Distributed energy storage power station fault diagnosis system and method and storage medium

Abstract

The invention discloses a distributed energy storage power station fault diagnosis system and method and a storage medium, and relates to the technical field of energy storage, and the method comprises the following steps: obtaining historical monitoring data and real-time monitoring data of an energy storage power station; dividing associated data packets through correlation analysis; performing digital processing on the associated data packet to obtain an associated function group; solving an intersection point of the correlation functions in the correlation function group to obtain an intersection point coordinate value, and calculating a distance value with the coordinate origin according to the intersection point coordinate value; determining a threshold range of a distance value according to the distance value obtained by processing the historical monitoring data, and analyzing the real-time monitoring data in the same step to obtain a distance value needing to be judged; whether the fault is diagnosed or not is determined by judging whether the distance value needing to be judged is within the threshold range of the distance value, fault omission caused by incomplete regular changes within the allowable range of monitoring data errors is avoided, and fault diagnosis is prospective.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage, and specifically to a fault diagnosis system, method, and storage medium for a distributed energy storage power station. Background Art

[0002] A distributed energy storage power station is an energy storage system with a decentralized layout, close to users or the demand side. By storing electrical energy and releasing it when needed, it can improve the flexibility and reliability of the power system. In existing distributed energy storage fault diagnosis systems, parameters are usually set as thresholds. When a certain parameter of the energy storage system exceeds the set parameter threshold, it is determined as a fault. For example, Chinese Invention Patent (CN118884257A) discloses a "fault diagnosis method for an energy storage power grid with distributed power access", which specifically discloses "based on an electrochemical characteristic diagnosis model and an electromagnetic characteristic diagnosis model, a preliminary diagnosis is performed on the energy storage battery pack to determine whether there is a fault in the energy storage battery pack, including the following steps: based on historical state data, set the electrochemical characteristic diagnosis threshold and the electromagnetic characteristic diagnosis threshold; when the electrochemical characteristic diagnosis value is less than or equal to the electrochemical characteristic diagnosis threshold, and at the same time the electromagnetic characteristic diagnosis value is less than or equal to the electromagnetic characteristic diagnosis threshold, it is determined that there is no fault in the energy storage battery pack; when the electrochemical characteristic diagnosis value is less than or equal to the electrochemical characteristic diagnosis threshold, and at the same time the electromagnetic characteristic diagnosis value is greater than the electromagnetic characteristic diagnosis threshold, it is determined that the electromagnetic field strength is abnormal and there is no fault in the energy storage battery pack; when the electrochemical characteristic diagnosis value is greater than the electrochemical characteristic diagnosis threshold, and at the same time the electromagnetic characteristic diagnosis value is greater than the electromagnetic characteristic diagnosis threshold, it is determined that multiple faults such as incorrect state of charge estimation, increased internal resistance, abnormal temperature, and abnormal electromagnetic field strength occur; when the electrochemical characteristic diagnosis value is greater than the electrochemical characteristic diagnosis threshold, and at the same time the electromagnetic characteristic diagnosis value is less than or equal to the electromagnetic characteristic diagnosis threshold, it is determined that multiple faults such as incorrect state of charge estimation, increased internal resistance, and abnormal temperature occur". In the above technical solution, once the monitoring data of the energy storage system is less than the set threshold due to abnormal values, although it is within the threshold range, long-term small-scale abnormalities will cause the faults of the energy storage system to slowly expand, which may lead to the inability to identify faults in real time and accurately, and further lead to the aggravation of faults, affecting the safety of the entire energy storage power station. Therefore, there is an urgent need for a real-time, accurate, and forward-looking distributed energy storage power station fault diagnosis system to solve the above technical problems. Summary of the Invention

[0003] The purpose of the present invention is to provide a fault diagnosis system, method, and storage medium for a distributed energy storage power station to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solution: a fault diagnosis method for a distributed energy storage power station, which obtains the historical monitoring data of the energy storage power station and receives the real-time monitoring data; According to the attributes of the monitoring data, the historical monitoring data is divided into relevance packets DP, and the real-time monitoring data is divided into the relevant relevance packets. The purpose is to facilitate the comparison and analysis of the real-time monitoring data with the historical monitoring data of the same attribute to ensure the reliability of the fault diagnosis; A plane rectangular coordinate system is established, and the historical monitoring data and the real-time monitoring data in the same relevance packet are digitized to obtain the correlation function group R and the correlation function group R ’ ; The historical monitoring data of a relevance packet is digitized to obtain a group of correlation function groups R, and the real-time monitoring data of a relevance packet is digitized to obtain a group of correlation function groups R ’ ; For any two adjacent timestamps of the same correlation function r in the correlation function group R, the intersection coordinate value CP is solved; If the same correlation function r of any two adjacent timestamps is completely parallel in the plane rectangular coordinate system, there is no intersection coordinate value. At this time, the historical monitoring data corresponding to the correlation function r of the later timestamp is discarded; Because in the actual analysis process, if the above situation exists, the historical monitoring data cannot be used as the comparison basis for the real-time monitoring data analysis because there is no abnormal situation; Calculate the distance value L between the intersection coordinate value CP and the coordinate origin. Because the intersection coordinate value of the same correlation function r of two adjacent timestamps reflects the non-fully regular change of the historical monitoring data of two adjacent timestamps, that is, the error change within the allowable range, and solving the distance value L between the intersection coordinate value and the coordinate origin can intuitively reflect whether this non-fully regular change is within the error allowable range, and based on the maximum distance value L max and the minimum distance value L min of the same correlation function r in the historical monitoring data, determine the distance range [L min , L max of the correlation function r. In the historical monitoring data, it is not determined that the energy storage power station has a fault. Then, it indicates that all non-fully regular changes of the historical monitoring data are within the allowable error range, that is, the threshold range can be set through the analysis results of the historical monitoring data; Solve between the correlation function r and the correlation function r ’ adjacent to the real-time monitoring data to obtain the intersection coordinate value CP ’ , and calculate the distance value L ’ between the intersection coordinate value CP and the coordinate origin’ ; Determine the distance value L ’ is within the distance range [L min , L max . If it is within the distance range, continue to analyze the received real-time monitoring data. If it is not within the distance range, diagnose the fault and give an alarm. When it is not within the distance range, it indicates that the real-time monitoring data may still be within the allowable error range. However, it has deviated from the corresponding historical monitoring data trajectory. In the long run, this will lead to the occurrence of a fault. Therefore, it is necessary to diagnose it as a fault and give an alarm for early maintenance.

[0005] According to the above technical solution, the number of historical monitoring data with the same timestamp in the associated data packet DP is n. A regular n-sided polygon is established based on the number of historical monitoring data n. After the real-time monitoring data is divided into the associated data packets it belongs to, through comparison and analysis, find the associated data packet corresponding to the historical monitoring data with the same attributes and quantity. The purpose is to ensure that when visualizing and positioning the historical monitoring data or real-time monitoring data, it is carried out under a unified standard. If the attributes or quantities are different, the probability of abnormal analysis and comparison between the historical monitoring data and the real-time monitoring data will increase, which will also lead to abnormal fault diagnosis.

[0006] According to the above technical solution, when performing digital processing on the historical monitoring data and real-time monitoring data in the same associated data packet, the following steps are included: Step1: Establish a plane rectangular coordinate system with the center of the regular n-sided polygon as the coordinate origin; Step2: Draw an extension line from the coordinate origin to any vertex angle of the regular n-sided polygon as the attribute line of the monitoring data; Step3: Locate the historical monitoring data and real-time monitoring data at the corresponding positions on the corresponding attribute lines according to the monitoring data attributes and monitoring data values; Step4: Connect the monitoring data of two different attributes of any group of monitoring data to obtain a function segment. Extend both ends of the function segment infinitely to obtain an associated function. Then, n*(n - 1) / 2 associated functions are obtained for a group of monitoring data, forming an associated function group, and obtaining an associated function group R of a group of historical monitoring data and an associated function group R of real-time monitoring data ’ .

[0007] According to the above technical solution, in the analysis of historical monitoring data, any two adjacent timestamps of the same associated function are respectively represented as r (x)-T and r (x)-T+t , where T represents the timestamp of historical monitoring data acquisition, and t represents the historical monitoring data acquisition time interval; For the associated function r (x)-T and the associated function r (x)-T+tSolve to obtain the intersection coordinate value CP = (x j , y j ) of the same correlation function at two adjacent timestamps; Calculate the distance value L between the intersection coordinate value CP = (x j , y j ) and the coordinate origin (0, 0). The distance value L is calculated using the formula [(x i - 0) 2 + (y i - 0) 2 1 / 2 Perform the calculation to obtain the set L 集 of the distance values of the same correlation function for any two adjacent timestamps in the historical monitoring data. Extract the maximum value L 集 and the minimum value L max in L min . Take [L min , L max as the distance value range threshold for this correlation function.

[0008] According to the above technical solution, determine the correlation data packet DP k of the real-time monitoring data to be analyzed, where k indicates that the real-time monitoring data belongs to the kth correlation data packet. Extract the historical monitoring data adjacent to the timestamp of the real-time monitoring data to be analyzed from the correlation data packet DP k ; Determine the correlation function group of this historical monitoring data as R k , and determine the correlation function group of the real-time monitoring data to be analyzed as R k ’ ; Among them, the correlation function group R k = {r k1 , r k2 , r k3 , …, r kn}, and the correlation function group R k ’ = {r k1 ’ , r k2 ’ , r k3 ’ , …, r kn ’}, where n represents n correlation functions; for example: n = 10, which means the real-time monitoring data contains 5 monitoring data with different attributes; Solve the same correlation function in the correlation function group R k and the correlation function group R k ’ respectively; ​Obtain the intersection coordinate values (x k-a , y k-a ) of any same correlation function, where a represents the a-th correlation function in the correlation function group R k and the correlation function group R k ’ ; Calculate the distance value between the intersection coordinate values (x k-a , y k-a ) and the coordinate origin (0, 0). The distance value L is calculated using the formula [(x k-a - 0) 2 + (y k-a - 0) 2 1 / 2 to obtain the distance value L k-a , and finally form a set L k of distance values in the correlation data packet DP 集 = {L k-a , L k-b , L k-c , …, L k-m}, where m represents m distance values finally calculated; Conduct a range analysis on the distance values in L 集 respectively, and determine whether the distance values in L 集 belong to [L min , L max . If there are distance values in L 集 that do not belong to the distance range, diagnose the fault and give an early warning.

[0009] Through the above technical solution, it can be found that this application does not directly determine whether there is a fault in the energy storage power station by setting the threshold of the monitoring data. Instead, after conducting a correlation analysis on the monitoring data, it conducts a step-by-step analysis through digital processing. Through the solution of this application, when the monitoring data shows non-fully regular anomalies, the fault can be diagnosed, that is, the monitoring data is within the allowable error range, but long-term non-fully regular anomalies will lead to final faults. Therefore, the above technical solution of this application is more forward-looking and avoids safety accidents in the energy storage power station caused by faults.

[0010] Distributed energy storage power station fault diagnosis system, the diagnosis system includes a data acquisition module, a data preprocessing module, a data analysis module, and a fault diagnosis module; ​The data acquisition module is used to acquire the historical monitoring data of the energy storage power station and receive the real-time monitoring data; the data preprocessing module is used to correlate the historical monitoring data and the real-time monitoring data and perform data preprocessing to obtain a number of associated function groups; the data analysis module is used to calculate and analyze the data processed by the data preprocessing module as the basis for fault diagnosis; the fault diagnosis module is used to diagnose the faults of the energy storage power station based on the analysis results of the data analysis module.

[0011] According to the above technical solution, the data acquisition module includes a historical monitoring data acquisition unit and a real-time monitoring data receiving unit; The historical monitoring data acquisition unit is used to acquire historical monitoring data from the database; the real-time monitoring data receiving unit is used to receive the real-time monitoring data of the energy storage power station and store the received real-time monitoring data into the database as historical monitoring data after the fault diagnosis is completed.

[0012] According to the above technical solution, the data preprocessing module includes a data association unit, a visualization processing unit, a coordinate system establishment unit, and a function generation unit; The data association unit correlates and binds the historical monitoring data and the real-time monitoring data according to the monitoring data attributes to obtain an associated data packet; the visualization processing unit is used to visualize the historical monitoring data and the real-time monitoring data in the associated data packet and establish a polygon according to the number of data attributes in the associated data packet; the coordinate system establishment unit is used to establish a plane rectangular coordinate system on the polygon processed by the visualization processing unit; the function generation unit is used to connect the different attribute data with the same time stamp in the historical monitoring data or the real-time monitoring data in the plane rectangular coordinate system to obtain a line segment function, and infinitely extend both ends of the line segment function to obtain an associated function, and the associated functions with the same time stamp form an associated function group.

[0013] According to the above technical solution, the data analysis module includes a function solving unit, a distance value calculation unit, and a range threshold determination unit; The function solving unit is used to solve the same associated function with different time stamps generated by the historical monitoring data in the associated data packet to obtain the intersection coordinate value CP, and is also used to solve the same associated function with adjacent time stamps respectively generated by the historical monitoring data and the real-time monitoring data in the same associated data packet to obtain the intersection coordinate value CP ’ ; the distance calculation unit calculates the distance values between the obtained intersection coordinate values CP and CP ’ and the coordinate origin to obtain the distance value L and the distance value L'; the range threshold determination unit determines the maximum distance value L max and the minimum distance value L of the same associated function in the historical monitoring datamin Determine the distance range [L min , L max for the correlation function r; The fault diagnosis module is used to determine whether the distance value L ’ is within the distance range [L min , L max . If it is within the range, there is no fault; if not, diagnose the fault of the energy storage power station and give an early warning.

[0014] A readable storage medium stores a computer program, which, when executed on a processor, implements the fault diagnosis method for a distributed energy storage power station.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention abandons the traditional form of directly setting a threshold value and comparing the monitoring data with the set threshold value to determine whether the energy storage power station is faulty. Instead, it uses digital and visual analysis of the monitoring data, so that even when the monitoring data is within the threshold range but there is a non-fully conventional change, it can be diagnosed as a fault. Moreover, in the actual data monitoring process of the energy storage power station, the long-term existence of such non-fully conventional changes will lead to the occurrence of faults. Therefore, the above-mentioned fault diagnosis method of the present invention has a certain degree of foresight and can perform fault diagnosis and maintenance before the occurrence of faults and when there are potential safety hazards, thus avoiding the occurrence of faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a schematic diagram of the logic flow of the fault diagnosis method for a distributed energy storage power station according to the present invention.

[0017] Figure 2 is a schematic diagram of the visual correlation function of the second embodiment of the fault diagnosis method for a distributed energy storage power station according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] Embodiment 1 As Figure 1 shown, the present invention provides a fault diagnosis method for a distributed energy storage power station, including the following steps: S1. Obtain the historical monitoring data of the energy storage power station and receive the real-time monitoring data; S2. Classify the historical monitoring data according to the attributes of the monitoring data to obtain the associated data packet DP. For example, there is a relationship among the current data, voltage data, temperature data, charging efficiency, and discharging efficiency of the energy storage battery in the energy storage power station. Therefore, the five can be classified into the same associated data packet, and the real-time monitoring data is classified into the associated data packet it belongs to. The purpose is to facilitate the comparison and analysis of the real-time monitoring data with the historical monitoring data of the same attribute to ensure the reliability of fault diagnosis; Specifically, the number of historical monitoring data with the same timestamp in the associated data packet DP is n, and a regular n-sided polygon is established based on the number of historical monitoring data n. For example, if the number of historical monitoring data in the 3rd associated data packet DP3 is 5, a regular pentagon is established. Here, it is emphasized that it must be the number of historical monitoring data with the same timestamp because only at the same timestamp can it be compared and analyzed with the real-time monitoring data. If the combination of historical monitoring data with different timestamps is compared with the real-time monitoring data, the comparison value of this application will be lost. After the real-time monitoring data is classified into the associated data packet it belongs to, through comparison and analysis, the associated data packet corresponding to the historical monitoring data with the same attribute and quantity is found. The purpose is to ensure that when visualizing and positioning the historical monitoring data or real-time monitoring data, it is carried out under a unified standard. If the attributes or quantities are different, the probability of abnormal analysis and comparison between the historical monitoring data and the real-time monitoring data will increase, which will lead to abnormal fault diagnosis.

[0020] S3. Establish a plane rectangular coordinate system, and digitally process the historical monitoring data and real-time monitoring data in the same associated data packet to obtain the associated function group R and the associated function group R respectively ’ ; The historical monitoring data of an associated data packet is digitally processed to obtain a set of associated function group R, and the real-time monitoring data of an associated data packet is digitally processed to obtain a set of associated function group R ’ ; Specifically, when digitally processing the historical monitoring data and real-time monitoring data in the same associated data packet, the following steps are included: Step1. Take the center of the regular n-sided polygon as the coordinate origin and establish a plane rectangular coordinate system; Step2. Extend a line from the coordinate origin to any vertex angle of the regular n-sided polygon as the attribute line of the monitoring data; Step3. Locate the historical monitoring data and real-time monitoring data at the corresponding positions on the corresponding attribute lines according to the monitoring data attributes and monitoring data values; Step 4. Connect the monitoring data of two different attributes in any group of monitoring data to obtain a function segment. Extend both ends of the function segment infinitely to obtain an associated function. Then, n*(n - 1) / 2 associated functions can be obtained from a group of monitoring data, forming an associated function group. Obtain the associated function group R of a group of historical monitoring data and the associated function group R of real-time monitoring data ’ .

[0021] In this embodiment, the number of monitoring data in the 3rd group of historical monitoring data and real-time monitoring data is 5 each. Then, both the associated function group R and the associated function group R ’ are composed of 10 associated functions, that is, R = {r1, r2, r3, …, r 10}, R ’ = {r1 ’ , r2 ’ , r3 ’ , …, r 10 ’}. The associated functions in the associated function group R and the associated function group R' are all associated functions generated at the same timestamp.

[0022] S4. Solve any two adjacent timestamps of the same associated function r in the associated function group R to obtain the intersection coordinate value CP; Specifically, in the analysis of historical monitoring data, the same associated functions of any two adjacent timestamps are respectively expressed as r (x)-T and r (x)-T+t , where T represents the timestamp of historical monitoring data collection, and t represents the historical monitoring data collection time interval; For example: r (x)-T represents the associated function generated from the current - temperature data collected at 12:10:10, and r (x)-T+t represents the associated function generated from the current - temperature data collected at 12:10:12, where t = 2s; Solve the associated function r (x)-T and the associated function r (x)-T+t to obtain the intersection coordinate value CP = (x j , y j ) of the same associated function of two adjacent timestamps; If the same associated function r of any two adjacent timestamps is completely parallel in the plane rectangular coordinate system, there is no intersection coordinate value. At this time, the historical monitoring data corresponding to the associated function r of the later timestamp is discarded; Because in the actual analysis process, if the above - mentioned situation exists, the historical monitoring data cannot be used as the comparison basis for real - time monitoring data analysis because there is no abnormal situation; S5. Calculate the distance value L between the intersection point coordinate value CP and the coordinate origin. Since the intersection point coordinate values of the same correlation function r for two adjacent timestamps reflect the non-fully regular changes in the historical monitoring data for the two adjacent timestamps, the non-fully regular changes refer to: the changes in the historical monitoring data are non-regular, not gradually increasing or decreasing, but are sudden changes, and such sudden changes are within the normal threshold range, that is, the error changes within the allowable range. By solving the distance value L between the intersection point coordinate value and the coordinate origin, it can intuitively reflect whether such non-fully regular changes are within the error allowable range, and based on the maximum distance value L and the minimum distance value L of the same correlation function r in the historical monitoring data max and the minimum distance value L min Determine the distance range [L min , L max of the correlation function r. In the historical monitoring data, it is not determined that there is a fault in the energy storage power station. Then, it indicates that all non-fully regular changes in the historical monitoring data are within the allowable error range, that is, the threshold range can be set based on the analysis results of the historical monitoring data. However, in the traditional abnormal judgment process, usually when the abnormal change does not exceed the set threshold, no warning will be generated. However, such abnormal changes for a long time or multiple times will also lead to the occurrence of faults. Therefore, it is particularly important to consider this situation; Specifically, calculate the distance value L between the intersection point coordinate value CP = (x j , y j ) and the coordinate origin (0, 0). The distance value L is calculated using the formula [(x i - 0) 2 + (y i - 0) 2 1 / 2 to calculate and obtain the set L 集 of the distance values of the same correlation function for any two adjacent timestamps in the historical monitoring data. Extract the maximum value L 集 and the minimum value L max in L min , and take [L min , L max as the threshold of the distance value range of this correlation function.

[0023] S6. Solve between the correlation function r adjacent to the real-time monitoring data and the correlation function r ’ to obtain the intersection point coordinate value CP ’ ​, for example: the current-temperature monitoring data is used to collect data from the energy storage power station every 2 seconds. The timestamp of this current-temperature real-time monitoring data collection is 12:02:02. Then, among the historical monitoring data, the correlation function r corresponding to the current-temperature historical monitoring data with a timestamp of 12:02:00 is used as the correlation function r corresponding to the current-temperature real-time monitoring data with a timestamp of 12:02:02 ’ as the object to be solved, and calculate the intersection coordinate value CP ’ the distance value L from the origin of coordinates ’ ; Specifically, determine the correlation data packet DP of the real-time monitoring data to be analyzed k , where k indicates that the real-time monitoring data belongs to the kth correlation data packet. Extract the historical monitoring data adjacent to the timestamp of the real-time monitoring data to be analyzed from the correlation data packet DP k ; For example: the timestamp of the real-time monitoring data in the correlation data packet DP k is 12:25:32, and the data collection time interval of the monitoring data is 2s. Then, the timestamp of the historical monitoring data extracted from the correlation data packet DP k is 12:25:30; Determine the correlation function group of this historical monitoring data as R k , and determine the correlation function group of the real-time monitoring data to be analyzed as R k ’ ; Among them, the correlation function group R k = {r k1 , r k2 , r k3 , …, r kn}, the correlation function group R k ’ = {r k1 ’ , r k2 ’ , r k3 ’ , …, r kn ’}, where n represents n correlation functions; for example: n = 10, which means the real-time monitoring data contains 5 monitoring data with different attributes; Solve the same correlation function in the correlation function group R k and the correlation function group R k ’ respectively; Obtain the intersection coordinate value (x k-a , y k-a ) of any same correlation function, where a represents the correlation function group Rk and the associated function group R k ’ the a-th associated function in Calculate the distance value between the intersection coordinate values (x k-a , y k-a ) and the coordinate origin (0, 0). The distance value L is calculated using the formula [(x k-a -0) 2 +(y k-a -0) 2 1 / 2 to obtain the distance value L k-a , and finally form the associated data packet DP k the set L of distance values in 集 ={L k-a , L k-b , L k-c ,…, L k-m}, where m represents m distance values finally calculated; S7. Determine whether the distance value L ’ is within the distance range [L min , L max . If it is within the distance range, continue to analyze the received real-time monitoring data. If it is not within the distance range, diagnose the fault and give an early warning. When it is not within the distance range, it indicates that the real-time monitoring data may still be within the error tolerance range, but it has deviated from the corresponding historical monitoring data trajectory. In the long run, this will lead to faults. Therefore, it needs to be diagnosed as a fault and an early warning given for early maintenance.

[0024] This application determines abnormal conditions through the distance value between the intersection coordinate value and the origin. The reason is that when the change amount of a set of adjacent monitoring data is within the threshold range, no early warning will be generated. However, in the digital processing process of this application, the monitoring data is transformed into the form of an extended function. On an extended line, if the monitoring data shows non-fully regular changes, that is, mutations occur, then there must be intersections between the extended functions corresponding to the associated functions. When the mutation change amount is larger, the extended functions of the associated functions will inevitably have a larger offset amount, and then the distance value between it and the coordinate origin will inevitably increase.

[0025] Specifically, perform a range analysis on the distance values in L 集 to determine whether the distance values in L 集 belong to [L min , L max . If there are distance values in L 集 that do not belong to the distance range, diagnose the fault and give an early warning.

[0026] ​Through the above technical solution, it can be found that this application does not directly determine whether there is a fault in the energy storage power station by setting the threshold of the monitoring data. Instead, after performing a correlation analysis on the monitoring data, it analyzes step by step through digital processing. Through the solution of this application, when the monitoring data shows non-fully regular anomalies, the fault can be diagnosed, that is, the monitoring data is within the allowable error range, but the long-term non-fully regular anomalies will lead to the final fault. Therefore, the above technical solution of this application is more forward-looking and avoids safety accidents in the energy storage power station caused by the occurrence of faults.

[0027] Embodiment 2 As Figure 2 shown, the data attributes in the associated data packet are: current, voltage, temperature, charging efficiency, and discharging efficiency, a total of five; Through the analysis of historical monitoring data, it is known that the distance range value is [2.16, 8.52]; The time stamp of the historical monitoring data is 12:32:18, and the generated set of correlation functions is R = {r1, r2, r3, …, r 10}; The time stamp of the real-time monitoring data is 12:32:20, and the generated set of correlation functions is R ’ = {r1 ’ , r2 ’ , r3 ’ , …, r 10 ’}; Solve the intersection coordinate values of the same correlation function in the set of correlation functions R and the set of correlation functions R', and obtain the set CP 集 ’ = {cp1 ’ , cp2 ’ , cp3 ’ , …, cp 10 ’}; Calculate the distance values between the set CP 集 ’ of the intersection coordinate values and the origin of coordinates respectively, and obtain the set of distance values: L 集 ’ = {3.25, 5.58, 5.83, 2.58, 4.32, 7.65, 10.02, 5.65, 6.11, 4.69}; Compare the distance values in L 集 ’ with the distance range value [2.16, 8.52], and it can be found that there is a distance value 10.02 that does not belong to the distance range value, then diagnose the fault and give an early warning.

[0028] Embodiment 3 A readable storage medium stores a computer program, and when the computer program is executed on a processor, a distributed energy storage power station fault diagnosis method is implemented.

[0029] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved.

Claims

1. A fault diagnosis method for a distributed energy storage power station, characterized in that: Obtain the historical monitoring data of the energy storage power station and receive the real-time monitoring data; According to the attributes of the monitoring data, divide the historical monitoring data into relevant data packets DP, and divide the real-time monitoring data into the corresponding relevant data packets; Establish a plane rectangular coordinate system, and perform digital processing on the historical monitoring data and real-time monitoring data in the same associated data packet to obtain the associated function group R and the associated function group R respectively ’ ; Solve the same correlation function r with any two adjacent timestamps in the correlation function group R to obtain the intersection coordinate value CP; Calculate the distance value L between the intersection point coordinate value CP and the coordinate origin, and based on the maximum distance value L of the same correlation function r in the historical monitoring data max and the minimum distance value L min Determine the distance range [L min , L max ; Solve between the correlation function r and the correlation function r of adjacent timestamps with the real-time monitoring data to obtain the intersection coordinate value CP ’ and calculate the distance value L between the intersection coordinate value CP ’ and the origin of coordinates ’ ; ’ ; Determine the distance value L ’ Is it within the distance range [L min , L max ? If it is within the distance range, continue to analyze the received real-time monitoring data. If it is not within the distance range, diagnose the fault and give an early warning.

2. The fault diagnosis method for a distributed energy storage power station according to claim 1, characterized in that: The number of historical monitoring data with the same timestamp in the relevant data packet DP is n. A regular n-sided polygon is established based on the number of historical monitoring data n. After the real-time monitoring data is divided into the corresponding relevant data packets, through comparison and analysis, find the relevant data packet corresponding to the historical monitoring data with the same attributes and quantity.

3. The fault diagnosis method of the distributed energy storage power station according to claim 2, wherein: When performing digital processing on the historical monitoring data and real-time monitoring data in the same relevant data packet, it includes the following steps: Step1: Establish a plane rectangular coordinate system with the center of the regular n-sided polygon as the coordinate origin; Step2: Extend a line from the coordinate origin to any vertex angle of the regular n-sided polygon as the attribute line of the monitoring data; Step3: Locate the historical monitoring data and real-time monitoring data at the corresponding positions on the corresponding attribute lines according to the attributes and values of the monitoring data; Step 4. Connect the monitoring data of two different attributes in any group of monitoring data to obtain a function segment. Extend the two ends of the function segment infinitely to obtain an associated function. Then, n*(n - 1) / 2 associated functions can be obtained from a group of monitoring data, forming an associated function group. Obtain the associated function group R of a group of historical monitoring data and the associated function group R of real-time monitoring data ’ .

4. The fault diagnosis method of the distributed energy storage power station according to claim 1, wherein: In the analysis of historical monitoring data, the same correlation function for any two adjacent timestamps is respectively denoted as r (x)-T and r (x)-T+t , where T represents the timestamp of historical monitoring data collection, and t represents the time interval of historical monitoring data collection; For the correlation function r (x)-T and the correlation function r (x)-T+t Solve to obtain the intersection coordinate values CP=(x j , y j ) of the same correlation function at two adjacent timestamps; Calculate the distance value L between the intersection coordinate value CP = (x j , y j ) and the coordinate origin (0, 0), and obtain the set L of the same correlation function distance values for any two adjacent timestamps in the historical monitoring data 集 . Extract the maximum value L 集 and the minimum value L max in L min . Take [L min , L max as the distance value range threshold of this correlation function.

5. The fault diagnosis method for a distributed energy storage power station according to claim 1, wherein: Determine the associated data packet DP of the real-time monitoring data to be analyzed k , where k indicates that the real-time monitoring data belongs to the k-th associated data packet, and extract the historical monitoring data adjacent to the timestamp of the real-time monitoring data to be analyzed from the associated data packet DP k ; Determine that the associated function group of the historical monitoring data is R k , determine that the associated function group of the real-time monitoring data to be analyzed is R k ’ ; Solve the same correlation function in the correlation function groups R k and the correlation function groups R k ’ respectively; Obtain the intersection coordinate values (x k-a , y k-a ) of any same correlation function, where a represents the a-th correlation function in the correlation function group R k and the correlation function group R k ’ . Calculate the distance value between the intersection point coordinate values (x k-a , y k-a ) and the coordinate origin (0, 0) to obtain the distance value L k-a . Finally, form the associated data packet DP k with the set of distance values L 集 = {L k-a , L k-b , L k-c , …, L k-m}, where m represents m distance values finally calculated; Perform range analysis on the distance values in L 集 respectively, and determine whether the distance values in L 集 belong to [L min , L max . If there are distance values in L 集 that do not fall within the distance range, diagnose the fault and give an early warning.

6. A distributed energy storage power station fault diagnosis system for implementing the distributed energy storage power station fault diagnosis method according to any one of claims 1-5, characterized in that: The diagnostic system includes a data acquisition module, a data preprocessing module, a data analysis module, and a fault diagnosis module; The data acquisition module is used to obtain the historical monitoring data of the energy storage power station and receive the real-time monitoring data; the data preprocessing module is used to correlate the historical monitoring data and real-time monitoring data and perform data preprocessing to obtain several correlation function groups; the data analysis module is used to calculate and analyze the data processed by the data preprocessing module as the basis for fault diagnosis; the fault diagnosis module is used to diagnose the faults of the energy storage power station based on the analysis results of the data analysis module.

7. The fault diagnosis system for a distributed energy storage power station according to claim 6, wherein: The data acquisition module includes a historical monitoring data acquisition unit and a real-time monitoring data receiving unit; The historical monitoring data acquisition unit is used to obtain the historical monitoring data from the database; the real-time monitoring data receiving unit is used to receive the real-time monitoring data of the energy storage power station and store the received real-time monitoring data into the database as historical monitoring data after the fault diagnosis is completed.

8. The distributed energy storage power station fault diagnosis system according to claim 6, wherein: The data preprocessing module includes a data correlation unit, a visualization processing unit, a coordinate system establishment unit, and a function generation unit; The data correlation unit correlates and binds the historical monitoring data and real-time monitoring data according to the attributes of the monitoring data to obtain relevant data packets; the visualization processing unit is used to visually process the historical monitoring data and real-time monitoring data in the relevant data packets and establish a polygon based on the quantity of data attributes in the relevant data packets; the coordinate system establishment unit is used to establish a plane rectangular coordinate system on the polygon processed by the visualization processing unit; the function generation unit is used to connect the different attribute data with the same timestamp in the historical monitoring data or real-time monitoring data in the plane rectangular coordinate system to obtain a line segment function, and infinitely extend both ends of the line segment function to obtain a correlation function. The correlation functions with the same timestamp form a correlation function group.

9. The distributed energy storage power station fault diagnosis system according to claim 6, wherein: The data analysis module includes a function solving unit, a distance value calculation unit, and a range threshold determination unit; The function solving unit is used to solve the same correlation function with different timestamps generated from the historical monitoring data in the associated data packet to obtain the intersection coordinate value CP, and is also used to solve the same correlation function with adjacent timestamps respectively generated from the historical monitoring data and the real-time monitoring data in the same associated data packet to obtain the intersection coordinate value CP ’ ; The distance calculation unit calculates the distance values between the obtained intersection coordinate values CP and CP ’ and the origin of coordinates to obtain the distance values L and L'; The range threshold determination unit determines the distance range [L max and the distance minimum value L min of the correlation function r according to the distance maximum value L min , L max ; The fault diagnosis module is used to judge the distance value L ’ whether it is within the distance range [L min , L max . If it is within the range, there is no fault; if not, the energy storage power station fault is diagnosed and a warning is issued.

10. A readable storage medium, characterized in that: It stores a computer program which, when executed on a processor, implements the distributed energy storage power station fault diagnosis method according to any one of claims 1-5.

Images