

According to the common historical normal operation records and the common historical abnormal operation records, the abnormal occurrence frequency of the common monitoring area is obtained, and recorded as the common abnormal frequency;

Whether the fault monitoring threshold needs to be optimized is determined according to the key abnormal frequency and the common abnormal frequency.

Further, judging whether the fault monitoring threshold needs to be optimized according to the key abnormal frequency and the common abnormal frequency includes:

Obtaining a ratio of the key abnormal frequency to the common abnormal frequency, recorded as an abnormal frequency ratio;

Comparing the abnormal frequency ratio with the abnormal frequency ratio threshold, and judging whether it is necessary to optimize the fault monitoring threshold according to the comparison result;

When the abnormal frequency ratio is greater than the abnormal frequency ratio threshold, determining to optimize the fault monitoring threshold;

When the abnormal frequency ratio is less than or equal to the abnormal frequency ratio threshold, it is determined not to optimize the fault monitoring threshold.

Furthermore, when setting the optimization coefficient corresponding to the fault monitoring threshold and obtaining the optimized fault monitoring threshold, it includes:

Setting an optimization coefficient interval, wherein the optimization coefficient interval includes a first optimization coefficient, a second optimization coefficient, and a third optimization coefficient;

Performing weighted summation on the key abnormal frequency and the common abnormal frequency to obtain a comprehensive abnormal frequency;

Comparing the comprehensive abnormal frequency with the first comprehensive abnormal frequency and the second comprehensive abnormal frequency, and determining the optimization coefficient corresponding to the fault monitoring threshold according to the comparison result; wherein the first comprehensive abnormal frequency is less than the second comprehensive abnormal frequency;

When the comprehensive abnormal frequency is less than the first comprehensive abnormal frequency, determining that the optimization coefficient corresponding to the fault monitoring threshold is the first optimization coefficient, and taking the product value of the first optimization coefficient and the fault monitoring threshold as the optimized fault monitoring threshold;

When the comprehensive abnormal frequency is greater than or equal to the first comprehensive abnormal frequency and less than the second comprehensive abnormal frequency, determining that the optimization coefficient corresponding to the fault monitoring threshold is the second optimization coefficient, and taking the product value of the second optimization coefficient and the fault monitoring threshold as the optimized fault monitoring threshold;

When the comprehensive abnormal frequency is greater than or equal to the second comprehensive abnormal frequency, the optimization coefficient corresponding to the fault monitoring threshold is determined to be the third optimization coefficient, and the product value of the third optimization coefficient and the fault monitoring threshold is used as the optimized fault monitoring threshold.

Furthermore, the abnormal operating condition data of the common monitoring area is collected, and the abnormal operating condition data is analyzed, and it is determined whether the optimized fault monitoring threshold needs to be compensated based on the analysis result, including:

Extracting features from the abnormal operating condition data to obtain abnormal operating condition feature data; wherein the abnormal operating condition feature data includes abnormal sound feature data, abnormal vibration feature data, and abnormal temperature feature data;

Acquire the abnormal operating condition characteristic value of each abnormal operating condition characteristic data, and the abnormal operating condition standard value corresponding to each abnormal operating condition characteristic value;

All the abnormal operating condition characteristic values are compared with the corresponding abnormal operating condition standard values, and it is determined whether the optimized fault monitoring threshold needs to be compensated based on the comparison result.

Furthermore, all the abnormal operating condition characteristic values are compared with the corresponding abnormal operating condition standard values, and it is determined whether the optimized fault monitoring threshold needs to be compensated according to the comparison result, including:

When all of the abnormal operating condition characteristic values are less than or equal to the corresponding abnormal operating condition standard values, it is determined that there is no need to compensate the optimized fault monitoring threshold;

When the abnormal operating condition characteristic value is greater than the corresponding abnormal operating condition standard value, it is determined that the optimized fault monitoring threshold needs to be compensated.

Further, when calculating the abnormal operating condition impact index of the abnormal operating condition data, it includes:

Performing weighted processing on each abnormal operating condition characteristic value that is greater than the corresponding abnormal operating condition standard value to obtain a weighted abnormal operating condition characteristic value;

All weighted abnormal operating condition characteristic values are summed up to obtain the abnormal operating condition impact index.

Further, when setting the compensation coefficient of the optimized fault monitoring threshold based on the abnormal operating condition impact index and obtaining the compensated fault monitoring threshold, it includes:

Comparing the abnormal operating condition impact index with historical data, setting a compensation coefficient of the optimized fault monitoring threshold according to the comparison result, and obtaining a compensated fault monitoring threshold;

When there is a historical abnormal operating condition impact index that is the same as the abnormal operating condition impact index in the historical data, the historical optimization coefficient corresponding to the historical abnormal operating condition impact index is used as the compensation coefficient, and the product value of the compensation coefficient and the optimized fault monitoring threshold is used as the compensated fault monitoring threshold;

When there is no historical abnormal operating condition impact index that is the same as the abnormal operating condition impact index in the historical data, the difference between the abnormal operating condition impact index and the historical data is calculated one by one, recorded as the abnormal difference, and an abnormal difference set is constructed. The compensation coefficient of the optimized fault monitoring threshold is set according to the abnormal difference set, and the compensated fault monitoring threshold is obtained.

Further, when the compensation coefficient of the optimized fault monitoring threshold is set according to the abnormal difference set and the compensated fault monitoring threshold is obtained, it includes:

Obtaining the minimum value of the abnormal difference values in the abnormal difference value set, and recording it as the minimum abnormal difference value;

Comparing the minimum abnormal difference with the first minimum abnormal difference and the second minimum abnormal difference, and determining the compensation coefficient of the optimized fault monitoring threshold according to the comparison result; wherein the first minimum abnormal difference is smaller than the second minimum abnormal difference;

When the minimum abnormal difference is less than the first minimum abnormal difference, determining that the compensation coefficient corresponding to the optimized fault monitoring threshold is the first compensation coefficient, and taking the product value of the first compensation coefficient and the optimized fault monitoring threshold as the compensated fault monitoring threshold;

When the minimum abnormal difference is greater than or equal to the first minimum abnormal difference and less than the second minimum abnormal difference, determining that the compensation coefficient corresponding to the optimized fault monitoring threshold is the second compensation coefficient, and taking the product value of the second compensation coefficient and the optimized fault monitoring threshold as the compensated fault monitoring threshold;

When the minimum abnormal difference is greater than or equal to the second minimum abnormal difference, the compensation coefficient corresponding to the optimized fault monitoring threshold is determined to be the third compensation coefficient, and the product value of the third compensation coefficient and the optimized fault monitoring threshold is used as the compensated fault monitoring threshold.

Further, when determining the fault warning level of the pump unit to be monitored according to the compensation fault monitoring threshold, it includes:

Comparing the compensation fault monitoring threshold with the first compensation fault monitoring threshold and the second compensation fault monitoring threshold, and determining the fault warning level of the pump unit to be monitored according to the comparison result; wherein the first compensation fault monitoring threshold is less than the second compensation fault monitoring threshold;

When the compensation fault monitoring threshold is less than the first compensation fault monitoring threshold, determining that the fault warning level of the pump unit to be monitored is a first warning level;

When the compensation fault monitoring threshold is greater than or equal to the first compensation fault monitoring threshold and less than the second compensation fault monitoring threshold, determining that the fault warning level of the pump unit to be monitored is a secondary warning level;

When the compensation fault monitoring threshold is greater than or equal to the second compensation fault monitoring threshold, determining that the fault warning level of the pump unit to be monitored is a third warning level;

Among them, the first-level warning level is lower than the second-level warning level, and the second-level warning level is lower than the third-level warning level.

Compared with the prior art, the beneficial effects of the present invention are as follows: the pump unit fault monitoring method based on multi-dimensional perception fusion recognition provided by the present invention realizes comprehensive control of the operating status of the pump unit by detailed division and monitoring of the key monitoring areas and general monitoring areas of the pump unit; a reasonable fault monitoring threshold is set according to the characteristics and operation stage information of the pump unit to ensure the accuracy and effectiveness of monitoring; dynamic optimization of the fault monitoring threshold is achieved through analysis of historical operation records, thereby improving the sensitivity and adaptability of monitoring; at the same time, in-depth analysis and compensation processing of abnormal operating condition data further enhances the accuracy and reliability of fault warning.

BRIEF DESCRIPTION OF THE DRAWINGS

Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

FIG1 is a flow chart of a method for monitoring pump unit faults based on multi-dimensional perception fusion and recognition according to an embodiment of the present invention.

DETAILED DESCRIPTION

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

Referring to FIG. 1 , in some embodiments of the present application, this embodiment provides a pump unit fault monitoring method based on multi-dimensional perception fusion recognition, comprising the following steps:

S100: Determine a pump unit to be monitored, divide the pump unit to be monitored into monitoring areas, and obtain key monitoring areas and common monitoring areas; obtain operation stage information of the key monitoring areas, and determine a fault monitoring threshold of the pump unit to be monitored based on the operation stage information;

S200: extracting corresponding historical operation records from a historical operation record library based on the key monitoring area, the common monitoring area and the fault monitoring threshold, analyzing the historical operation records, and judging whether it is necessary to optimize the fault monitoring threshold based on the analysis result, and if so, setting the optimization coefficient corresponding to the fault monitoring threshold, and obtaining the optimized fault monitoring threshold;

S300: collecting abnormal operating condition data of the common monitoring area, analyzing the abnormal operating condition data, judging whether the optimized fault monitoring threshold needs to be compensated based on the analysis result, and if so, calculating the abnormal operating condition impact index of the abnormal operating condition data, and setting the compensation coefficient of the optimized fault monitoring threshold based on the abnormal operating condition impact index, and obtaining the compensated fault monitoring threshold;

S400: Determine the fault warning level of the pump unit to be monitored according to the compensation fault monitoring threshold.

In this embodiment, the key monitoring areas include core components such as bearings, seals and motors of the pump unit; the common monitoring areas include inlet and outlet pipes, connectors and ancillary equipment of the pump unit.

In this embodiment, the operation stage information includes the startup stage, the stable operation stage and the shutdown stage of the pump unit.

In this embodiment, the fault monitoring threshold in the startup phase is preferably 1.5 times the vibration amplitude of the pump group, the fault monitoring threshold in the stable operation phase is preferably 1.2 times the vibration amplitude of the pump group, and the fault monitoring threshold in the shutdown phase is set to 1 times the vibration amplitude of the pump group. Such settings are based on the characteristics and fault probability of the pump group in different operation phases, aiming to more accurately capture potential fault signals and improve monitoring efficiency.

It can be understood that the pump unit fault monitoring method based on multi-dimensional perception fusion identification provided in this embodiment realizes comprehensive control of the operating status of the pump unit by detailed division and monitoring of the key monitoring areas and general monitoring areas of the pump unit; according to the characteristics and operation stage information of the pump unit, a reasonable fault monitoring threshold is set to ensure the accuracy and effectiveness of monitoring; through the analysis of historical operation records, the dynamic optimization of the fault monitoring threshold is realized, and the sensitivity and adaptability of monitoring are improved; at the same time, the in-depth analysis and compensation processing of abnormal operating condition data further enhances the accuracy and reliability of fault warning.

Specifically, analyzing the historical operation records and judging whether the fault monitoring threshold needs to be optimized based on the analysis results includes:

Extracting the historical operation records of the key monitoring areas respectively and recording them as key historical operation records; extracting the historical operation records of the common monitoring areas and recording them as common historical operation records;

Parsing the key historical operation records to obtain key historical normal operation records and key historical abnormal operation records;

According to the key historical normal operation records and the key historical abnormal operation records, the abnormal occurrence frequency of the key monitoring area is obtained, and recorded as the key abnormal frequency;

Parsing the common historical operation records to obtain common historical normal operation records and common historical abnormal operation records;

According to the common historical normal operation records and the common historical abnormal operation records, the abnormal occurrence frequency of the common monitoring area is obtained, and recorded as the common abnormal frequency;

Whether the fault monitoring threshold needs to be optimized is determined according to the key abnormal frequency and the common abnormal frequency.

In this embodiment, in the key historical operation records, the number of key historical normal operation records is recorded as the first number, and the number of key historical abnormal operation records is recorded as the second number, then the key abnormal frequency is equal to the ratio of the second number to the sum of the first number and the second number; in the ordinary historical operation records, the number of ordinary historical normal operation records is recorded as the third number, and the number of ordinary historical abnormal operation records is recorded as the fourth number, then the ordinary abnormal frequency is equal to the ratio of the fourth number to the sum of the third number and the fourth number.

Specifically, judging whether the fault monitoring threshold needs to be optimized according to the key abnormal frequency and the common abnormal frequency includes:

Obtaining a ratio of the key abnormal frequency to the common abnormal frequency, recorded as an abnormal frequency ratio;

Comparing the abnormal frequency ratio with the abnormal frequency ratio threshold, and judging whether it is necessary to optimize the fault monitoring threshold according to the comparison result;

When the abnormal frequency ratio is greater than the abnormal frequency ratio threshold, determining to optimize the fault monitoring threshold;

When the abnormal frequency ratio is less than or equal to the abnormal frequency ratio threshold, it is determined not to optimize the fault monitoring threshold.

It can be understood that the abnormal frequency ratio indicates the relative relationship between the abnormal occurrence frequency of the key monitoring area and the ordinary monitoring area, reflecting the stability and fault tendency of different parts of the pump unit during operation. When the abnormal occurrence frequency of the key monitoring area increases significantly relative to the ordinary monitoring area, it means that the key components may face a higher risk of failure. At this time, optimizing and adjusting the fault monitoring threshold can more effectively capture these potential fault signals and avoid missing or false alarms, thereby improving the accuracy and timeliness of pump unit fault monitoring. In addition, through the continuous optimization of the fault monitoring threshold, it can also adapt to the dynamic changes in the operating status of the pump unit to ensure the long-term effectiveness and stability of the monitoring system.

Specifically, setting the optimization coefficient corresponding to the fault monitoring threshold and obtaining the optimized fault monitoring threshold includes:

Setting an optimization coefficient interval, wherein the optimization coefficient interval includes a first optimization coefficient, a second optimization coefficient, and a third optimization coefficient;

Performing weighted summation on the key abnormal frequency and the common abnormal frequency to obtain a comprehensive abnormal frequency;

Comparing the comprehensive abnormal frequency with the first comprehensive abnormal frequency and the second comprehensive abnormal frequency, and determining the optimization coefficient corresponding to the fault monitoring threshold according to the comparison result; wherein the first comprehensive abnormal frequency is less than the second comprehensive abnormal frequency;

When the comprehensive abnormal frequency is less than the first comprehensive abnormal frequency, determining that the optimization coefficient corresponding to the fault monitoring threshold is the first optimization coefficient, and taking the product value of the first optimization coefficient and the fault monitoring threshold as the optimized fault monitoring threshold;

When the comprehensive abnormal frequency is greater than or equal to the first comprehensive abnormal frequency and less than the second comprehensive abnormal frequency, determining that the optimization coefficient corresponding to the fault monitoring threshold is the second optimization coefficient, and taking the product value of the second optimization coefficient and the fault monitoring threshold as the optimized fault monitoring threshold;

When the comprehensive abnormal frequency is greater than or equal to the second comprehensive abnormal frequency, the optimization coefficient corresponding to the fault monitoring threshold is determined to be the third optimization coefficient, and the product value of the third optimization coefficient and the fault monitoring threshold is used as the optimized fault monitoring threshold.

In this embodiment, the first optimization coefficient<the second optimization coefficient<the third optimization coefficient.

It is understandable that by setting different optimization coefficients, the fault monitoring threshold can be flexibly adjusted according to the abnormal frequency of the actual operating status of the pump unit, so as to achieve more sensitive and accurate capture of potential fault signals. When the comprehensive abnormal frequency is low, a smaller optimization coefficient is used to maintain the relative stability of the fault monitoring threshold and avoid false alarms; when the comprehensive abnormal frequency is high, a larger optimization coefficient is used to appropriately reduce the fault monitoring threshold, improve the response speed to potential faults, and ensure the timeliness and effectiveness of monitoring. This dynamic optimization strategy can significantly improve the intelligence level and practicality of pump unit fault monitoring.

Specifically, the abnormal operating condition data of the common monitoring area is collected, and the abnormal operating condition data is analyzed, and based on the analysis result, it is determined whether the optimized fault monitoring threshold needs to be compensated, including:

Extracting features from the abnormal operating condition data to obtain abnormal operating condition feature data; wherein the abnormal operating condition feature data includes abnormal sound feature data, abnormal vibration feature data, and abnormal temperature feature data;

Acquire the abnormal operating condition characteristic value of each abnormal operating condition characteristic data, and the abnormal operating condition standard value corresponding to each abnormal operating condition characteristic value;

All the abnormal operating condition characteristic values are compared with the corresponding abnormal operating condition standard values, and it is determined whether the optimized fault monitoring threshold needs to be compensated based on the comparison result.

It is understandable that abnormal operating condition data often contains important information about the operating status of the pump unit. Through in-depth analysis of these data, potential fault hazards can be discovered and handled in a timely manner. In this embodiment, abnormal sound characteristic data can reflect the operating status of the mechanical components inside the pump unit, such as bearing wear, seal leakage, etc.; abnormal vibration characteristic data can reveal the vibration of the pump unit during operation, which helps to determine whether there are problems such as imbalance and looseness; and abnormal temperature characteristic data can reflect the temperature distribution of various components of the pump unit, which is of great significance for discovering faults such as overheating and poor cooling.

When comparing the abnormal operating condition characteristic value with the abnormal operating condition standard value, if a certain abnormal operating condition characteristic value exceeds its corresponding abnormal operating condition standard value, the characteristic data is considered abnormal, which may indicate that the pump unit has corresponding fault hazards. At this time, in order to more accurately reflect the actual operating status of the pump unit, it is necessary to compensate and adjust the optimized fault monitoring threshold.

Specifically, all the abnormal operating condition characteristic values are compared with the corresponding abnormal operating condition standard values, and judging whether it is necessary to compensate the optimized fault monitoring threshold value according to the comparison result includes:

When all of the abnormal operating condition characteristic values are less than or equal to the corresponding abnormal operating condition standard values, it is determined that there is no need to compensate the optimized fault monitoring threshold;

When the abnormal operating condition characteristic value is greater than the corresponding abnormal operating condition standard value, it is determined that the optimized fault monitoring threshold needs to be compensated.

Specifically, the calculation of the abnormal operating condition impact index of the abnormal operating condition data includes:

Performing weighted processing on each abnormal operating condition characteristic value that is greater than the corresponding abnormal operating condition standard value to obtain a weighted abnormal operating condition characteristic value;

All weighted abnormal operating condition characteristic values are summed up to obtain the abnormal operating condition impact index.

It is understandable that the abnormal operating condition impact index comprehensively reflects the severity and potential impact of abnormal operating conditions in the general monitoring area, and is an important basis for determining whether it is necessary to compensate for the optimized fault monitoring threshold and the degree of compensation. Weighted processing can give different weights according to the importance and sensitivity of different abnormal operating condition characteristic values, so as to more accurately evaluate their impact on the operating status of the pump unit. By summing the weighted abnormal operating condition characteristic values, the abnormal operating condition impact index obtained can intuitively reflect the overall situation of the abnormal operating condition and provide strong support for subsequent optimization and adjustment.

Specifically, when the compensation coefficient of the optimized fault monitoring threshold is set based on the abnormal operating condition impact index and the compensated fault monitoring threshold is obtained, it includes:

Comparing the abnormal operating condition impact index with historical data, setting a compensation coefficient of the optimized fault monitoring threshold according to the comparison result, and obtaining a compensated fault monitoring threshold;

When there is a historical abnormal operating condition impact index that is the same as the abnormal operating condition impact index in the historical data, the historical optimization coefficient corresponding to the historical abnormal operating condition impact index is used as the compensation coefficient, and the product value of the compensation coefficient and the optimized fault monitoring threshold is used as the compensated fault monitoring threshold;

When there is no historical abnormal operating condition impact index that is the same as the abnormal operating condition impact index in the historical data, the difference between the abnormal operating condition impact index and the historical data is calculated one by one, recorded as the abnormal difference, and an abnormal difference set is constructed. The compensation coefficient of the optimized fault monitoring threshold is set according to the abnormal difference set, and the compensated fault monitoring threshold is obtained.

In this embodiment, the historical data includes a number of historical abnormal operating condition impact indexes.

Specifically, when the compensation coefficient of the optimized fault monitoring threshold is set according to the abnormal difference set and the compensated fault monitoring threshold is obtained, it includes:

Obtaining the minimum value of the abnormal difference values in the abnormal difference value set, and recording it as the minimum abnormal difference value;

Comparing the minimum abnormal difference with the first minimum abnormal difference and the second minimum abnormal difference, and determining the compensation coefficient of the optimized fault monitoring threshold according to the comparison result; wherein the first minimum abnormal difference is smaller than the second minimum abnormal difference;

When the minimum abnormal difference is less than the first minimum abnormal difference, determining that the compensation coefficient corresponding to the optimized fault monitoring threshold is the first compensation coefficient, and taking the product value of the first compensation coefficient and the optimized fault monitoring threshold as the compensated fault monitoring threshold;

When the minimum abnormal difference is greater than or equal to the first minimum abnormal difference and less than the second minimum abnormal difference, determining that the compensation coefficient corresponding to the optimized fault monitoring threshold is the second compensation coefficient, and taking the product value of the second compensation coefficient and the optimized fault monitoring threshold as the compensated fault monitoring threshold;

When the minimum abnormal difference is greater than or equal to the second minimum abnormal difference, the compensation coefficient corresponding to the optimized fault monitoring threshold is determined to be the third compensation coefficient, and the product value of the third compensation coefficient and the optimized fault monitoring threshold is used as the compensated fault monitoring threshold.

It is understandable that after determining that the optimized fault monitoring threshold needs to be compensated, the compensation coefficient of the optimized fault monitoring threshold is set based on the abnormal operating condition impact index. Specifically, a compensation coefficient interval is set, and the compensation coefficient interval includes a first compensation coefficient, a second compensation coefficient, and a third compensation coefficient, and the first compensation coefficient < second compensation coefficient < third compensation coefficient. The setting of the compensated fault monitoring threshold further improves the accuracy and flexibility of pump unit fault monitoring. Through in-depth analysis and compensation processing of abnormal operating condition data, it can more accurately reflect the actual operating status of the pump unit, timely discover and warn of potential fault hazards, and provide strong guarantees for the stable operation of the pump unit. At the same time, this fault monitoring method based on multi-dimensional perception fusion and identification integrates a variety of monitoring means and information, realizes comprehensive, accurate and timely monitoring of the operating status of the pump unit, and provides a scientific basis and technical support for the maintenance and management of the pump unit.

Specifically, when determining the fault warning level of the pump unit to be monitored according to the compensation fault monitoring threshold, it includes:

Comparing the compensation fault monitoring threshold with the first compensation fault monitoring threshold and the second compensation fault monitoring threshold, and determining the fault warning level of the pump unit to be monitored according to the comparison result; wherein the first compensation fault monitoring threshold is less than the second compensation fault monitoring threshold;

When the compensation fault monitoring threshold is less than the first compensation fault monitoring threshold, determining that the fault warning level of the pump unit to be monitored is a first warning level;

When the compensation fault monitoring threshold is greater than or equal to the first compensation fault monitoring threshold and less than the second compensation fault monitoring threshold, determining that the fault warning level of the pump unit to be monitored is a secondary warning level;

When the compensation fault monitoring threshold is greater than or equal to the second compensation fault monitoring threshold, determining that the fault warning level of the pump unit to be monitored is a third warning level;

Among them, the first-level warning level is lower than the second-level warning level, and the second-level warning level is lower than the third-level warning level.

It is understandable that the setting of fault warning levels is intended to issue warning signals of different levels in a timely manner according to the actual operating status of the pump unit, so that the operation and maintenance personnel can respond quickly and take corresponding treatment measures. The first-level warning level means that the pump unit is relatively loose, indicating that the operating status of the pump unit is basically normal, but regular inspections and maintenance are still required to ensure its long-term stable operation; the second-level warning level means that the pump unit has certain hidden dangers of failure, and the operation and maintenance personnel need to strengthen monitoring, pay close attention to the operating status of the pump unit, and be prepared to take corresponding maintenance or replacement measures; the third-level warning level indicates that the pump unit may have a serious risk of failure. At this time, measures should be taken immediately to conduct a comprehensive inspection of the pump unit, eliminate the fault in time, and avoid greater losses or accidents. Through the reasonable setting and timely adjustment of the fault warning level, it is possible to achieve refined management of the operating status of the pump unit, improve the operation and maintenance efficiency and fault handling capabilities, and provide strong guarantees for the safe and stable operation of the pump unit.

Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

The present application is described with reference to the flowcharts and/or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the processes and/or boxes in the flowchart and/or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device for implementing the functions specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims (9)

1. A pump unit fault monitoring method based on multidimensional sensing fusion identification is characterized by comprising the following steps:

Determining a pump unit to be monitored, dividing a monitoring area of the pump unit to be monitored to obtain a key monitoring area and a common monitoring area, acquiring operation phase information of the key monitoring area, and determining a fault monitoring threshold value of the pump unit to be monitored based on the operation phase information;

Extracting corresponding historical operation records from a historical operation record library based on the key monitoring area, the common monitoring area and the fault monitoring threshold value, analyzing the historical operation records, judging whether the fault monitoring threshold value needs to be optimized based on an analysis result, if so, setting an optimization coefficient corresponding to the fault monitoring threshold value, and obtaining an optimized fault monitoring threshold value;

Collecting abnormal working condition data of the common monitoring area, analyzing the abnormal working condition data, judging whether the optimized fault monitoring threshold is required to be compensated based on an analysis result, if so, calculating an abnormal working condition influence index of the abnormal working condition data, setting a compensation coefficient of the optimized fault monitoring threshold based on the abnormal working condition influence index, and obtaining a compensated fault monitoring threshold;

Determining a fault early warning grade of the pump unit to be monitored according to the compensation fault monitoring threshold value;

setting a compensation coefficient of the optimized fault monitoring threshold based on the abnormal working condition influence index, and when the compensated fault monitoring threshold is obtained, comprising:

Comparing the abnormal working condition influence index with historical data, setting a compensation coefficient of the optimized fault monitoring threshold according to a comparison result, and obtaining a compensated fault monitoring threshold;

When the historical data has the historical abnormal condition influence index which is the same as the abnormal condition influence index, taking a historical optimization coefficient corresponding to the historical abnormal condition influence index as the compensation coefficient, and taking the product value of the compensation coefficient and the optimization fault monitoring threshold as the compensation fault monitoring threshold;

when the historical data does not have the historical abnormal condition influence indexes which are the same as the abnormal condition influence indexes, calculating the differences of the abnormal condition influence indexes and the historical data one by one, recording the differences as abnormal differences, constructing an abnormal difference set, setting the compensation coefficient of the optimized fault monitoring threshold according to the abnormal difference set, and obtaining the compensated fault monitoring threshold.

2. The method for monitoring the pump unit fault based on the multidimensional sensing fusion recognition according to claim 1, wherein the step of analyzing the historical operation record and judging whether the fault monitoring threshold needs to be optimized based on the analysis result comprises the following steps:

extracting the history operation records of the common monitoring area and recording the history operation records as common history operation records;

analyzing the key history operation record to obtain a key history normal operation record and a key history abnormal operation record;

acquiring abnormal occurrence frequency of the key monitoring area according to the key history normal operation record and the key history abnormal operation record, and recording the abnormal occurrence frequency as key abnormal frequency;

analyzing the common historical operation record to obtain a common historical normal operation record and a common historical abnormal operation record;

Acquiring abnormal occurrence frequency of the common monitoring area according to the common historical normal operation record and the common historical abnormal operation record, and marking the abnormal occurrence frequency as common abnormal frequency;

judging whether the fault monitoring threshold value needs to be optimized according to the key abnormal frequency and the common abnormal frequency.

3. The method for monitoring the pump unit fault based on multidimensional sensing fusion recognition according to claim 2, wherein when judging whether the fault monitoring threshold needs to be optimized according to the key abnormal frequency and the common abnormal frequency, the method comprises the following steps:

acquiring the ratio of the key abnormal frequency to the common abnormal frequency, and marking the ratio as an abnormal frequency ratio;

Comparing the abnormal frequency ratio with an abnormal frequency ratio threshold, and judging whether the fault monitoring threshold needs to be optimized according to a comparison result;

when the abnormal frequency ratio is larger than the abnormal frequency ratio threshold, determining to optimize the fault monitoring threshold;

and when the abnormal frequency ratio is smaller than or equal to the abnormal frequency ratio threshold, judging that the fault monitoring threshold is not optimized.

4. The pump unit fault monitoring method based on multidimensional sensing fusion recognition according to claim 3, wherein when an optimization coefficient corresponding to the fault monitoring threshold is set and the optimized fault monitoring threshold is obtained, the method comprises the following steps:

setting an optimization coefficient interval, wherein the optimization coefficient interval comprises a first optimization coefficient, a second optimization coefficient and a third optimization coefficient;

The key abnormal frequency and the common abnormal frequency are weighted and summed to obtain a comprehensive abnormal frequency;

comparing the comprehensive abnormal frequency with a first comprehensive abnormal frequency and a second comprehensive abnormal frequency, and determining an optimization coefficient corresponding to the fault monitoring threshold according to a comparison result, wherein the first comprehensive abnormal frequency is smaller than the second comprehensive abnormal frequency;

When the comprehensive abnormal frequency is smaller than the first comprehensive abnormal frequency, determining an optimization coefficient corresponding to the fault monitoring threshold value as a first optimization coefficient, and taking the product value of the first optimization coefficient and the fault monitoring threshold value as the optimized fault monitoring threshold value;

When the comprehensive abnormal frequency is greater than or equal to the first comprehensive abnormal frequency and smaller than the second comprehensive abnormal frequency, determining an optimization coefficient corresponding to the fault monitoring threshold value as a second optimization coefficient, and taking the product value of the second optimization coefficient and the fault monitoring threshold value as the optimized fault monitoring threshold value;

and when the comprehensive abnormal frequency is greater than or equal to the second comprehensive abnormal frequency, determining an optimization coefficient corresponding to the fault monitoring threshold value as a third optimization coefficient, and taking the product value of the third optimization coefficient and the fault monitoring threshold value as the optimized fault monitoring threshold value.

5. The pump unit fault monitoring method based on multidimensional sensing fusion recognition according to claim 1, wherein the method is characterized by collecting abnormal working condition data of the common monitoring area, analyzing the abnormal working condition data, and judging whether the optimized fault monitoring threshold needs to be compensated based on an analysis result, and comprises the following steps:

Extracting features of the abnormal working condition data to obtain abnormal working condition feature data, wherein the abnormal working condition feature data comprises abnormal sound feature data, abnormal vibration feature data and abnormal temperature feature data;

obtaining an abnormal working condition characteristic value of each abnormal working condition characteristic data and an abnormal working condition standard value corresponding to each abnormal working condition characteristic value;

And comparing all the abnormal working condition characteristic values with corresponding abnormal working condition standard values, and judging whether the optimized fault monitoring threshold value needs to be compensated according to the comparison result.

6. The pump unit fault monitoring method based on multidimensional sensing fusion recognition according to claim 5, wherein comparing all the abnormal condition characteristic values with corresponding abnormal condition standard values, and judging whether the optimized fault monitoring threshold needs to be compensated according to the comparison result comprises:

when all the abnormal working condition characteristic values are smaller than or equal to the corresponding abnormal working condition standard values, judging that the optimized fault monitoring threshold value does not need to be compensated;

When the abnormal working condition characteristic value is larger than the corresponding abnormal working condition standard value, judging that the optimized fault monitoring threshold value needs to be compensated.

7. The method for monitoring pump unit faults based on multidimensional sensing fusion recognition of claim 6, when calculating an abnormal condition impact index of the abnormal condition data, comprising:

Weighting each abnormal working condition characteristic value larger than the corresponding abnormal working condition standard value to obtain a weighted abnormal working condition characteristic value;

And summing all weighted abnormal condition characteristic values to obtain the abnormal condition influence index.

8. The pump unit fault monitoring method based on multidimensional sensing fusion recognition according to claim 1, wherein when setting the compensation coefficient of the optimized fault monitoring threshold according to the abnormal difference value set and obtaining the compensated fault monitoring threshold, the method comprises the following steps:

obtaining the minimum value of the abnormal difference value in the abnormal difference value set and marking the minimum value as the minimum abnormal difference value;

Comparing the minimum abnormal difference value with a first minimum abnormal difference value and a second minimum abnormal difference value, and determining a compensation coefficient of the optimized fault monitoring threshold according to a comparison result, wherein the first minimum abnormal difference value is smaller than the second minimum abnormal difference value;

When the minimum abnormal difference value is smaller than the first minimum abnormal difference value, determining a compensation coefficient corresponding to the optimized fault monitoring threshold value as a first compensation coefficient, and taking the product value of the first compensation coefficient and the optimized fault monitoring threshold value as the compensated fault monitoring threshold value;

When the minimum abnormal difference value is larger than or equal to the first minimum abnormal difference value and smaller than the second minimum abnormal difference value, determining a compensation coefficient corresponding to the optimized fault monitoring threshold value as a second compensation coefficient, and taking the product value of the second compensation coefficient and the optimized fault monitoring threshold value as the compensated fault monitoring threshold value;

And when the minimum abnormal difference value is larger than or equal to the second minimum abnormal difference value, determining a compensation coefficient corresponding to the optimized fault monitoring threshold value as a third compensation coefficient, and taking the product value of the third compensation coefficient and the optimized fault monitoring threshold value as the compensated fault monitoring threshold value.

9. The method for monitoring the pump unit fault based on multidimensional sensing fusion recognition according to claim 1, wherein when determining the fault pre-warning level of the pump unit to be monitored according to the compensating fault monitoring threshold value, the method comprises the following steps:

Comparing the compensation fault monitoring threshold value with a first compensation fault monitoring threshold value and a second compensation fault monitoring threshold value, and determining a fault early warning level of the pump unit to be monitored according to a comparison result, wherein the first compensation fault monitoring threshold value is smaller than the second compensation fault monitoring threshold value;

When the compensation fault monitoring threshold value is smaller than the first compensation fault monitoring threshold value, determining that the fault early warning level of the pump unit to be monitored is a primary early warning level

When the compensation fault monitoring threshold value is larger than or equal to the first compensation fault monitoring threshold value and smaller than the second compensation fault monitoring threshold value, determining that the fault early warning level of the pump unit to be monitored is a secondary early warning level;

when the compensation fault monitoring threshold value is larger than or equal to the second compensation fault monitoring threshold value, determining that the fault early warning level of the pump unit to be monitored is a three-level early warning level;

The primary early warning level is smaller than the secondary early warning level, and the secondary early warning level is smaller than the tertiary early warning level.