A fault diagnosis method, system, device and storage medium of an energy storage system

Abstract

The application provides a fault diagnosis method, system, device and storage medium of an energy storage system. The method comprises: acquiring first measurement point data, second measurement point data, third measurement point data and fourth measurement point data; calling a preset AI time sequence algorithm and multiple mechanism models to perform data processing on the first measurement point data, the second measurement point data, the third measurement point data and the fourth measurement point data respectively to obtain first diagnosis results, second diagnosis results, third diagnosis results and fourth diagnosis results; and generating a target diagnosis result corresponding to the energy storage system according to the first diagnosis results, the second diagnosis results, the third diagnosis results, the fourth diagnosis results and real-time operation data of a power grid side. In the application, the measurement point data corresponding to a system level is processed by calling the preset AI time sequence algorithm and the multiple mechanism models to generate a target diagnosis result that can accurately represent the system level to which a fault belongs, so that accurate fault tracing is realized.

Description

Technical Field

[0001] This invention relates to the field of equipment management technology, and in particular to a fault diagnosis method, system, device and storage medium for an energy storage system. Background Technology

[0002] Energy storage systems are composed of a deeply coupled battery management system (BMS), power conversion system (PCS), energy management system (EMS), and grid connection system. The interconnection between the various energy storage devices in an energy storage system is complex, and the faults of energy storage systems generally exhibit characteristics of cross-system conduction, explicit suddenness, implicit gradual change, and strong correlation of operating conditions.

[0003] Currently, the common fault diagnosis method for energy storage systems is the single-device independent threshold alarm mode. That is, for each energy storage device, a separate current threshold, voltage threshold, and temperature threshold are set, and the fault of the energy storage device is determined based on the above current threshold, voltage threshold, and temperature threshold.

[0004] Therefore, existing fault diagnosis methods for energy storage systems can only identify obvious faults that have already occurred, and fault location is limited to the equipment as a whole, which cannot achieve accurate fault tracing. Furthermore, independent diagnosis of each device and isolated data lead to a high false alarm rate and false negative rate for coupled faults. Summary of the Invention

[0005] The main objective of this invention is to provide a fault diagnosis method, system, device, and storage medium for energy storage systems. It aims to address the technical problems of existing fault diagnosis methods for energy storage systems, such as the inability to accurately pinpoint the fault source and the high false alarm and false negative rates associated with coupled faults.

[0006] To achieve the above objectives, the present invention provides a fault diagnosis method for an energy storage system. The method is applied to an energy storage system, which includes a battery management system, a power conversion system, an energy management system, and a grid connection system. The method includes: Acquire the first measurement point data corresponding to the battery management system, the second measurement point data corresponding to the power conversion system, the third measurement point data corresponding to the energy management system, and the fourth measurement point data corresponding to the grid connection system; By calling preset AI timing algorithms and multiple mechanism models, data processing is performed on the first measurement point data, the second measurement point data, the third measurement point data, and the fourth measurement point data, respectively, to obtain the first diagnostic result corresponding to the battery management system, the second diagnostic result corresponding to the power conversion system, the third diagnostic result corresponding to the energy management system, and the fourth diagnostic result corresponding to the grid connection system. Based on the first diagnostic result, the second diagnostic result, the third diagnostic result, the fourth diagnostic result, and the real-time operation data of the power grid, a target diagnostic result corresponding to the energy storage system is generated. The target diagnostic result is used to characterize the system level to which the fault belongs.

[0007] Optionally, the plurality of mechanism models include a first mechanism model corresponding to the battery management system, a second mechanism model corresponding to the power conversion system, a third mechanism model corresponding to the energy management system, and a fourth mechanism model corresponding to the grid-connected system; The process involves calling a preset AI timing algorithm and multiple mechanism models to process the data at the first, second, third, and fourth measurement points, respectively, to obtain a first diagnostic result corresponding to the battery management system, a second diagnostic result corresponding to the power conversion system, a third diagnostic result corresponding to the energy management system, and a fourth diagnostic result corresponding to the grid-connected system, including: The first mechanism model is called to calculate the standard voltage of a single cell, the theoretical voltage difference of equalization, and the theoretical temperature rise of the module. The AI ​​time-series algorithm is called to identify the hardware characteristics of the battery. Based on the standard voltage of the single cell, the theoretical voltage difference of equalization, the theoretical temperature rise of the module, and the hardware characteristics of the battery, a first diagnostic result is generated. The second mechanism model is called to calculate the theoretical loss of the device, the theoretical fluctuation of the bus voltage, and the theoretical temperature difference of heat dissipation. The AI ​​time series algorithm is called to identify the characteristics of the power device. Based on the theoretical loss of the device, the theoretical fluctuation of the bus voltage, the theoretical temperature difference of heat dissipation, and the characteristics of the power device, a second diagnostic result is generated. The third mechanism model is invoked to calculate the difference between the instruction issued and the execution feedback value. The AI ​​time series algorithm is invoked to identify communication characteristics. Based on the difference and the communication characteristics, a third diagnostic result is generated. The fourth mechanism model is invoked to calculate the power difference between the theoretical power and the actual power. The AI ​​time series algorithm is invoked to identify grid connection characteristics. Based on the power difference and the grid connection characteristics, a fourth diagnostic result is generated.

[0008] Optionally, after generating the target diagnostic results corresponding to the energy storage system, the method further includes: Based on the fault type represented by the target diagnostic result, the first diagnostic result, the second diagnostic result, the system level and duration of the fault represented by the target diagnostic result, it is determined whether the fault corresponding to the energy storage system is a self-healing fault. If the fault corresponding to the energy storage system is a self-healing fault, then a self-healing strategy is executed on the energy storage system. If the fault corresponding to the energy storage system is not a self-healing fault, then graded fault handling is performed on the energy storage system.

[0009] Optionally, determining whether the fault corresponding to the energy storage system is a self-healing fault based on the fault type characterized by the target diagnostic result, the first diagnostic result, the second diagnostic result, and the system level and duration of the fault characterized by the target diagnostic result includes: Under the condition that the preset conditions are met, the fault corresponding to the energy storage system is determined to be a self-healing fault; If the preset conditions are not met, it is determined that the fault corresponding to the energy storage system is not a self-healing fault; The preset conditions include: The fault type characterized by the target diagnostic results is not a hardware fault, and; The degradation score determined based on the first diagnostic result and the second diagnostic result is less than or equal to the preset score, and; The fault represented by the target diagnostic result belongs to a single system level, and the duration of the fault represented by the target diagnostic result is less than a preset time.

[0010] Optionally, the implementation of a self-healing strategy for the energy storage system includes: When the fault represented by the target diagnostic result belongs to the battery management system, adjust the battery active balancing start-stop strategy, adjust the single cell overvoltage protection threshold, adjust the single cell undervoltage protection threshold, and balance the charging compensation voltage difference. If the fault described by the target diagnostic result belongs to the power conversion system, then calibrate the voltage sampling zero point, calibrate the current sampling zero point, correct the SPWM modulation control parameters, restart the heat dissipation control module, and adjust the fan speed range. If the fault represented by the target diagnostic result belongs to the energy management system, reset the upper and lower limits of the scheduling power, reset the charging and discharging timing control parameters, switch the scheduling control strategy, reset the abnormal command issuance logic, and restart the communication link. When the fault described in the target diagnostic result belongs to the grid-connected system, the load of multiple units is reorganized and allocated, the grid-connected voltage parameters are corrected, the grid-connected power parameters are corrected, and the reactive power compensation logic of the grid connection point is adjusted.

[0011] Optionally, the step of performing graded fault handling on the energy storage system includes: Based on the target diagnostic result, the first diagnostic result, and the second diagnostic result, determine the fault level corresponding to the fault represented by the target diagnostic result; If the fault characterized by the target diagnosis result is a level one fault, then the energy storage system will be shut down and disconnected from the grid, and an emergency repair prompt will be sent to the outside world. If the fault characterized by the target diagnostic result is a level two fault, then the energy storage system will be subjected to derating, capacity limiting and power reduction, and a maintenance reminder will be sent to the outside world; If the fault characterized by the target diagnostic result is a level three fault, then the energy storage system will be tracked and monitored.

[0012] Optionally, determining the fault level corresponding to the fault represented by the target diagnostic result based on the target diagnostic result, the first diagnostic result, and the second diagnostic result includes: The fault level weight is determined based on the system level to which the fault belongs, as characterized by the target diagnostic results. The hardware damage weight is determined based on the degree of hardware damage characterized by the first diagnostic result and the second diagnostic result. Based on the fault propagation risk characterized by the target diagnostic results, determine the safety risk weights; Based on the downtime loss risk characterized by the target diagnostic results, the downtime loss weight is determined; The fault level weight, hardware damage weight, safety risk weight, and downtime loss weight are weighted and summed to obtain the comprehensive fault risk score. If the comprehensive fault risk score is in the first interval, then the fault represented by the target diagnosis result is determined to be a level one fault. If the comprehensive fault risk score is in the second interval, then the fault represented by the target diagnostic result is determined to be a level two fault. If the comprehensive fault risk score is in the third interval, then the fault represented by the target diagnostic result is determined to be a level three fault. The lowest score in the first interval is higher than the highest score in the second interval, and the lowest score in the second interval is higher than the highest score in the third interval.

[0013] In addition, to achieve the above objectives, the present invention also provides an energy storage system. The energy storage system includes a processing module, a battery management system, a power conversion system, an energy management system, and a grid connection system. The processing module is communicatively connected to the battery management system, the power conversion system, the energy management system, and the grid connection system, respectively. The processing module is used to acquire the first measurement point data corresponding to the battery management system, the second measurement point data corresponding to the power conversion system, the third measurement point data corresponding to the energy management system, and the fourth measurement point data corresponding to the grid connection system. By calling preset AI timing algorithms and multiple mechanism models, data processing is performed on the first measurement point data, the second measurement point data, the third measurement point data, and the fourth measurement point data, respectively, to obtain the first diagnostic result corresponding to the battery management system, the second diagnostic result corresponding to the power conversion system, the third diagnostic result corresponding to the energy management system, and the fourth diagnostic result corresponding to the grid connection system. Based on the first diagnostic result, the second diagnostic result, the third diagnostic result, the fourth diagnostic result, and the real-time operation data of the power grid, a target diagnostic result corresponding to the energy storage system is generated. The target diagnostic result is used to characterize the system level to which the fault belongs.

[0014] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution: The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the fault diagnosis method for the energy storage system proposed in any of the embodiments of this application.

[0015] To address the aforementioned technical problems, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the fault diagnosis method for the energy storage system proposed in any of the embodiments of this application.

[0016] This application provides a fault diagnosis method, system, device, and storage medium for an energy storage system. The method is applied to an energy storage system, which includes a battery management system, a power conversion system, an energy management system, and a grid-connected system. The method includes: acquiring first measurement point data corresponding to the battery management system, second measurement point data corresponding to the power conversion system, third measurement point data corresponding to the energy management system, and fourth measurement point data corresponding to the grid-connected system; calling a preset AI time-series algorithm and multiple mechanism models to process the first, second, third, and fourth measurement point data respectively, obtaining a first diagnostic result corresponding to the battery management system, a second diagnostic result corresponding to the power conversion system, a third diagnostic result corresponding to the energy management system, and a fourth diagnostic result corresponding to the grid-connected system; and generating a target diagnostic result corresponding to the energy storage system based on the first, second, third, and fourth diagnostic results and real-time grid-side operating data, whereby the target diagnostic result characterizes the system level to which the fault belongs. In this application, by calling a preset AI time series algorithm and multiple mechanism models, the measurement point data corresponding to each system level is processed to generate target diagnostic results that can accurately characterize the system level to which the fault belongs, thereby achieving accurate fault tracing and reducing the false alarm rate and false negative rate of coupled faults. Attached Figure Description

[0017] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the fault diagnosis method for an energy storage system provided in an embodiment of this application; Figure 2 This is a schematic diagram of the application process of the fault diagnosis method for the energy storage system provided in the embodiments of this application; Figure 3 This is a schematic diagram of the energy storage system provided in the embodiments of this application; Figure 4 This is a basic structural block diagram of a computer device provided in the embodiments of this application. Detailed Implementation

[0019] All technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or accompanying drawings of this application are used to distinguish different objects, not to describe a particular order.

[0020] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] Please see Figure 1 This document illustrates a flowchart of an embodiment of a fault diagnosis method for an energy storage system according to this application. The fault diagnosis method for an energy storage system provided in this application is applied to an energy storage system, which includes a battery management system, a power conversion system, an energy management system, and a grid-connected system. The method includes the following steps: Step 101: Obtain the first measurement point data corresponding to the battery management system, the second measurement point data corresponding to the power conversion system, the third measurement point data corresponding to the energy management system, and the fourth measurement point data corresponding to the grid connection system.

[0022] The aforementioned measurement data includes, but is not limited to: electrical parameters, such as voltage, current, power, and frequency; temperature data, such as IGBT temperature, bus temperature, cell temperature, and ambient temperature; equipment operating status, such as start / stop status, fault codes, and operating modes; dispatching instructions, such as power dispatch values ​​and charging / discharging sequence instructions; grid connection data, such as harmonic content, power factor, and grid connection point voltage; and communication messages, such as communication delay, packet loss rate, and frame verification results.

[0023] In this step, after acquiring the data from the first, second, third, and fourth measurement points, preprocessing can be performed on this data. Specifically, the preprocessing includes filtering and denoising, outlier removal, multi-source time series alignment, and sampling deviation calibration. Through this preprocessing, data distortion caused by instantaneous power grid disturbances, line sampling noise, and communication packet loss can be effectively eliminated, ensuring that the input data for the four-layer hierarchical diagnostics and subsequent AI time series prediction models is synchronized, accurate, effective, stable, and controllable.

[0024] Step 102: Invoke the preset AI timing algorithm and multiple mechanism models to process the first measurement point data, the second measurement point data, the third measurement point data and the fourth measurement point data respectively, to obtain the first diagnostic result corresponding to the battery management system, the second diagnostic result corresponding to the power conversion system, the third diagnostic result corresponding to the energy management system and the fourth diagnostic result corresponding to the grid connection system.

[0025] In this step, a preset AI time-series algorithm and multiple mechanistic models are invoked to process the acquired first, second, third, and fourth measurement point data, respectively, to obtain independent diagnostic results for each level. For specific implementation details, please refer to subsequent embodiments.

[0026] Among them, multiple mechanism models include a first mechanism model corresponding to the battery management system, a second mechanism model corresponding to the power conversion system, a third mechanism model corresponding to the energy management system, and a fourth mechanism model corresponding to the grid connection system.

[0027] It should be noted that the mechanistic model diagnosis and AI time-series algorithm diagnosis at each level can be executed synchronously using parallel computing to shorten the overall diagnosis latency. Each level independently outputs the initial judgment results of the fault type, degradation degree, and fault location, providing basic data for hierarchical diagnosis for subsequent cross-validation.

[0028] Step 103: Based on the first diagnostic result, the second diagnostic result, the third diagnostic result, the fourth diagnostic result, and the real-time operation data of the power grid, generate the target diagnostic result corresponding to the energy storage system. The target diagnostic result is used to characterize the system level to which the fault belongs.

[0029] The system integrates the first, second, third, and fourth diagnostic results output in step S102, as well as the real-time operating data from the power grid side. The operating data includes power grid voltage, power grid frequency, grid connection point status, and dispatch plan values. Multi-unit joint cross-verification and comprehensive analysis are then performed.

[0030] Specifically: (1) If the BMS layer reports abnormal battery voltage, the PCS layer reports DC overvoltage, and the EMS layer reports power command exceeding limit at the same time, the root cause can be determined by timing alignment comparison: whether it is an erroneous command issued by the EMS, a PCS regulation failure, or a sudden rise in grid voltage.

[0031] (2) If there is a diagnostic anomaly in only a single level, while the corresponding data in other levels are normal and the power grid is operating smoothly, then it is determined to be a single-level equipment failure.

[0032] (3) If all levels report anomalies and are accompanied by fluctuations in power grid data, it is determined to be a chain response caused by external power grid disturbance.

[0033] (4) If the abnormal value is repeatedly changed within the normal steady-state fluctuation range of the equipment and there is no continuous deterioration trend, it is determined to be a normal operating condition fluctuation and no fault alarm is triggered.

[0034] In this step, through the multi-source data linkage comparison and verification described above, it is possible to distinguish between inherent faults in the equipment itself, abnormalities in scheduling logic control, instantaneous disturbances in the external power grid, and fluctuations in the normal operating conditions of the unit. This eliminates the problems of false alarms, missed alarms, and duplicate alarms caused by independent diagnosis at a single level. Target diagnostic results corresponding to the energy storage system are generated. These results characterize the system level to which the fault belongs, the fault type, and the precise location of the root cause, enabling accurate fault localization.

[0035] In this embodiment, by calling a preset AI time-series algorithm and multiple mechanism models, the measurement point data corresponding to each system level is processed to generate a target diagnostic result that can accurately characterize the system level to which the fault belongs, thereby achieving accurate fault tracing and reducing the false alarm rate and false negative rate of coupled faults.

[0036] Optionally, the plurality of mechanism models include a first mechanism model corresponding to the battery management system, a second mechanism model corresponding to the power conversion system, a third mechanism model corresponding to the energy management system, and a fourth mechanism model corresponding to the grid connection system; The process involves calling a preset AI timing algorithm and multiple mechanism models to process the data at the first, second, third, and fourth measurement points, respectively, to obtain a first diagnostic result corresponding to the battery management system, a second diagnostic result corresponding to the power conversion system, a third diagnostic result corresponding to the energy management system, and a fourth diagnostic result corresponding to the grid-connected system, including: The first mechanism model is called to calculate the standard voltage of a single cell, the theoretical voltage difference of equalization, and the theoretical temperature rise of the module. The AI ​​time-series algorithm is called to identify the hardware characteristics of the battery. Based on the standard voltage of the single cell, the theoretical voltage difference of equalization, the theoretical temperature rise of the module, and the hardware characteristics of the battery, a first diagnostic result is generated. The second mechanism model is called to calculate the theoretical loss of the device, the theoretical fluctuation of the bus voltage, and the theoretical temperature difference of heat dissipation. The AI ​​time series algorithm is called to identify the characteristics of the power device. Based on the theoretical loss of the device, the theoretical fluctuation of the bus voltage, the theoretical temperature difference of heat dissipation, and the characteristics of the power device, a second diagnostic result is generated. The third mechanism model is invoked to calculate the difference between the instruction issued and the execution feedback value. The AI ​​time series algorithm is invoked to identify communication characteristics. Based on the difference and the communication characteristics, a third diagnostic result is generated. The fourth mechanism model is invoked to calculate the power difference between the theoretical power and the actual power. The AI ​​time series algorithm is invoked to identify grid connection characteristics. Based on the power difference and the grid connection characteristics, a fourth diagnostic result is generated.

[0037] In this embodiment, a first mechanism model is invoked, which includes a battery electrochemical equivalent model, a single-cell equilibrium voltage difference model, and a heat accumulation conduction model. The battery electrochemical equivalent model outputs the theoretical calculated value of the standard voltage of a single cell, the single-cell equilibrium voltage difference model calculates the equilibrium theoretical voltage difference between any two cells, and the heat accumulation conduction model calculates the theoretical temperature rise of the module under specific charge and discharge conditions.

[0038] By employing AI time-series algorithms and taking the full life-cycle voltage sequence, temperature sequence, and capacity estimation sequence as input, the system identifies battery hardware characteristics such as slow voltage drift, continuous temperature difference expansion, and gradual capacity decay through time-series feature extraction and cluster analysis.

[0039] The standard voltage of a single cell, the theoretical voltage difference of equilibrium, and the theoretical temperature rise of the module output by the first mechanism model are subtracted from the actual collected values ​​one by one to obtain various deviations. Each deviation is then weighted and fused with the battery hardware features identified by the AI ​​algorithm to accurately identify battery-level faults such as electrical anomalies of single cells, potential heat accumulation hazards, consistency degradation, and capacity decay, and to generate the first diagnostic result.

[0040] In this embodiment, a second mechanism model is invoked, which includes an IGBT loss model, a DC bus RC voltage regulation model, and a thermal resistance model. The IGBT loss model calculates the theoretical device loss of the IGBT; the DC bus RC voltage regulation model calculates the theoretical bus voltage fluctuation; and the thermal resistance model calculates the theoretical temperature difference.

[0041] By employing AI timing algorithms and taking IGBT temperature, bus ripple voltage, cooling fan speed, and modulation deviation sequence as inputs, characteristics of power devices such as slow aging, capacitor value decay, and continuous decline in heat dissipation performance are obtained.

[0042] The deviation between the theoretical device loss, theoretical bus voltage fluctuation, and theoretical heat dissipation temperature difference output by the second mechanism model and the actual collected values ​​is calculated, and the power device characteristics identified by the AI ​​algorithm are integrated to generate a second diagnostic result.

[0043] In this embodiment, a third mechanism model is invoked, which is constructed based on the scheduling instruction issuance logic, power execution response logic, and multi-unit power allocation logic mechanism. This model calculates the difference between the instruction issuance value and the execution feedback value at the mechanism level.

[0044] The AI ​​timing algorithm is invoked, taking the scheduling instruction timing sequence, power response curve sequence, and communication delay sequence as inputs, to identify communication characteristics such as long-term instruction offset, continuous increase in communication delay, and gradual increase in power response lag.

[0045] The difference calculated by the third mechanism model is combined with the communication characteristics identified by the AI ​​algorithm to generate a third diagnostic result.

[0046] In this embodiment, the fourth mechanism model is invoked. This model is constructed based on the power conservation mechanism of the entire station, the grid-connected stability mechanism, and the multi-unit collaborative energy efficiency mechanism. It is used to compare the difference between the theoretical power balance value and the actual operating power.

[0047] By employing AI time-series algorithms and taking long-term power series, grid connection point voltage series, and multi-unit load distribution series as inputs, the system identifies grid connection characteristics such as long-term power offset across the entire station, slow degradation of grid-connected harmonics, and long-term imbalance in load distribution among multiple units.

[0048] Finally, the power difference calculated by the fourth mechanism model is fused and analyzed with the grid connection characteristics identified by the AI ​​algorithm to generate the fourth diagnostic result.

[0049] This embodiment integrates the mechanistic models at each level with AI time-series algorithms for diagnosis, fully combining the dual advantages of strong interpretability of physical mechanisms and deep feature mining capabilities of AI algorithms. It relies on physical mechanisms to achieve accurate tracing of device-level latent faults, and relies on AI algorithms to capture subtle signs of equipment degradation, complete intelligent fault classification and remaining life prediction, overcome the performance shortcomings of single technical solutions, and achieve integrated intelligent upgrade of diagnosis and prediction.

[0050] Optionally, after generating the target diagnostic results corresponding to the energy storage system, the method further includes: Based on the fault type represented by the target diagnostic result, the first diagnostic result, the second diagnostic result, the system level and duration of the fault represented by the target diagnostic result, it is determined whether the fault corresponding to the energy storage system is a self-healing fault. If the fault corresponding to the energy storage system is a self-healing fault, then a self-healing strategy is executed on the energy storage system. If the fault corresponding to the energy storage system is not a self-healing fault, then graded fault handling is performed on the energy storage system.

[0051] In this embodiment, after obtaining the target diagnostic result, the first diagnostic result, and the second diagnostic result, the system level and duration of the fault represented by the target diagnostic result are used to determine whether the fault corresponding to the energy storage system is a self-healing fault. For specific implementation details, please refer to the following embodiments.

[0052] If the fault corresponding to the energy storage system is a self-healing fault, then a self-healing strategy is executed for the energy storage system. Specifically, if the fault belongs to the BMS battery layer, then the battery-level self-healing strategy is executed; if the fault belongs to the PCS converter layer, then the PCS-level self-healing strategy is executed; if the fault belongs to the EMS scheduling layer, then the EMS-level self-healing strategy is executed; if the fault belongs to the system level, then the system-level self-healing strategy is executed.

[0053] If the fault corresponding to the energy storage system is not a self-healing fault, then a graded fault handling procedure will be performed on the energy storage system. For specific implementation details, please refer to subsequent embodiments. In this embodiment, by determining whether the fault corresponding to the energy storage system is a self-healing fault, recoverable soft faults are guided to the self-healing channel, and irreversible hard faults are guided to the graded handling channel. This achieves precision and differentiation in fault handling, minimizes unnecessary downtime operations, and ensures the operating benefits of the energy storage power station.

[0054] Optionally, determining whether the fault corresponding to the energy storage system is a self-healing fault based on the fault type characterized by the target diagnostic result, the first diagnostic result, the second diagnostic result, and the system level and duration of the fault characterized by the target diagnostic result includes: Under the condition that the preset conditions are met, the fault corresponding to the energy storage system is determined to be a self-healing fault; If the preset conditions are not met, it is determined that the fault corresponding to the energy storage system is not a self-healing fault; The preset conditions include: The fault type characterized by the target diagnostic results is not a hardware fault, and; The degradation score determined based on the first diagnostic result and the second diagnostic result is less than or equal to the preset score, and; The fault represented by the target diagnostic result belongs to a single system level, and the duration of the fault represented by the target diagnostic result is less than a preset time.

[0055] In this embodiment, if all preset conditions are met, the fault corresponding to the energy storage system is determined to be a self-healing fault; if any preset condition is not met, the fault corresponding to the energy storage system is determined to be a non-self-healing fault.

[0056] The aforementioned preset conditions include: Condition 1: The fault type characterized by the target diagnostic results is not a hardware fault.

[0057] The fault types characterized by the target diagnostic results are non-hardware faults such as communication link faults, scheduling parameter faults, sampling faults, logic control faults, and battery management faults.

[0058] Condition 2: The degradation score is less than or equal to the preset score.

[0059] The degradation severity scores for each diagnostic item are extracted from both the first and second diagnostic results. All diagnostic items are scored from 0 to 100, where 0 represents complete normality with no degradation and 100 represents extremely severe degradation. The maximum or weighted average of all degradation scores is taken to obtain the overall degradation score. If the overall degradation score is less than or equal to a preset score, then condition two is satisfied. Optionally, the preset score is 30 points.

[0060] Condition 3: The fault represented by the target diagnosis result belongs to a single system level, and the duration of the fault represented by the target diagnosis result is less than the preset time.

[0061] Extract the system level information of the fault and the duration of the fault from the target diagnostic results. If the fault occurs only at a single level among the system level, EMS scheduling level, PCS converter level, and BMS battery level, and the fault duration is less than a preset time, then condition three is met. Optionally, the preset time is 10 seconds.

[0062] This embodiment determines whether a fault is self-healing based on three dimensions: fault type, degree of degradation, duration, and scope, thereby ensuring that soft faults with self-healing feasibility enter the autonomous self-healing channel.

[0063] Optionally, the implementation of a self-healing strategy for the energy storage system includes: When the fault represented by the target diagnostic result belongs to the battery management system, adjust the battery active balancing start-stop strategy, adjust the single cell overvoltage protection threshold, adjust the single cell undervoltage protection threshold, and balance the charging compensation voltage difference. If the fault described by the target diagnostic result belongs to the power conversion system, then calibrate the voltage sampling zero point, calibrate the current sampling zero point, correct the SPWM modulation control parameters, restart the heat dissipation control module, and adjust the fan speed range. If the fault represented by the target diagnostic result belongs to the energy management system, reset the upper and lower limits of the scheduling power, reset the charging and discharging timing control parameters, switch the scheduling control strategy, reset the abnormal command issuance logic, and restart the communication link. When the fault described in the target diagnostic result belongs to the grid-connected system, the load of multiple units is reorganized and allocated, the grid-connected voltage parameters are corrected, the grid-connected power parameters are corrected, and the reactive power compensation logic of the grid connection point is adjusted.

[0064] In this embodiment, when the fault represented by the target diagnostic result belongs to the BMS battery layer, the following autonomous repair operation is performed: Adjust the active battery balancing start / stop strategy: Based on the current voltage distribution of each cell, adaptively adjust the start and stop voltage thresholds of the active balancing circuit to make the balancing action more compatible with the current battery pack consistency state and eliminate small voltage deviations caused by improper balancing strategy settings. Adjust the overvoltage and undervoltage protection thresholds for individual cells: Based on the battery aging level and the current temperature environment, fine-tune the overvoltage / undervoltage protection thresholds within the safe allowable range to avoid frequent false alarms caused by the mismatch between the fixed thresholds and the actual working characteristics of the battery.

[0065] Equalizing charging to compensate for voltage difference: For battery cells with small voltage difference deviations, active equalization circuits are used to perform charging or discharging compensation to restore the voltage of each cell to a balanced state.

[0066] In this embodiment, when the system level to which the fault represented by the target diagnostic result belongs is the PCS converter level, the following autonomous repair operation is performed: Calibrate the voltage sampling zero point and calibrate the current sampling zero point: Recalibrate the zero point offset coefficient of the sampling channel to eliminate sampling deviation; Correcting SPWM modulation control parameters: The SPWM modulation parameters are adaptively corrected based on the theoretical calculation values ​​of the mechanism model, so that the output waveform quality is restored to the normal level.

[0067] Restart the thermal control module: Automatically execute the software reset operation of the thermal control module to reset the fan control logic.

[0068] Adjust fan speed range: Adaptively adjust the upper and lower limits of the fan PWM speed control duty cycle based on the current equipment temperature rise and ambient temperature.

[0069] In this embodiment, when the fault represented by the target diagnostic result belongs to the EMS scheduling layer, the following autonomous repair operation is performed: Reset scheduling power upper and lower limits: Restore the EMS scheduling power upper and lower limit parameters to the factory default values ​​or the most recent valid operating values.

[0070] Reset charging and discharging timing control parameters: Resets the EMS charging and discharging timing control parameters to their default valid values.

[0071] Switching scheduling control strategy: If the current scheduling strategy encounters an execution error, it will automatically switch to the redundant backup scheduling strategy.

[0072] Reset Abnormal Instruction Issuance Logic: In case of deadlock or transition abnormality in the state machine of instruction issuance logic, perform software logic reset operation to reset the instruction issuance process.

[0073] Restart communication link: Automatically executes communication protocol stack reset and link reconnection operations, restarts handshake interaction process, and restores normal communication.

[0074] When the fault characterized by the target diagnostic results belongs to the system level, the following autonomous repair operations are performed: Reorganizing and redistributing loads across multiple units: Based on the current operating status and rated capacity of each unit, the optimal load distribution ratio is recalculated, and load reorganization commands are automatically issued to enable each unit to bear a balanced load.

[0075] Correct grid-connected voltage and power parameters: Automatically adjust the voltage reference value and active power reference value of the reactive power compensation device of the entire station to restore the operating parameters of the grid connection point to the range required by the dispatching.

[0076] Adjust the reactive power compensation logic at the grid connection point: automatically switch the reactive power compensation control mode to restore the power quality at the grid connection point to a qualified level.

[0077] This embodiment achieves precise autonomous repair of recoverable soft faults at different levels by setting different self-healing strategies for each level of fault, without the need for manual intervention, and greatly shortens the fault recovery cycle.

[0078] Optionally, the step of performing graded fault handling on the energy storage system includes: Based on the target diagnostic result, the first diagnostic result, and the second diagnostic result, determine the fault level corresponding to the fault represented by the target diagnostic result; If the fault characterized by the target diagnosis result is a level one fault, then the energy storage system will be shut down and disconnected from the grid, and an emergency repair prompt will be sent to the outside world. If the fault characterized by the target diagnostic result is a level two fault, then the energy storage system will be subjected to derating, capacity limiting and power reduction, and a maintenance reminder will be sent to the outside world; If the fault characterized by the target diagnostic result is a level three fault, then the energy storage system will be tracked and monitored.

[0079] In this embodiment, the fault level corresponding to the fault represented by the target diagnostic result is determined based on the target diagnostic result, the first diagnostic result, and the second diagnostic result. For specific implementation details, please refer to the following embodiments.

[0080] In this embodiment, if the fault characterized by the target diagnosis result is determined to be a level one fault, the following actions are taken: shut down the energy storage system, disconnect it from the grid, send an emergency repair notice to the outside world, and prohibit the equipment from being put into operation.

[0081] Typical scenarios for Level 1 faults include, but are not limited to: the BMS battery layer reporting a battery thermal runaway warning, the PCS converter layer reporting an IGBT power device breakdown short circuit, a DC bus short circuit, and the system level reporting a serious grid-connected power exceeding the limit or a grid-connected point voltage collapse.

[0082] In this embodiment, if the fault characterized by the target diagnostic result is determined to be a level two fault, the following actions are taken: the energy storage system is derated, its capacity is limited, and its power is reduced; maintenance reminders are sent to external parties; and hardware replacement is completed within a specified window period.

[0083] Typical scenarios for Level 2 faults include, but are not limited to: PCS converter layer reporting significant IGBT aging, BMS battery layer reporting severe degradation of battery module consistency, and DC bus capacitor value decaying to below 70% of the rated value.

[0084] In this embodiment, if the fault characterized by the target diagnostic result is determined to be a Level 3 fault, the following actions are taken: Tracking and monitoring the energy storage system; marking the equipment health record: marking the faulty equipment as a "Level 3 potential fault" and its details in the equipment's full lifecycle health management record as a reference for daily operation and maintenance inspections; automatically including the equipment in the key inspection targets when generating daily inspection task sheets, prompting maintenance personnel to conduct on-site visual inspections and auxiliary tests; and selecting an opportune time for shutdown and maintenance: the system does not force an immediate shutdown or throttling, and the equipment continues to operate normally. The system automatically selects an appropriate time (such as a low grid load period or a planned shutdown window) in subsequent scheduling strategies to recommend a shutdown and maintenance time window to maintenance personnel.

[0085] Typical scenarios for Level 3 faults include, but are not limited to: permanent damage to a single voltage sampling channel in the PCS converter layer, failure of a single temperature sampling point in the BMS battery layer, and slight degradation of the performance of a local cooling fan.

[0086] In this embodiment, by setting different handling methods for each level of fault, namely alarm, capacity limit, and tripping three-level differentiated handling methods, the traditional extensive mode of forced shutdown is abandoned, and the commercial benefits of power generation, arbitrage, and frequency regulation of the energy storage system are preserved to the maximum extent, thus achieving a dynamic optimal balance between equipment safety and power plant economic benefits.

[0087] Optionally, determining the fault level corresponding to the fault represented by the target diagnostic result based on the target diagnostic result, the first diagnostic result, and the second diagnostic result includes: The fault level weight is determined based on the system level to which the fault belongs, as characterized by the target diagnostic results. The hardware damage weight is determined based on the degree of hardware damage characterized by the first diagnostic result and the second diagnostic result. Based on the fault propagation risk characterized by the target diagnostic results, determine the safety risk weights; Based on the downtime loss risk characterized by the target diagnostic results, the downtime loss weight is determined; The fault level weight, hardware damage weight, safety risk weight, and downtime loss weight are weighted and summed to obtain the comprehensive fault risk score. If the comprehensive fault risk score is in the first interval, then the fault represented by the target diagnosis result is determined to be a level one fault. If the comprehensive fault risk score is in the second interval, then the fault represented by the target diagnostic result is determined to be a level two fault. If the comprehensive fault risk score is in the third interval, then the fault represented by the target diagnostic result is determined to be a level three fault. The lowest score in the first interval is higher than the highest score in the second interval, and the lowest score in the second interval is higher than the highest score in the third interval.

[0088] In this embodiment, the fault level weight is determined according to the system level to which the fault belongs as characterized by the target diagnosis result. For example, if the fault belongs to the BMS battery level, the highest level weight score is assigned; if the fault belongs to the PCS converter level, the next higher level weight score is assigned; if the fault belongs to the EMS scheduling level, the middle level weight score is assigned; and if the fault belongs to the system level, the lower level weight score is assigned.

[0089] In this embodiment, diagnostic information related to the degree of hardware damage is extracted from the first and second diagnostic results, and a corresponding hardware damage weight score is assigned according to the severity of the hardware damage. For example, permanent breakdown / burnout or potential thermal runaway is assigned the highest hardware damage weight score; significant performance degradation without breakdown is assigned a medium hardware damage weight score; and slight performance degradation is assigned a low hardware damage weight score.

[0090] In this embodiment, information on the potential spread risk of the fault is extracted from the target diagnostic results, including but not limited to: whether the fault may trigger a thermal runaway cascade, whether it may cause short-circuit current to spread to other branches, and whether it may cause a multi-unit cascading shutdown. Based on the level of safety spread risk, a corresponding safety risk weight score is assigned. For example, if there is a risk of thermal runaway cascade or short-circuit current spreading to multiple branches, the highest safety risk weight score is assigned; if local components are damaged but there is no condition for spread to other equipment, a medium safety risk weight score is assigned; if equipment performance is slightly degraded and there is no spread risk, a lower safety risk weight score is assigned.

[0091] In this embodiment, information on potential outage losses caused by a fault is extracted from the target diagnostic results. This includes, but is not limited to, the scope of the outage impact, the urgency of the current power grid's energy storage dispatch, and whether the fault occurs during peak electricity price periods / frequency regulation service periods. Based on the magnitude of the outage loss risk, a corresponding outage loss weight score is assigned. For example, a fault that could cause a complete outage of a large-capacity energy storage station is assigned the highest outage loss weight score; a fault that only causes a single unit to shut down is assigned a medium outage loss weight score; and a fault that only causes reduced-rate operation of a local branch without requiring a complete unit shutdown is assigned a lower outage loss weight score.

[0092] The fault level weight, hardware damage weight, safety risk weight, and downtime loss weight are weighted and summed to obtain the comprehensive fault risk score. If the comprehensive fault risk score is in the first interval, the fault represented by the target diagnosis result is determined to be a level 1 fault. If the comprehensive fault risk score is in the second interval, the fault represented by the target diagnosis result is determined to be a level 2 fault. If the comprehensive fault risk score is in the third interval, the fault represented by the target diagnosis result is determined to be a level 3 fault.

[0093] The first interval is also known as the high-score interval, the second interval as the medium-score interval, and the third interval as the low-score interval. The specific scores for each interval can be customized.

[0094] In this embodiment, the faults corresponding to the energy storage system are scored from four dimensions based on the target diagnosis result, the first diagnosis result, and the second diagnosis result, thereby achieving an objective assessment of the fault level. This avoids the subjectivity and arbitrariness of traditional manual experience-based grading, making the triggering of the graded disposal strategy more accurate and reliable.

[0095] In addition, after the fault stratification self-healing is completed or the manual stratification handling is completed, the full data of the four-layer architecture's full-time operation, stratified diagnosis, AI degradation assessment, alarms, and self-healing operations are archived, and the health records of all equipment throughout the entire life cycle are updated. Based on massive historical time-series data, incremental learning and parameter adaptive calibration are carried out on the four-layer architecture's supporting physical mechanism model, AI time-series prediction model, and stratified self-healing strategy library to continuously optimize the accuracy of the four-layer stratified diagnosis and identification, the effect of time-series trend prediction, and the adaptability of the stratified self-healing strategy. After the optimization iteration is completed, the system enters the next round of periodic data collection and stratified diagnosis cycle operation.

[0096] For a better understanding of the overall technical solution, please refer to [link / reference]. Figure 2 .like Figure 2 As shown, real-time data is collected from all dimensions of the battery management system, power conversion system, energy management system, and grid connection system; a four-level hierarchical fault mechanism is applied to accurately diagnose faults.

[0097] Specifically, the battery management system uses AI timing algorithms and a first mechanism model to obtain a first diagnostic result; the power conversion system uses AI timing algorithms and a second mechanism model to obtain a second diagnostic result; the energy management system uses AI timing algorithms and a third mechanism model to obtain a third diagnostic result; and the grid-connected system uses AI timing algorithms and a fourth mechanism model to obtain a fourth diagnostic result.

[0098] Based on the first, second, third, and fourth diagnostic results and real-time operation data from the grid side, the target diagnostic results corresponding to the energy storage system are generated.

[0099] Furthermore, it is determined whether the fault corresponding to the energy storage system is a self-healing fault; if it is a self-healing fault, a self-healing strategy is executed on the energy storage system. Specifically, if the fault belongs to the battery management system, a first self-healing strategy is executed. The first self-healing strategy includes adjusting the battery active balancing start-stop strategy, adjusting the single cell overvoltage protection threshold, adjusting the single cell undervoltage protection threshold, and balancing the charging compensation voltage difference.

[0100] If the fault pertains to a power conversion system, a second self-healing strategy is executed. This second self-healing strategy includes calibrating the voltage sampling zero point, calibrating the current sampling zero point, correcting the SPWM modulation control parameters, restarting the thermal control module, and adjusting the fan speed range.

[0101] If the fault belongs to the energy management system, the third self-healing strategy is executed. The third self-healing strategy includes resetting the upper and lower limits of the scheduling power, resetting the charging and discharging timing control parameters, switching the scheduling control strategy, resetting the abnormal command issuance logic, and restarting the communication link.

[0102] If the fault belongs to a grid-connected system, the fourth self-healing strategy is executed. The fourth self-healing strategy includes reorganizing and redistributing the load of multiple units, correcting the grid-connected voltage parameters, correcting the grid-connected power parameters, and adjusting the reactive power compensation logic of the grid connection point.

[0103] If the fault is not self-healing, then determine the fault level corresponding to the fault represented by the target diagnostic result, and then execute the corresponding fault handling method according to the fault level.

[0104] For a Level 1 fault, the energy storage system will be shut down and disconnected from the grid, and an emergency repair notification will be sent to the outside world.

[0105] For Level 2 faults, the energy storage system will be derated, its capacity and power will be reduced, and a maintenance notification will be sent to the outside world.

[0106] For Level 3 faults, the energy storage system will be tracked and monitored.

[0107] Further reading Figure 3 As a response to the above Figure 1 The implementation of the method shown in this application provides an embodiment of an energy storage system 300, which is similar to... Figure 1 The method embodiments shown correspond to those described.

[0108] An energy storage system 300 provided in this embodiment of the invention includes a processing module 301, a battery management system 302, a power conversion system 303, an energy management system 304, and a grid-connected system 305. The processing module 301 is communicatively connected to the battery management system 302, the power conversion system 303, the energy management system, and the grid-connected system 305. The processing module 301 is used to acquire the first measurement point data corresponding to the battery management system 302, the second measurement point data corresponding to the power conversion system 303, the third measurement point data corresponding to the energy management system, and the fourth measurement point data corresponding to the grid connection system 305. By calling preset AI timing algorithms and multiple mechanism models, data processing is performed on the first measurement point data, the second measurement point data, the third measurement point data, and the fourth measurement point data, respectively, to obtain the first diagnostic result corresponding to the battery management system 302, the second diagnostic result corresponding to the power conversion system 303, the third diagnostic result corresponding to the energy management system, and the fourth diagnostic result corresponding to the grid-connected system 305; Based on the first diagnostic result, the second diagnostic result, the third diagnostic result, the fourth diagnostic result, and the real-time operation data of the power grid, a target diagnostic result corresponding to the energy storage system 300 is generated. The target diagnostic result is used to characterize the system level to which the fault belongs.

[0109] Optionally, the plurality of mechanism models include a first mechanism model corresponding to the battery management system 302, a second mechanism model corresponding to the power conversion system 303, a third mechanism model corresponding to the energy management system, and a fourth mechanism model corresponding to the grid-connected system 305; The processing module 301 is further configured to: The first mechanism model is called to calculate the standard voltage of a single cell, the theoretical voltage difference of equalization, and the theoretical temperature rise of the module. The AI ​​time-series algorithm is called to identify the hardware characteristics of the battery. Based on the standard voltage of the single cell, the theoretical voltage difference of equalization, the theoretical temperature rise of the module, and the hardware characteristics of the battery, a first diagnostic result is generated. The second mechanism model is called to calculate the theoretical loss of the device, the theoretical fluctuation of the bus voltage, and the theoretical temperature difference of heat dissipation. The AI ​​time series algorithm is called to identify the characteristics of the power device. Based on the theoretical loss of the device, the theoretical fluctuation of the bus voltage, the theoretical temperature difference of heat dissipation, and the characteristics of the power device, a second diagnostic result is generated. The third mechanism model is invoked to calculate the difference between the instruction issued and the execution feedback value. The AI ​​time series algorithm is invoked to identify communication characteristics. Based on the difference and the communication characteristics, a third diagnostic result is generated. The fourth mechanism model is invoked to calculate the power difference between the theoretical power and the actual power. The AI ​​time series algorithm is invoked to identify grid connection characteristics. Based on the power difference and the grid connection characteristics, a fourth diagnostic result is generated.

[0110] Optionally, the processing module 301 is further specifically used for: Based on the fault type represented by the target diagnostic result, the first diagnostic result, the second diagnostic result, the system level and duration of the fault represented by the target diagnostic result, it is determined whether the fault corresponding to the energy storage system 300 is a self-healing fault. If the fault corresponding to the energy storage system 300 is a self-healing fault, then a self-healing strategy is executed on the energy storage system 300. If the fault corresponding to the energy storage system 300 is not a self-healing fault, then the energy storage system 300 will be subjected to graded fault handling.

[0111] Optionally, the processing module 301 is further specifically used for: Under the condition that the preset conditions are met, the fault corresponding to the energy storage system 300 is determined to be a self-healing fault; If the preset conditions are not met, it is determined that the fault corresponding to the energy storage system 300 is not a self-healing fault; The preset conditions include: The fault type characterized by the target diagnostic results is not a hardware fault, and; The degradation score determined based on the first diagnostic result and the second diagnostic result is less than or equal to the preset score, and; The fault represented by the target diagnostic result belongs to a single system level, and the duration of the fault represented by the target diagnostic result is less than a preset time.

[0112] Optionally, the processing module 301 is further specifically used for: When the fault represented by the target diagnostic result belongs to the battery management system 302, adjust the battery active balancing start-stop strategy, adjust the single cell overvoltage protection threshold, adjust the single cell undervoltage protection threshold, and balance the charging compensation voltage difference. When the fault described by the target diagnostic result belongs to the power conversion system 303, the following steps are taken: calibrate the voltage sampling zero point, calibrate the current sampling zero point, correct the SPWM modulation control parameters, restart the heat dissipation control module, and adjust the fan speed range. If the fault represented by the target diagnostic result belongs to the energy management system, reset the upper and lower limits of the scheduling power, reset the charging and discharging timing control parameters, switch the scheduling control strategy, reset the abnormal command issuance logic, and restart the communication link. When the fault level represented by the target diagnostic result is grid-connected system 305, the load of multiple units is reorganized and distributed, the grid-connected voltage parameters are corrected, the grid-connected power parameters are corrected, and the reactive power compensation logic of the grid connection point is adjusted.

[0113] Optionally, the processing module 301 is further specifically used for: Based on the target diagnostic result, the first diagnostic result, and the second diagnostic result, determine the fault level corresponding to the fault represented by the target diagnostic result; If the fault characterized by the target diagnosis result is a level one fault, then the energy storage system 300 will be shut down and disconnected from the power grid, and an emergency repair prompt will be sent to the outside world. If the fault characterized by the target diagnostic result is a level two fault, then the energy storage system 300 will be subjected to derating, capacity limiting and power reduction, and a maintenance reminder will be sent to the outside world. If the fault characterized by the target diagnostic result is a level three fault, then the energy storage system 300 will be tracked and monitored.

[0114] Optionally, the processing module 301 is further specifically used for: The fault level weight is determined based on the system level to which the fault belongs, as characterized by the target diagnostic results. The hardware damage weight is determined based on the degree of hardware damage characterized by the first diagnostic result and the second diagnostic result. Based on the fault propagation risk characterized by the target diagnostic results, determine the safety risk weights; Based on the downtime loss risk characterized by the target diagnostic results, the downtime loss weight is determined; The fault level weight, hardware damage weight, safety risk weight, and downtime loss weight are weighted and summed to obtain the comprehensive fault risk score. If the comprehensive fault risk score is in the first interval, then the fault represented by the target diagnosis result is determined to be a level one fault. If the comprehensive fault risk score is in the second interval, then the fault represented by the target diagnostic result is determined to be a level two fault. If the comprehensive fault risk score is in the third interval, then the fault represented by the target diagnostic result is determined to be a level three fault. The lowest score in the first interval is higher than the highest score in the second interval, and the lowest score in the second interval is higher than the highest score in the third interval.

[0115] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0116] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0117] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0118] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating platform and various application software installed on the computer device 4, such as the program code of the fault diagnosis method for the energy storage system. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0119] In some embodiments, the processor 42 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run program code stored in the memory 41 or process data, for example, to run program code for a fault diagnosis method of the energy storage system.

[0120] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0121] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described energy storage system fault diagnosis method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0122] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware online platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0123] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0124] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A fault diagnosis method for an energy storage system, characterized in that, The method is applied to an energy storage system, which includes a battery management system, a power conversion system, an energy management system, and a grid connection system. The method includes: Acquire the first measurement point data corresponding to the battery management system, the second measurement point data corresponding to the power conversion system, the third measurement point data corresponding to the energy management system, and the fourth measurement point data corresponding to the grid connection system; By calling preset AI timing algorithms and multiple mechanism models, data processing is performed on the first measurement point data, the second measurement point data, the third measurement point data, and the fourth measurement point data, respectively, to obtain the first diagnostic result corresponding to the battery management system, the second diagnostic result corresponding to the power conversion system, the third diagnostic result corresponding to the energy management system, and the fourth diagnostic result corresponding to the grid connection system. Based on the first diagnostic result, the second diagnostic result, the third diagnostic result, the fourth diagnostic result, and the real-time operation data of the power grid, a target diagnostic result corresponding to the energy storage system is generated. The target diagnostic result is used to characterize the system level to which the fault belongs.

2. The method according to claim 1, characterized in that, The plurality of mechanism models include a first mechanism model corresponding to the battery management system, a second mechanism model corresponding to the power conversion system, a third mechanism model corresponding to the energy management system, and a fourth mechanism model corresponding to the grid connection system; The process involves calling a preset AI timing algorithm and multiple mechanism models to process the data at the first, second, third, and fourth measurement points, respectively, to obtain a first diagnostic result corresponding to the battery management system, a second diagnostic result corresponding to the power conversion system, a third diagnostic result corresponding to the energy management system, and a fourth diagnostic result corresponding to the grid-connected system, including: The first mechanism model is called to calculate the standard voltage of a single cell, the theoretical voltage difference of equalization, and the theoretical temperature rise of the module. The AI ​​time-series algorithm is called to identify the hardware characteristics of the battery. Based on the standard voltage of the single cell, the theoretical voltage difference of equalization, the theoretical temperature rise of the module, and the hardware characteristics of the battery, a first diagnostic result is generated. The second mechanism model is called to calculate the theoretical loss of the device, the theoretical fluctuation of the bus voltage, and the theoretical temperature difference of heat dissipation. The AI ​​time series algorithm is called to identify the characteristics of the power device. Based on the theoretical loss of the device, the theoretical fluctuation of the bus voltage, the theoretical temperature difference of heat dissipation, and the characteristics of the power device, a second diagnostic result is generated. The third mechanism model is invoked to calculate the difference between the instruction issued and the execution feedback value. The AI ​​time series algorithm is invoked to identify communication characteristics. Based on the difference and the communication characteristics, a third diagnostic result is generated. The fourth mechanism model is invoked to calculate the power difference between the theoretical power and the actual power. The AI ​​time series algorithm is invoked to identify grid connection characteristics. Based on the power difference and the grid connection characteristics, a fourth diagnostic result is generated.

3. The method according to claim 1, characterized in that, After generating the target diagnostic results corresponding to the energy storage system, the method further includes: Based on the fault type represented by the target diagnostic result, the first diagnostic result, the second diagnostic result, the system level and duration of the fault represented by the target diagnostic result, it is determined whether the fault corresponding to the energy storage system is a self-healing fault. If the fault corresponding to the energy storage system is a self-healing fault, then a self-healing strategy is executed on the energy storage system. If the fault corresponding to the energy storage system is not a self-healing fault, then graded fault handling is performed on the energy storage system.

4. The method according to claim 3, characterized in that, The step of determining whether the fault corresponding to the energy storage system is a self-healing fault based on the fault type characterized by the target diagnostic result, the first diagnostic result, the second diagnostic result, and the system level and duration of the fault characterized by the target diagnostic result includes: Under the condition that the preset conditions are met, the fault corresponding to the energy storage system is determined to be a self-healing fault; If the preset conditions are not met, it is determined that the fault corresponding to the energy storage system is not a self-healing fault; The preset conditions include: The fault type characterized by the target diagnostic results is not a hardware fault, and; The degradation score determined based on the first diagnostic result and the second diagnostic result is less than or equal to the preset score, and; The fault represented by the target diagnostic result belongs to a single system level, and the duration of the fault represented by the target diagnostic result is less than a preset time.

5. The method according to claim 3, characterized in that, The self-healing strategy for the energy storage system includes: When the fault represented by the target diagnostic result belongs to the battery management system, adjust the battery active balancing start-stop strategy, adjust the single cell overvoltage protection threshold, adjust the single cell undervoltage protection threshold, and balance the charging compensation voltage difference. If the fault described by the target diagnostic result belongs to the power conversion system, then calibrate the voltage sampling zero point, calibrate the current sampling zero point, correct the SPWM modulation control parameters, restart the heat dissipation control module, and adjust the fan speed range. If the fault represented by the target diagnostic result belongs to the energy management system, reset the upper and lower limits of the scheduling power, reset the charging and discharging timing control parameters, switch the scheduling control strategy, reset the abnormal command issuance logic, and restart the communication link. When the fault described in the target diagnostic result belongs to the grid-connected system, the load of multiple units is reorganized and allocated, the grid-connected voltage parameters are corrected, the grid-connected power parameters are corrected, and the reactive power compensation logic of the grid connection point is adjusted.

6. The method according to claim 3, characterized in that, The implementation of graded fault handling for the energy storage system includes: Based on the target diagnostic result, the first diagnostic result, and the second diagnostic result, determine the fault level corresponding to the fault represented by the target diagnostic result; If the fault characterized by the target diagnosis result is a level one fault, then the energy storage system will be shut down and disconnected from the grid, and an emergency repair prompt will be sent to the outside world. If the fault characterized by the target diagnostic result is a level two fault, then the energy storage system will be subjected to derating, capacity limiting and power reduction, and a maintenance reminder will be sent to the outside world; If the fault characterized by the target diagnostic result is a level three fault, then the energy storage system will be tracked and monitored.

7. The method according to claim 6, characterized in that, The step of determining the fault level corresponding to the fault represented by the target diagnostic result based on the target diagnostic result, the first diagnostic result, and the second diagnostic result includes: The fault level weight is determined based on the system level to which the fault belongs, as characterized by the target diagnostic results. The hardware damage weight is determined based on the degree of hardware damage characterized by the first diagnostic result and the second diagnostic result. Based on the fault propagation risk characterized by the target diagnostic results, determine the safety risk weights; Based on the downtime loss risk characterized by the target diagnostic results, the downtime loss weight is determined; The fault level weight, hardware damage weight, safety risk weight, and downtime loss weight are weighted and summed to obtain the comprehensive fault risk score. If the comprehensive fault risk score is in the first interval, then the fault represented by the target diagnosis result is determined to be a level one fault. If the comprehensive fault risk score is in the second interval, then the fault represented by the target diagnostic result is determined to be a level two fault. If the comprehensive fault risk score is in the third interval, then the fault represented by the target diagnostic result is determined to be a level three fault. The lowest score in the first interval is higher than the highest score in the second interval, and the lowest score in the second interval is higher than the highest score in the third interval.

8. An energy storage system, characterized in that, The energy storage system includes a processing module, a battery management system, a power conversion system, an energy management system, and a grid connection system. The processing module is communicatively connected to the battery management system, the power conversion system, the energy management system, and the grid connection system, respectively. The processing module is used to acquire the first measurement point data corresponding to the battery management system, the second measurement point data corresponding to the power conversion system, the third measurement point data corresponding to the energy management system, and the fourth measurement point data corresponding to the grid connection system. By calling preset AI timing algorithms and multiple mechanism models, data processing is performed on the first measurement point data, the second measurement point data, the third measurement point data, and the fourth measurement point data, respectively, to obtain the first diagnostic result corresponding to the battery management system, the second diagnostic result corresponding to the power conversion system, the third diagnostic result corresponding to the energy management system, and the fourth diagnostic result corresponding to the grid connection system. Based on the first diagnostic result, the second diagnostic result, the third diagnostic result, the fourth diagnostic result, and the real-time operation data of the power grid, a target diagnostic result corresponding to the energy storage system is generated. The target diagnostic result is used to characterize the system level to which the fault belongs.

9. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the fault diagnosis method for the energy storage system as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the fault diagnosis method for the energy storage system as described in any one of claims 1 to 7.

Images