Three-phase photovoltaic inverter fault management and control system based on data integration

Abstract

The application discloses a three-phase photovoltaic inverter fault management and control system based on data integration, relates to the technical field of inverter fault management and control, and aims to solve the technical problem that in the prior art, iterative operation analysis cannot be performed through operation simulation model construction, so that blind detection is performed during fault identification, specifically, the system integrates data of three units of operation simulation, fault feature extraction and environment influence analysis through a fault management and control platform, realizes full-process data closed loop of fault from positioning-diagnosis-regulation-environment adaptation, continuously accumulates and updates failure mechanism and regulation mechanism of the same type operation end, forms a reusable operation and maintenance knowledge base, and provides data support for long-term operation and maintenance and preventive maintenance of the photovoltaic inverter.

Description

Technical Field

[0001] This invention relates to the field of inverter fault management technology, specifically a three-phase photovoltaic inverter fault management system based on data integration. Background Technology

[0002] Three-phase photovoltaic inverters are core power electronic devices that efficiently convert direct current (DC) generated by photovoltaic arrays into three-phase alternating current (AC). They are mainly used in medium-to-high power photovoltaic grid-connected power generation systems (typically with a power of ≥10kW). They can be directly connected to a three-phase power grid or drive three-phase loads (such as industrial motors). They are a key hub connecting photovoltaic energy with the power consumption / grid side.

[0003] For example, patent application number 2025105830990 discloses an inverter drive control method, system, and inverter for a synchronous motor. The synchronous motor is driven by a current source inverter. The method includes: constructing a mathematical model of the synchronous motor in the dq coordinate system to obtain the actual current and voltage of the synchronous motor on the d-axis and q-axis; defining state variables; constructing a Lyapunov function to ensure that the Lyapunov function is globally asymptotically stable at the equilibrium point to obtain voltage reference values ​​of the synchronous motor on the d-axis and q-axis; calculating the ideal setpoint value of the output current on the AC side of the current source inverter based on the voltage reference value; and converting the ideal setpoint value of the output current to obtain the drive signal of the inverter bridge switching transistor in the inverter.

[0004] However, in the existing technology, it is still impossible to perform iterative calculation and analysis through the construction of an operation simulation model during the inverter operation phase. As a result, blind detection is performed during fault identification, making fault location without reference basis and unable to control the location time. It is also impossible to extract failure mechanism and adjustment mechanism based on fault feature extraction, lacking a basis for fault management and reducing the efficiency of fault management.

[0005] Therefore, a solution is proposed. Summary of the Invention

[0006] The purpose of this invention is to solve the problems mentioned above by proposing a fault management system for three-phase photovoltaic inverters based on data integration.

[0007] The objective of this invention can be achieved through the following technical solution: a three-phase photovoltaic inverter fault management system based on data integration, including a fault management platform, wherein the fault management platform is communicatively connected to an operation simulation model construction unit, a fault feature extraction unit, and an environmental impact analysis unit;

[0008] The operation simulation model construction unit establishes a simulation model based on the operation process of the three-phase photovoltaic inverter, and performs operation detection on the simulation model by inputting operation parameters, and obtains various types of time moments and waveforms based on the detection process;

[0009] The fault feature extraction unit combines various time points and waveforms to extract features, compares the features of the photovoltaic inverter operating parameters, and obtains the failure mechanism and regulation mechanism based on the feature comparison.

[0010] The environmental impact analysis unit conducts environmental impact analysis based on the failure mechanism, and performs synchronous monitoring of environmental parameters for both the failure mechanism and the regulation mechanism to ensure the efficiency of monitoring and regulation.

[0011] Furthermore, the process of operating the simulation model building unit is as follows:

[0012] The simulation model is constructed by collecting data on the internal components and connection circuits of the photovoltaic inverter, and the internal components involved in the operation in the simulation model are marked as the operating end.

[0013] The task electrical parameters are input, and the operating terminals are sorted according to the operation process of the photovoltaic inverter. The task electrical parameters are represented as current and voltage. The first operating terminal in the sorted order carries the task electrical parameters, and after receiving the input task electrical parameters, it operates the component and produces electrical parameters after completing the component processing. These parameters are marked as processed electrical parameters. The process is iteratively processed according to all sorted operating terminals, and finally, electrical parameters are produced and marked as result electrical parameters.

[0014] Based on the task electrical parameters and the equipment characteristics of the photovoltaic inverter, the preset output electrical parameters are obtained. If the deviation between the preset output electrical parameters and the result electrical parameters is within the set result parameter deviation range, the current operation process of the photovoltaic inverter is marked as a qualified process; if the deviation between the preset output electrical parameters and the result electrical parameters is not within the set result parameter deviation range, the current operation process of the photovoltaic inverter is marked as a non-qualified process.

[0015] Furthermore, during the non-qualified process, the processing electrical parameters at each operating moment are statistically analyzed and recorded. The operating moment is classified according to whether the processing electrical parameters are consistent with the output electrical parameters of the operating end components. If they are consistent, they are classified as slow-state operating ends; if they are inconsistent, they are classified as fast-state operating ends.

[0016] Furthermore, during the non-qualified process, if a fast-state operating end appears, the corresponding operating end is marked as the fault source point, and the fluctuation trend of the input electrical parameters and the fluctuation trend of the output electrical parameters of the fault source point are recorded.

[0017] If the fluctuation trend of the input electrical parameters at the fault source point does not exceed the set input parameter threshold, but the fluctuation trend of the output electrical parameters exceeds the set output parameter threshold, then the corresponding fluctuation time will be marked as the component abnormal time, and the output electrical parameters of the corresponding operating end will be summarized to form the component abnormal waveform.

[0018] If the fluctuation trend of the input electrical parameters at the fault source point exceeds the fluctuation value of the set input parameter threshold, and the fluctuation trend of the output electrical parameters exceeds the value fluctuation of the set output parameter threshold, then the corresponding value fluctuation time is marked as the load abnormal time, and the output electrical parameters of the corresponding operating end are summarized to form a load abnormal waveform.

[0019] If the fluctuation trend of the input electrical parameters at the fault source point exceeds the fluctuation value of the set input parameter threshold, and the fluctuation trend of the output electrical parameters does not exceed the fluctuation value of the set output parameter threshold, then the corresponding fluctuation time will be marked as the operation adjustment time, and the output electrical parameters of the corresponding operation end will be summarized to form the operation adjustment waveform.

[0020] If the fluctuation trend of the input electrical parameters at the fault source does not exceed the set input parameter threshold, and the fluctuation trend of the output electrical parameters does not exceed the set output parameter threshold, then the corresponding fluctuation time is marked as the stable operation time, and the output electrical parameters at the corresponding operating end are summarized to form a stable operation waveform. The stable operation time and the stable operation waveform are then sent to the fault control platform together.

[0021] Furthermore, the process of the fault feature extraction unit is as follows:

[0022] The load anomaly waveform is extracted. Based on the load anomaly time of the non-load anomaly waveform and a waveform statistics period is set, the non-load anomaly waveform and the load anomaly waveform are obtained with the load anomaly time as the dividing point. The numerical waveforms of the input electrical parameters and output electrical parameters during the non-load anomaly waveform stage are recorded and marked as the input waveform before the anomaly and the output waveform before the anomaly, respectively. The numerical waveforms of the input electrical parameters and output electrical parameters are obtained based on the load anomaly waveform and marked as the input waveform after the anomaly and the output waveform after the anomaly, respectively.

[0023] The input waveforms before and after the anomaly are compared and the non-overlapping waveforms are extracted. The failure mechanism is set based on the non-overlapping waveforms. Combined with the historical running period, if the parameter fluctuation span corresponding to the non-overlapping waveform is generated and the operating end is running normally, this type of non-overlapping waveform is marked as the accumulation effect waveform; if the parameter fluctuation span corresponding to the non-overlapping waveform is generated and the operating end is running abnormally, this type of non-overlapping waveform is marked as the failure effect waveform.

[0024] Furthermore, based on the time of load anomaly, the cumulative duration of the accumulated effect waveform is obtained. If the cumulative duration of the accumulated effect waveform is inconsistent and the deviation fluctuates repeatedly when the same type of operating end is repeatedly used as the fault source, then the accumulated effect waveform is marked as the non-failure mechanism of the current fault source. If the cumulative duration of the accumulated effect waveform is consistent or the deviation fluctuates repeatedly within a set value range when the same type of operating end is repeatedly used as the fault source, then the accumulated effect waveform and the failure effect waveform are marked as the failure mechanism of the current fault source.

[0025] Furthermore, the stable operating waveform is extracted, and adjacent load anomaly moments and operation adjustment moments are collected. The interval is marked as a stable period. The execution time of operation adjustment measures within the stable period is recorded. After the operation adjustment measures are executed, the load anomaly waveform and the operation adjustment waveform are compared and overlapped. The trend changes of non-overlapping waveforms are recorded. The current operation adjustment measure execution progress is correlated with the trend changes of non-overlapping waveforms and uniformly marked as the adjustment mechanism. The adjustment mechanism is updated in conjunction with the adjustment process of multiple fault sources, and the determined adjustment mechanism is sent to the fault control platform. The fault control platform monitors the adjustment of all fault sources at the current moment and compares the waveform trend changes according to the progress of adjustment measures at different stages. If the actual waveform trend change is inconsistent with the corresponding waveform trend change, it indicates an adjustment anomaly, and the adjustment measures are changed. Conversely, if the actual waveform trend change is consistent with the corresponding waveform trend change, it indicates normal adjustment, and the current adjustment measures continue to be executed.

[0026] Furthermore, the process for the environmental impact analysis unit is as follows:

[0027] Based on the failure mechanism, the waveform transition period is obtained and marked as the gradual failure stage; based on the regulation mechanism, the waveform transition period is obtained and marked as the regulation effectiveness stage; the operating environment parameters are collected at the operating end for the gradual failure stage and the regulation effectiveness stage, where the environmental parameters are represented by humidity and temperature in the operating environment;

[0028] During the gradual failure phase, environmental parameter types that fluctuate beyond the set range are collected and set as widespread interference types. During the phase where the fluctuation of environmental parameters of the widespread interference type intensifies, if the waveform fluctuation slope increases during the gradual failure phase, the corresponding type of environmental parameter is extracted from the widespread interference type and marked as precise interference type.

[0029] During the adjustment effect phase, if the real-time parameter waveform evolves according to the operation adjustment waveform, then the precise disturbance type environmental parameter or the wide disturbance type environmental parameter that floats within the set range is marked as the adjustment control type; otherwise, it is marked as the adjustment non-control type.

[0030] Compared with the prior art, the beneficial effects of the present invention are:

[0031] 1. By constructing a simulation model unit to model the internal components and connection circuits of the photovoltaic inverter, and combining iterative calculation and deviation analysis of the task electrical parameters, the fault source point in the non-qualified process can be quickly located. Based on the fluctuation trend of input / output electrical parameters, different fault types such as component abnormality and load abnormality can be accurately classified, avoiding blind testing of the entire system in traditional troubleshooting and greatly shortening the fault location time.

[0032] 2. The fault feature extraction unit clarifies the failure mechanism of the fault source by comparing the overlap of waveforms before and after the load abnormality, classifying non-overlapping waveforms, and marking the failure mechanism (such as the correlation determination between the waveform affected by accumulation and the waveform affected by failure). At the same time, it continuously monitors historical data from similar operating terminals, breaking through the traditional fault diagnosis that only stays at the level of "finding the fault" and realizing in-depth diagnosis of "clarifying the root cause of the fault", providing accurate basis for subsequent maintenance.

[0033] 3. By extracting stable operating waveforms and analyzing the correspondence between adjustment measures and waveform trends during stable periods, the adjustment mechanism is constructed and dynamically updated. The fault control platform monitors the adjustment process in real time. If the actual waveform is inconsistent with the expected trend, the measures are adjusted in a timely manner to avoid the expansion of the fault due to adjustment lag or inappropriate measures. This ensures that the inverter can quickly restore stable operation after fault handling and reduces downtime losses.

[0034] 4. The environmental impact analysis unit accurately identifies the "widespread interference type" and "precise interference type" environmental parameters that affect the failure by dividing the gradual failure stage and the adjustment effect stage, and marks the "control type" of environmental parameters during the adjustment process, so that the fault management extends from the "equipment itself" to the "equipment-environment" collaborative level. Based on this, the fault management platform monitors environmental parameters to avoid environmental factors from aggravating the failure or offsetting the adjustment effect, and improves the overall anti-interference capability of the system. Attached Figure Description

[0035] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0036] Figure 1 This is a system principle block diagram of the present invention;

[0037] Figure 2 This is a flowchart illustrating the system coordination process of the present invention. Detailed Implementation

[0038] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Please see Figures 1-2 As shown, the fault management system for a three-phase photovoltaic inverter based on data integration includes a fault management platform. The fault management platform is connected to the operation simulation model construction unit, the fault feature extraction unit, and the environmental impact analysis unit via an industrial Ethernet (or RS485 bus) to receive electrical parameter data (current, voltage), waveform data, and environmental parameter data (temperature, humidity) sent by each unit.

[0040] The fault management platform generates a signal for constructing an operation simulation model and sends it to the operation simulation model construction unit.

[0041] After receiving the operation simulation model construction signal, the operation simulation model construction unit establishes a simulation model based on the operation process of the three-phase photovoltaic inverter, and performs operation detection on the input operation parameters of the simulation model.

[0042] The simulation model is constructed by collecting data on the internal components and connection circuits of the photovoltaic inverter, and the internal components involved in the operation in the simulation model are marked as the operating end.

[0043] Input the task electrical parameters and sort the operating terminals according to the operation process of the photovoltaic inverter. The task electrical parameters are represented as current and voltage.

[0044] The first operating end in the sequence carries the task electrical parameters, and after receiving the input task electrical parameters, it operates the components and produces electrical parameters after completing the component processing, and marks them as processed electrical parameters. Iterative processing is performed according to all the sorted operating ends, and finally electrical parameters are produced and marked as result electrical parameters.

[0045] Based on the task electrical parameters and the equipment characteristics of the photovoltaic inverter, the preset output electrical parameters are obtained. If the deviation between the preset output electrical parameters and the result electrical parameters is within the set result parameter deviation range, the current operation process of the photovoltaic inverter is marked as a qualified process; if the deviation between the preset output electrical parameters and the result electrical parameters is not within the set result parameter deviation range, the current operation process of the photovoltaic inverter is marked as a non-qualified process.

[0046] Specific use case: During operation testing, specific task electrical parameters are input; assuming the current is set to 10A and the voltage to 500V, these parameters are input into the simulation model; according to the operation process of the photovoltaic inverter, each operating terminal is sorted; the first operating terminal in the sorted order first carries the task electrical parameters and begins component operation; during this process, the operating terminal processes the input current and voltage, and after completing the component processing, it produces new electrical parameters, which are marked as processed electrical parameters; then, the subsequent operating terminals are iteratively processed in sequence, with each operating terminal using the processed electrical parameters of the previous operating terminal as input to perform its own component operation and parameter processing, and finally producing the result electrical parameters.

[0047] Based on the task's electrical parameters and the characteristics of the photovoltaic inverter, preset output electrical parameters are obtained through theoretical calculations or empirical formulas. For example, after calculation, the preset output electrical parameters are a current of 9.8-10.2A and a voltage of 495-505V. The preset output electrical parameters are compared with the actual obtained electrical parameters. If the actual electrical parameters are within the set deviation range (i.e., current between 9.8-10.2A and voltage between 495-505V), the current operation of the photovoltaic inverter is marked as a qualified process. If the actual electrical parameters exceed this range (e.g., current of 9.5A and voltage of 490V), and are not within the preset range, the current operation is marked as a non-qualified process.

[0048] During the non-conforming process, the processing electrical parameters at each operating moment are statistically analyzed and recorded. The operating moment is classified according to whether the processing electrical parameters are consistent with the output electrical parameters of the operating end components. If they are consistent, they are classified as slow-state operating end; if they are inconsistent, they are classified as fast-state operating end.

[0049] During non-conforming processes, if a fast-state operation occurs, the corresponding operation point is marked as a fault source. The fluctuation trends of the input electrical parameters and the output electrical parameters of the fault source are recorded. If the fluctuation trend of the input electrical parameters of the fault source does not exceed the set input parameter threshold, but the fluctuation trend of the output electrical parameters exceeds the set output parameter threshold, the corresponding fluctuation time is marked as the component abnormal time. The output electrical parameters of the corresponding operation point are summarized to form the component abnormal waveform. The component abnormal time and the component abnormal waveform are sent to the fault control platform together.

[0050] If the fluctuation trend of the input electrical parameters at the fault source point exceeds the fluctuation value of the set input parameter threshold, and the fluctuation trend of the output electrical parameters exceeds the value fluctuation of the set output parameter threshold, then the corresponding value fluctuation time is marked as the load abnormal time, and the output electrical parameters of the corresponding operating end are summarized to form the load abnormal waveform. The load abnormal time and the load abnormal waveform are sent to the fault management platform together.

[0051] If the fluctuation trend of the input electrical parameters at the fault source exceeds the fluctuation value of the set input parameter threshold, and the fluctuation trend of the output electrical parameters does not exceed the fluctuation value of the set output parameter threshold, then the corresponding fluctuation time will be marked as the operation adjustment time, and the output electrical parameters of the corresponding operation end will be summarized to form the operation adjustment waveform. The operation adjustment time and the operation adjustment waveform will be sent to the fault management platform together.

[0052] If the fluctuation trend of the input electrical parameters at the fault source point does not exceed the set input parameter threshold, and the fluctuation trend of the output electrical parameters does not exceed the set output parameter threshold, then the corresponding fluctuation time will be marked as the stable operation time, and the output electrical parameters at the corresponding operating end will be summarized to form a stable operation waveform. The stable operation time and the stable operation waveform will be sent to the fault control platform together.

[0053] Specific application scenario: During non-compliance processes, detailed statistics are collected on the processing electrical parameters at each operating moment, and the processing electrical parameters at all operating moments are recorded.

[0054] The type of operating terminal is classified based on whether the processed electrical parameters are consistent with the output electrical parameters of the operating terminal components. For example, if the output electrical parameters of a certain operating terminal component are 10A current and 500V voltage, while the actual processed electrical parameters are 9.5A current and 490V voltage, the two are inconsistent, and this operating terminal is classified as a fast-state operating terminal. If the processed electrical parameters are consistent with the output electrical parameters of the component components, it is classified as a slow-state operating terminal.

[0055] When a fast-state operation occurs, the corresponding operation point is marked as the fault source point; the fluctuation trend of the input electrical parameters and the fluctuation trend of the output electrical parameters of the fault source point are recorded; assuming that the fluctuation trend of the input electrical parameters of the fault source point is relatively stable and does not exceed the set input parameter threshold (such as the current threshold of 9-11A and the voltage threshold of 490-510V), while the fluctuation trend of the output electrical parameters exceeds the set output parameter threshold (such as the current threshold of 9.5-10.5A and the voltage threshold of 495-505V), for example, the output current is 9.3A, which exceeds the lower limit of the threshold, then the corresponding value fluctuation time is marked as the component abnormal time;

[0056] After receiving various types of time and waveforms, the fault management platform generates a fault feature extraction signal and sends it to the fault feature extraction unit.

[0057] After receiving the fault feature extraction signal, the fault feature extraction unit performs feature extraction by combining various time points and waveforms to facilitate feature comparison of photovoltaic inverter operating parameters and accurately identify the operational risks of the photovoltaic inverter.

[0058] The load anomaly waveform is extracted. Based on the load anomaly time of the non-load anomaly waveform and a waveform statistics period is set, the non-load anomaly waveform and the load anomaly waveform are obtained with the load anomaly time as the dividing point. The numerical waveforms of the input electrical parameters and output electrical parameters during the non-load anomaly waveform stage are recorded and marked as the input waveform before the anomaly and the output waveform before the anomaly, respectively. The numerical waveforms of the input electrical parameters and output electrical parameters are obtained based on the load anomaly waveform and marked as the input waveform after the anomaly and the output waveform after the anomaly, respectively.

[0059] The input waveforms before and after the anomaly are compared and the non-overlapping waveforms are extracted. The failure mechanism is set based on the non-overlapping waveforms. Combined with the historical running period, if the parameter fluctuation span corresponding to the non-overlapping waveform is generated and the operating end is running normally, then this type of non-overlapping waveform is marked as the accumulation effect waveform; if the parameter fluctuation span corresponding to the non-overlapping waveform is generated and the operating end is running abnormally, then this type of non-overlapping waveform is marked as the failure effect waveform.

[0060] Specific application scenario: For the analysis of abnormal load waveforms, the waveform statistical period is set to 5 seconds. This is a time length that has been verified in practice and can reflect the changes in the inverter's operating status relatively comprehensively and accurately. The waveform is divided into detailed segments based on the moment of load abnormality. For example, if the moment of load abnormality is the 3rd second in a certain monitoring, the waveform of the first 3 seconds in this 5-second statistical period is the non-load abnormal waveform, and the waveform of the last 2 seconds is the load abnormal waveform.

[0061] When recording the numerical waveforms of the corresponding input and output electrical parameters, high-precision data acquisition equipment is used to sample at a frequency of 1000 data points per second. For example, for the current in the input electrical parameters, in the non-load abnormal waveform stage (first 3 seconds), the acquired data shows a relatively stable sine wave shape, with a peak value of about 10A and a valley value of about -10A. However, in the load abnormal waveform stage (last 3-5 seconds), the current waveform shows obvious distortion, with the peak value reaching 15A and the valley value dropping to -15A. The frequency of the waveform also changes from the original 50Hz to 60Hz.

[0062] Regarding the voltage in the output electrical parameters, during the non-load abnormal waveform phase, its value remained stable at around 220V, with a fluctuation range of ±5V; during the load abnormal waveform phase, the voltage plummeted to 200V, and the fluctuations were severe, with the fluctuation range expanding to ±20V. These detailed data records provided a solid foundation for subsequent fault analysis.

[0063] Based on the time of load anomaly, the cumulative duration of the accumulated impact waveform is obtained. If the cumulative duration of the accumulated impact waveform is inconsistent and the deviation fluctuates repeatedly when the same type of operating end is repeatedly used as the fault source, then the accumulated impact waveform is marked as the non-failure mechanism of the current fault source. If the cumulative duration of the accumulated impact waveform is consistent or the deviation fluctuates repeatedly within a set value range when the same type of operating end is repeatedly used as the fault source, then the accumulated impact waveform and the failure impact waveform are marked as the failure mechanism of the current fault source.

[0064] The failure mechanisms collected in real time are sent to the fault management platform, which then continuously monitors the failure mechanisms of similar operating devices.

[0065] The stable operating waveform is extracted, and adjacent load anomaly moments and operation adjustment moments are collected. The interval is marked as a stable period. The execution time of operation adjustment measures within the stable period is recorded. After the operation adjustment measures are executed, the load anomaly waveform and the operation adjustment waveform are compared and overlapped. The trend changes of non-overlapping waveforms are recorded. The current operation adjustment measure execution progress is correlated with the trend changes of non-overlapping waveforms and uniformly marked as the adjustment mechanism. The adjustment mechanism is updated based on the adjustment process of multiple fault sources and the determined adjustment mechanism is sent to the fault control platform. The fault control platform monitors the adjustment of all fault sources at the current moment and compares the waveform trend changes according to the progress of adjustment measures at different stages. If the actual waveform trend change is inconsistent with the corresponding waveform trend change, it indicates an adjustment anomaly, and the adjustment measures are changed. Conversely, if the actual waveform trend change is consistent with the corresponding waveform trend change, it indicates normal adjustment, and the current adjustment measures continue to be executed.

[0066] Specific application scenario: During a certain historical operating period, when the parameter fluctuation span corresponding to the non-overlapping waveform occurs, the operating status of the operating end is closely observed. If the operating end is operating normally, for example, all its electrical performance indicators are within the normal range, and the parameters such as component temperature and power consumption are stable, then this type of non-overlapping waveform is marked as an accumulation effect waveform. Suppose that in a monitoring, the current parameter fluctuation span of the non-overlapping waveform is ±2A, while the actual operating current of the operating end always remains within ±5% of the normal operating current. The operating end is stable and there are no phenomena such as overheating or abnormal vibration. Therefore, it can be determined that the non-overlapping waveform is an accumulation effect waveform.

[0067] If the operating end malfunctions, such as component overheating, increased noise, or decreased output power, and the parameters corresponding to the non-overlapping waveform fluctuate, then this type of non-overlapping waveform is marked as a failure-affecting waveform. For example, when the voltage parameter fluctuation of the non-overlapping waveform reaches ±20V, and the operating end experiences component overheating with the temperature exceeding the normal operating temperature by 20°C and the output power decreases by 10%, then the non-overlapping waveform can be identified as a failure-affecting waveform.

[0068] After determining the cumulative effect waveform, accurately calculate its cumulative duration based on the time of load anomaly. When the same type of operating end is repeatedly used as the fault source, conduct a detailed comparison of the cumulative duration of the cumulative effect waveform in different stages. If the cumulative duration of the cumulative effect waveform in different stages is inconsistent and the deviation fluctuates back and forth, for example, the cumulative duration of the cumulative effect waveform is 2 seconds during the first fault and 5 seconds during the second fault, with a large deviation between the two, and this deviation fluctuates back and forth irregularly, then the cumulative effect waveform is marked as a non-failure mechanism of the current fault source. This means that the appearance of this waveform is not the direct cause of the fault, but may be caused by other temporary factors.

[0069] If the cumulative duration of the accumulated effect waveform is consistent across different stages, or if the fluctuation of the deviation does not exceed the set value range (e.g., the set value range is ±1 second), and in multiple faults, the cumulative duration of the accumulated effect waveform is between 4 and 5 seconds, with the deviation not exceeding the set range, then the accumulated effect waveform and the failure effect waveform are jointly marked as the failure mechanism of the current fault source. This indicates that these waveforms are closely related to the occurrence of the fault and are the key basis for fault diagnosis and handling.

[0070] After the fault feature extraction is completed, the fault management platform generates an environmental impact analysis signal and sends it to the environmental impact analysis unit.

[0071] After receiving the environmental impact analysis signal, the environmental impact analysis unit performs environmental impact analysis based on the failure mechanism, which facilitates synchronous monitoring of environmental parameters based on the failure mechanism and regulation mechanism, and ensures the efficiency of monitoring and regulation.

[0072] The waveform transition period obtained based on the failure mechanism is marked as the gradual failure stage; the waveform transition period obtained based on the regulation mechanism is marked as the regulation effectiveness stage.

[0073] During the gradual failure stage and the adjustment effectiveness stage, the operating environment parameters of the operation end are collected, where the environmental parameters are represented by parameters such as humidity and temperature in the operating environment.

[0074] During the gradual failure phase, environmental parameter types that fluctuate beyond the set range are collected and set as widespread interference types. During the phase where the fluctuation of environmental parameters of the widespread interference type intensifies, if the waveform fluctuation slope increases during the gradual failure phase, the corresponding type of environmental parameter is extracted from the widespread interference type and marked as precise interference type.

[0075] During the adjustment effect phase, if the real-time parameter waveform evolves according to the operation adjustment waveform, then the precise disturbance type environmental parameter or the wide disturbance type environmental parameter that floats within the set range is marked as the adjustment control type; otherwise, it is marked as the adjustment non-control type.

[0076] The corresponding environmental parameters are sent to the fault management platform. The fault management platform monitors all types of environmental parameters at each stage to ensure that the operation performed at the corresponding stage achieves the preset effect of the operation itself.

[0077] Specific application scenario: During the gradual failure phase, the normal range for environmental parameters is set as 40%-60% for humidity and 25℃-35℃ for temperature. High-precision environmental sensors are used to collect humidity and temperature parameters in the operating environment in real time. Assuming that at 10:05, the collected humidity value is 70%, exceeding the upper limit of the normal range, and the temperature value is 40℃, also exceeding the upper limit of the normal range, these two environmental parameters are set as broad interference types. As time progresses, at 10:08, the humidity value further increases to 80%, and the temperature value rises to 45℃. Simultaneously, the waveform fluctuation slope increases during the gradual failure phase; for example, the voltage waveform distortion rate increases from 10% to 15%, and the current waveform fluctuation range expands from ±5A to ±8A. At this point, humidity and temperature are extracted from the broad interference type and marked as precise interference types. This indicates that changes in these two environmental parameters have a direct and critical impact on inverter failure.

[0078] During the adjustment effectiveness phase, the evolution of parameter waveforms is monitored in real time. Assuming that at 10:20, after implementing adjustment measures, the real-time collected waveforms of parameters such as voltage and current evolve according to the operational adjustment waveform, such as the voltage waveform gradually recovering from a distorted state to a sine wave shape, and the fluctuation range of the current waveform gradually narrowing, with a similarity to the operational adjustment waveform exceeding 80%, then the environmental parameters (such as humidity and temperature) fluctuating within the current set range are marked as adjustment control types. This indicates that the current adjustment measures have effectively suppressed the negative impact of these environmental parameters on inverter operation, allowing the inverter to recover towards normal operation. Conversely, if the real-time parameter waveforms do not evolve according to the operational adjustment waveform, such as the voltage waveform still exhibiting severe distortion, the current waveform fluctuating abnormally, and a similarity to the operational adjustment waveform below 50%, then the environmental parameters are marked as non-control types. In this case, further analysis of the cause and adjustment of the adjustment measures are needed to ensure the inverter can return to normal operation.

[0079] In summary, the system integrates data from three major units—operation simulation, fault feature extraction, and environmental impact analysis—through a fault management platform, achieving a closed-loop data process from fault location to diagnosis, adjustment, and environmental adaptation. Simultaneously, it continuously accumulates and updates failure and adjustment mechanisms for similar operating units, forming a reusable operation and maintenance knowledge base. This provides data support for the long-term operation and maintenance and preventative maintenance of photovoltaic inverters, reducing overall operation and maintenance costs.

[0080] Thresholds, preset values, preset ranges, etc. are set for result comparison and analysis to determine whether they are good or bad. The value of these thresholds is determined by a combination of large-scale model analysis of sample data and human experience. They can also be adjusted appropriately based on seasonal or common-sense influences.

[0081] Furthermore, the settings for weighting ratios, influence factors, etc., are based on the magnitude of each parameter's influence on the results. The specific values ​​are allocated to ultimately reflect the impact on the results. The settings for input and storage are also determined by a combination of large-scale model analysis of sample data and human experience. Appropriate adjustments can also be made based on seasonal or rational influence conditions.

[0082] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A three-phase photovoltaic inverter fault control system based on data integration, characterized in that, This includes a fault management platform, which is connected to an operation simulation model construction unit, a fault feature extraction unit, and an environmental impact analysis unit. The operation simulation model construction unit establishes a simulation model based on the operation process of the three-phase photovoltaic inverter, and performs operation detection on the simulation model by inputting operation parameters, and obtains various types of time moments and waveforms based on the detection process; The fault feature extraction unit combines various time points and waveforms to extract features, compares the features of the photovoltaic inverter operating parameters, and obtains the failure mechanism and regulation mechanism based on the feature comparison. The environmental impact analysis unit conducts environmental impact analysis based on the failure mechanism, and synchronously monitors environmental parameters for both the failure mechanism and the regulation mechanism to ensure the efficiency of monitoring and regulation. The process of building the simulation model unit is as follows: The simulation model is constructed by collecting data on the internal components and connection circuits of the photovoltaic inverter, and the internal components involved in the operation in the simulation model are marked as the operating end. The task electrical parameters are input, and the operating terminals are sorted according to the operation process of the photovoltaic inverter. The task electrical parameters are represented as current and voltage. The first operating terminal in the sorted order carries the task electrical parameters, and after receiving the input task electrical parameters, it operates the component and produces electrical parameters after completing the component processing. These parameters are marked as processed electrical parameters. The process is iteratively processed according to all sorted operating terminals, and finally, electrical parameters are produced and marked as result electrical parameters. Based on the task electrical parameters and the equipment characteristics of the photovoltaic inverter, the preset output electrical parameters are obtained. If the deviation between the preset output electrical parameters and the result electrical parameters is within the set result parameter deviation range, the current operation process of the photovoltaic inverter is marked as a qualified process. If the deviation between the preset output electrical parameters and the result electrical parameters is not within the set result parameter deviation range, the current operation process of the photovoltaic inverter will be marked as a non-qualified process. During the non-conforming process, the processing electrical parameters at each operating moment are statistically analyzed and recorded. The operating moment is classified according to whether the processing electrical parameters are consistent with the output electrical parameters of the operating end components. If they are consistent, they are classified as slow-state operating end; if they are inconsistent, they are classified as fast-state operating end. In the process of non-compliance, if a fast-state operation end appears, the corresponding operation end is marked as the fault source point, and the fluctuation trend of the input electrical parameters and the fluctuation trend of the output electrical parameters of the fault source point are recorded. If the fluctuation trend of the input electrical parameters at the fault source point does not exceed the set input parameter threshold, but the fluctuation trend of the output electrical parameters exceeds the set output parameter threshold, then the corresponding fluctuation time will be marked as the component abnormal time, and the output electrical parameters of the corresponding operating end will be summarized to form the component abnormal waveform. If the fluctuation trend of the input electrical parameters at the fault source point exceeds the fluctuation value of the set input parameter threshold, and the fluctuation trend of the output electrical parameters exceeds the value fluctuation of the set output parameter threshold, then the corresponding value fluctuation time is marked as the load abnormal time, and the output electrical parameters of the corresponding operating end are summarized to form a load abnormal waveform. If the fluctuation trend of the input electrical parameters at the fault source point exceeds the fluctuation value of the set input parameter threshold, and the fluctuation trend of the output electrical parameters does not exceed the fluctuation value of the set output parameter threshold, then the corresponding fluctuation time will be marked as the operation adjustment time, and the output electrical parameters of the corresponding operation end will be summarized to form the operation adjustment waveform. If the fluctuation trend of the input electrical parameters at the fault source does not exceed the set input parameter threshold, and the fluctuation trend of the output electrical parameters does not exceed the set output parameter threshold, then the corresponding fluctuation time is marked as the stable operation time, and the output electrical parameters at the corresponding operating end are summarized to form a stable operation waveform. The stable operation time and the stable operation waveform are then sent to the fault control platform together.

2. The three-phase photovoltaic inverter fault control system based on data integration according to claim 1, characterized in that, The process of the fault feature extraction unit is as follows: The load anomaly waveform is extracted. Based on the load anomaly time of the non-load anomaly waveform and a waveform statistics period is set, the non-load anomaly waveform and the load anomaly waveform are obtained with the load anomaly time as the dividing point. The numerical waveforms of the input electrical parameters and output electrical parameters during the non-load anomaly waveform stage are recorded and marked as the input waveform before the anomaly and the output waveform before the anomaly, respectively. The numerical waveforms of the input electrical parameters and output electrical parameters are obtained based on the load anomaly waveform and marked as the input waveform after the anomaly and the output waveform after the anomaly, respectively. The input waveforms before and after the anomaly are compared and the non-overlapping waveforms are extracted. The failure mechanism is set based on the non-overlapping waveforms. Combined with the historical running period, if the parameter fluctuation span corresponding to the non-overlapping waveform is generated and the operating end is running normally, this type of non-overlapping waveform is marked as the accumulation effect waveform; if the parameter fluctuation span corresponding to the non-overlapping waveform is generated and the operating end is running abnormally, this type of non-overlapping waveform is marked as the failure effect waveform.

3. The three-phase photovoltaic inverter fault control system based on data integration according to claim 2, characterized in that, Based on the moment of load abnormality, the cumulative duration of the accumulated effect waveform is obtained. If the accumulated duration of the accumulated effect waveform is inconsistent and the deviation fluctuates repeatedly when the same type of operating end is repeatedly used as the source of failure, the accumulated effect waveform is marked as the non-failure mechanism of the current fault source. If the cumulative duration of the accumulated impact waveform is consistent or the deviation fluctuates within a set range when the same type of operating end is repeatedly used as the source of failure, then the accumulated impact waveform and the failure impact waveform are marked as the failure mechanism of the current fault source.

4. The three-phase photovoltaic inverter fault control system based on data integration according to claim 3, characterized in that, The stable operating waveform is extracted, and adjacent load anomaly moments and operation adjustment moments are collected. The interval is marked as a stable period. The execution time of operation adjustment measures within the stable period is recorded. After the operation adjustment measures are executed, the load anomaly waveform and the operation adjustment waveform are compared and overlapped. The trend changes of non-overlapping waveforms are recorded. The current operation adjustment measure execution progress is correlated with the trend changes of non-overlapping waveforms and uniformly marked as adjustment mechanism. The adjustment mechanism is updated in combination with the adjustment process of multiple fault sources. The determined adjustment mechanism is sent to the fault control platform. The fault control platform monitors the adjustment of all fault sources at the current moment and compares the waveform trend changes according to the progress of adjustment measures at different stages. If the actual waveform trend change is inconsistent with the corresponding waveform trend change, it indicates an adjustment anomaly, and the adjustment measures are changed. Conversely, if the actual waveform trend change is consistent with the corresponding waveform trend change, it indicates that the adjustment is normal, and the current adjustment measures should continue to be implemented.

5. The three-phase photovoltaic inverter fault control system based on data integration according to claim 1, characterized in that, The process of environmental impact analysis unit is as follows: Based on the failure mechanism, the waveform transition period is obtained and marked as the gradual failure stage; based on the regulation mechanism, the waveform transition period is obtained and marked as the regulation effectiveness stage; the operating environment parameters are collected at the operating end for the gradual failure stage and the regulation effectiveness stage, where the environmental parameters are represented by humidity and temperature in the operating environment; During the gradual failure phase, environmental parameter types that fluctuate beyond the set range are collected and set as widespread interference types. During the phase where the fluctuation of environmental parameters of the widespread interference type intensifies, if the waveform fluctuation slope increases during the gradual failure phase, the corresponding type of environmental parameter is extracted from the widespread interference type and marked as precise interference type. During the adjustment effect phase, if the real-time parameter waveform evolves according to the operation adjustment waveform, then the precise disturbance type environmental parameter or the wide disturbance type environmental parameter that floats within the set range is marked as the adjustment control type. Conversely, it is marked as a regulation-non-control type.

Images