Robust diagnosis method and device for open-circuit fault of grid-connected three-phase inverter

Abstract

The invention provides a robust diagnosis method and device for an open-circuit fault of a grid-connected three-phase inverter, and relates to the technical field of fault diagnosis, and the method comprises the steps: determining the fault characteristics of a robust working condition, and calculating the detectable degrees of various faults in an operation process under different measurement conditions; determining the robustness degree of the fault to the measurement noise, and screening to obtain a recommended measurement condition; training fault features used for model training are obtained according to the recommended measurement conditions in combination with historical operation data of the grid-connected three-phase inverter in different fault states; training a model according to the training fault features; acquiring real-time operation data in the grid-connected three-phase inverter, and extracting real-time fault features for fault detection according to the real-time operation data; and inputting the real-time fault features into a fault classifier to obtain a diagnosis result. Through the fault detectability, the robustness degree of fault features to sensor noise is analyzed, so that the design of measurement conditions is optimized, and the accuracy of fault diagnosis is improved.

Description

Technical Field

[0001] The present application relates to the technical field of fault diagnosis, and in particular to a robust diagnosis method and device for an open-circuit fault of a grid-connected three-phase inverter. Background Art

[0002] As the core part of the main power electronic equipment and its various new energy power generation devices and drive systems, the safety and reliability of the inverter are related to the safety and stability of the entire system. As one of the main application directions of inverter equipment, the safe and stable operation of the grid-connected inverter has an important impact on improving the reliability of the entire conversion system. Studies have shown that inverters are prone to failure due to the influence of many factors. In particular, damage to power semiconductor switching devices accounts for 38% of inverter failures. Therefore, the fault diagnosis of the power switch tube of the grid-connected inverter is of great significance and value. However, the fault diagnosis of the power switch tube of the grid-connected inverter in the related technology will be affected by many factors, resulting in low accuracy of fault diagnosis, which affects the safe operation of the power switch tube of the grid-connected inverter. Summary of the invention

[0003] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0004] To this end, the first objective of the present application is to propose a robust diagnosis method for open-circuit faults of a grid-connected three-phase inverter.

[0005] The second object of the present application is to provide a device.

[0006] The third objective of the present application is to provide an electronic device.

[0007] A fourth objective of the present application is to provide a computer-readable storage medium.

[0008] A fifth object of the present application is to provide a computer program product.

[0009] To achieve the above-mentioned purpose, the first embodiment of the present application proposes a robust diagnosis method for open-circuit fault of a grid-connected three-phase inverter, comprising: Determine the fault characteristics of the robust operating conditions and calculate the detectability of various faults during the operation of the grid-connected three-phase inverter under different measurement conditions; Determining the robustness of the fault to measurement noise according to the detectability of the fault, and screening and obtaining recommended measurement conditions according to the robustness; According to the recommended measurement conditions and in combination with historical operation data of the grid-connected three-phase inverter under different fault conditions, training fault features for model training are obtained; Training a model according to the training fault features to obtain a fault classifier; Collecting real-time operating data from the grid-connected three-phase inverter, and extracting real-time fault features for fault detection based on the real-time operating data; The real-time fault features are input into the fault classifier to obtain a diagnosis result.

[0010] Optionally, the calculation formula for the relative detectability of the fault is:

[0011]

[0012] in, and For fault conditions, For failure and The relative detectability of and is the total set of fault characteristics, Indications and faults The corresponding probability distribution of the features, Indications and faults The probability distribution of the corresponding features, where , is the kernel function, for The mth sample in for The nth sample in for The mth sample in for The nth sample in for The sum of the samples in for The sum of the samples in .

[0013] Optionally, the calculation formula for the robustness of the fault to measurement noise is:

[0014] in, is the robustness of the fault to measurement noise, , , Indicates a fault in the presence of noise The probability distribution of the corresponding feature, Indicates a fault in the presence of noise The probability distribution of the corresponding feature.

[0015] Optionally, the screening according to the robustness to obtain the recommended measurement conditions includes: Obtain the noise robustness corresponding to different time windows at each signal-to-noise ratio, and determine the signal-to-noise ratio range in which the robustness corresponding to different time windows is greater than a preset threshold; The time window corresponding to the maximum signal-to-noise ratio range that meets the robustness threshold is used as the recommended measurement condition.

[0016] Optionally, the training fault features are obtained according to the recommended measurement conditions combined with historical operation data of the grid-connected three-phase inverter under different fault states, including: Fault features are extracted from the historical operation data, and the training fault features are generated in combination with the time window in the recommended measurement condition.

[0017] To achieve the above-mentioned purpose, the second embodiment of the present application proposes a robust diagnostic device for open-circuit fault of a grid-connected three-phase inverter, comprising: A first calculation module is used to determine the fault characteristics of the robust operating condition and calculate the detectability of various faults during the operation of the grid-connected three-phase inverter under different measurement conditions; A condition screening module, used to determine the robustness of the fault to measurement noise according to the detectability of the fault, and to screen and obtain recommended measurement conditions according to the robustness; A first acquisition module is used to obtain training fault features for model training based on the recommended measurement conditions combined with historical operation data of the grid-connected three-phase inverter under different fault states; A training module, used for training a model according to the training fault features to obtain a fault classifier; A second acquisition module is used to collect real-time operation data in the grid-connected three-phase inverter, and extract real-time fault features for fault detection based on the real-time operation data; The classification module is used to input the real-time fault characteristics into the fault classifier to obtain a diagnosis result.

[0018] To achieve the above-mentioned purpose, the third aspect of the present application provides an electronic device, including: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method as described in any one of the first aspects.

[0019] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present application proposes a computer-readable storage medium, in which computer-readable storage medium is stored computer execution instructions, and when the computer execution instructions are executed by a processor, they are used to implement the method as described in any one of the first aspects.

[0020] To achieve the above-mentioned purpose, the fifth aspect of the present application proposes a computer program product, which implements any method in the first aspect when executed by a processor.

[0021] The robust diagnosis method, device, electronic device and storage medium for open-circuit fault of a grid-connected three-phase inverter provided by the present application analyze the robustness of fault characteristics to sensor noise through fault detectability, thereby optimizing the design of measurement conditions and improving the accuracy of fault diagnosis. The fault characteristics obtained after optimizing the measurement conditions enhance the robustness of the diagnostic method to sensor noise.

[0022] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 A schematic flow chart of a robust diagnosis method for an open-circuit fault of a grid-connected three-phase inverter provided in an embodiment of the present application; Figure 2 It is a schematic diagram of the topological structure of a grid-connected three-phase inverter; Figure 3 A unified circuit model for IGBT faults; Figure 4 It is a fault diagnosis flow chart; Figure 5 A schematic diagram of the structure of a robust diagnostic device for open-circuit faults of a grid-connected three-phase inverter provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0025] As the core part of the main power electronic equipment and its various new energy power generation devices and drive systems, the safety and reliability of the inverter are related to the safety and stability of the entire system. As one of the main application directions of inverter equipment, the safe and stable operation of the grid-connected inverter has an important impact on improving the reliability of the entire conversion system. Studies have shown that inverters are prone to failure due to the influence of many factors. In particular, damage to power semiconductor switching devices accounts for 38% of inverter failures. Therefore, fault diagnosis of power switch tubes of grid-connected inverters is of great significance and value.

[0026] Insulated gate bipolar transistor (IGBT) is the most commonly used power switching device in inverters. It is very easy to fail because it is subjected to large electrothermal stress for a long time, and IGBT failures are mainly manifested as open circuit failures and short circuit failures. At present, there are very mature technologies to quickly cut off short circuit failures. However, after an open circuit failure occurs, the inverter can still operate for a short time. If it is not handled in time, it will cause secondary failures. Therefore, it is of great significance to study the open circuit fault diagnosis method of IGBT in grid-connected inverters.

[0027] Existing fault diagnosis technologies for power switch tubes can be roughly divided into model-based, signal-based and data-driven fault diagnosis methods. The accuracy of model-based fault diagnosis methods is highly dependent on the accuracy and detail of the circuit model, but building such a circuit model is very complex and time-consuming. Signal-based fault diagnosis methods generally have problems such as difficulty in setting thresholds, the need for prior knowledge of faults, the impact of load fluctuations (based on current signals) and increased system costs (based on voltage signals). Compared with the above two methods, data-driven fault diagnosis methods do not require too deep an understanding of the system and mainly rely on the data itself. The quality of the data largely determines the accuracy of the diagnosis method. This method only uses data to train the diagnostic model and then performs fault diagnosis by establishing a classifier. Therefore, it has strong adaptability and has attracted widespread attention.

[0028] However, in order to improve the fault diagnosis capability of power switch tubes of grid-connected inverters based on data-driven methods, existing research mainly focuses on the improvement and optimization of specific diagnostic algorithms, and there is less research on fault detectability. However, the improvement and optimization of fault diagnosis algorithms are meaningful only when the fault can be diagnosed. Therefore, blindly designing fault diagnosis algorithms without analyzing fault detectability will inevitably waste manpower and material resources, and it is very likely that satisfactory results will not be achieved. Fault detectability, as the basis of fault diagnosis, analyzes the attributes that characterize the system's fault diagnosis capability from the essential level. Its research provides new ideas for improving fault diagnosis capabilities and is of great significance. However, there are few reports on fault detectability analysis for the fault diagnosis problem of power switch tubes of grid-connected inverters.

[0029] To this end, this paper studies a fault diagnosis method that is robust to operating conditions and sensor noise for open-circuit faults of IGBTs. A model-based fault detectability analysis method is used to determine fault characteristics that are robust to operating conditions and to determine an optimal measurement condition. Specifically, through fault detectability analysis theory and a fault state equation model, fault characteristics that are robust to changes in operating conditions are obtained; through the calculation of fault detectability analysis, the influence of sensor noise on fault detectability is analyzed, and then an optimal time window that satisfies fault diagnosability is obtained, so that the design of measurement conditions can be optimized.

[0030] Robust working condition refers to the working state in which the system can maintain stable operation and meet performance requirements in the face of various uncertain factors (such as parameter changes, external interference, model errors, etc.). This working condition emphasizes the reliability and stability of the system in a complex environment, ensuring that basic functions can be maintained even in the worst case.

[0031] In a grid-connected three-phase inverter system, the output power of the grid-connected inverter is different under different working conditions. The robust working condition refers to the working condition where the fault characteristics are not affected under different inverter output powers.

[0032] The embodiment of the present application provides a robust diagnosis method for open circuit fault of a grid-connected three-phase inverter. Figure 1 The flowchart of a robust diagnosis method for open circuit fault of a grid-connected three-phase inverter provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method comprises the following steps: Step 101, determining the fault characteristics of the robust operating condition, and calculating the detectability of various faults during the operation of the grid-connected three-phase inverter under different measurement conditions; Step 102, determining the robustness of the fault to measurement noise according to the detectability of the fault, and screening to obtain recommended measurement conditions according to the robustness; Step 103, obtaining training fault features for model training according to the recommended measurement conditions combined with historical operation data of the grid-connected three-phase inverter under different fault states; Step 104, training a model according to the training fault features to obtain a fault classifier; Step 105, collecting real-time operation data from the grid-connected three-phase inverter, and extracting real-time fault features for fault detection based on the real-time operation data; Step 106: input the real-time fault feature into the fault classifier to obtain a diagnosis result.

[0033] The state space expression of the discrete-time system is (1) in, and are state variables, input signals, actuator fault signals, and output signals, respectively. Indicates that the sensor measures a noisy signal. , , , , , and is an unknown constant real matrix of appropriate dimension.

[0034] We introduce the following notations: (2) (3) (4) have In the following research, Can be , , , and ; express , , , and ; and is an integer; represents the time window length; and Indicates the number of test data.

[0035] Using the above notation, (1) can be written as: (5) in, , , , , .

[0036] Has the following form

[0037] Introducing superscript Indicates data under various fault conditions, where and is a positive integer, in particular, Indicates no fault condition.

[0038] Therefore, according to the introduced superscript, we can obtain the value corresponding to the first The formula for the fault condition is: (6) express The left null space of By simultaneously left-multiplying Eliminating the state term, we can obtain (7): (7) (8) Use (8) to express The time distribution of the fault conditions, where Represents at discrete time points When the fault vector In the The value of the fault condition. From formula (7-8), we can get (9): (9) From (9), we can see that the dynamic behavior of the system is a random probability distribution described based on the input and output data, and this distribution is only affected by the fault vector and the noise vector.

[0039] If the difference between the random probability distributions of the system dynamic behavior corresponding to the two fault conditions is greater, it means that it is easier to isolate the two faults. That is, the fault detectability can be analyzed by the difference between the probability distributions.

[0040] Figure 2 is a schematic diagram of the topological structure of a grid-connected three-phase inverter, as shown in Figure 2 As shown, the grid-connected three-phase inverter is the DC side voltage; There are 6 power switch tubes; is the filter inductor; is the equivalent resistance; is the grid-side voltage; is a three-phase current.

[0041] Figure 3 The unified circuit model of IGBT fault is as follows: Figure 3 As shown in the figure, since the open-circuit fault of the IGBT will affect the output voltage of the bridge arm, this paper adopts the form of superimposed controlled sources to represent the unified circuit model of the IGBT fault.

[0042] in, Represents normal situation The voltage between two points, They respectively represent the changes introduced to the voltage at the corresponding two points during a fault.

[0043] Based on the unified circuit model of the above faults and combined with Kirchhoff's law, the state equation model of the fault can be obtained as follows: (10) in: Represents the current path.

[0044] The state variable , input signal , output signal , IGBT fault signal , sensor sampling noise signal .

[0045] According to the above preliminary knowledge, the expression of the dynamic behavior of the grid-connected three-phase inverter can be obtained as follows: (11) in, for The left null space of . , , .

[0046]

[0047]

[0048] is the sampling step length, for The identity matrix of .

[0049] make (12) According to (11), the fault characteristics It is only affected by fault vectors and noise vectors, so it is robust to changes in system operating conditions.

[0050] Optionally, the calculation formula for the relative detectability of the fault is:

[0051] (13) in, and For fault conditions, For failure and The relative detectability of and is the total set of fault characteristics, Indications and faults The corresponding probability distribution of the features, Indications and faults The probability distribution of the corresponding features, where , is the kernel function, for The mth sample in for The nth sample in for The mth sample in for The nth sample in for The sum of the samples in for The sum of the samples in .

[0052] From (12), we can see that the fault characteristics is a random vector that satisfies a certain probability distribution. We can use the difference in the probability distribution of the characteristics of different fault types to define the degree of fault detectability. The reason for this definition is that by analyzing the probability distribution of different fault characteristics, the difference between them can be determined. And the more obvious the difference, the easier it is to isolate the two faults.

[0053] The calculation expression of detectability can be: (14) in, , is the kernel function in the reproducing kernel Hilbert space.

[0054] Optionally, the calculation formula for the robustness of the fault to measurement noise is: (15) in, is the robustness of the fault to measurement noise, , , Indicates a fault in the presence of noise The probability distribution of the corresponding feature, Indicates a fault in the presence of noise The probability distribution of the corresponding feature.

[0055] like , it means that the noise reduces the detection ability, and the larger the value, the greater the degree of reduction. , it means that the noise improves the detection ability.

[0056] Optionally, the step 102 of obtaining recommended measurement conditions by screening according to the robustness level includes: Obtain the noise robustness corresponding to different time windows at each signal-to-noise ratio, and determine the signal-to-noise ratio range in which the robustness corresponding to different time windows is greater than a preset threshold; The time window corresponding to the maximum signal-to-noise ratio range that meets the robustness threshold is used as the recommended measurement condition.

[0057] In this embodiment, in order to ensure the robustness of fault diagnosis to measurement noise as much as possible, the greater the robustness of the fault feature to measurement noise, the better. Therefore, by enumerating different signal-to-noise ratios and time window lengths, the robustness of the fault feature to measurement noise under each corresponding condition can be obtained, and then a three-dimensional graph can be obtained. Through this graph, we can select the time window according to the size of the robustness of the fault feature to measurement noise. The corresponding robustness under this time window has a good robustness in a large signal-to-noise ratio range.

[0058] Optionally, in step 103, obtaining training fault features according to the recommended measurement conditions combined with historical operation data of the grid-connected three-phase inverter under different fault states includes: Fault features are extracted from the historical operation data, and the training fault features are generated in combination with the time window in the recommended measurement condition.

[0059] Figure 4 It is a fault diagnosis flowchart, such as Figure 4As shown in the figure, the main process includes offline training and online diagnosis. In the offline training stage, the fault characteristics of the robust working condition are first obtained by using the preliminary knowledge and the fault state equation model. Then, through the fault detectability, the robustness of the characteristics to the measurement noise is analyzed, and the measurement conditions are optimized. The data collected under different fault states are used to form the fault characteristics of the robust working condition and noise through the optimization of measurement conditions and the feature extraction of the robust working condition. This feature is used for model training to obtain a fault classifier. In the online diagnosis stage, the real-time collected data is combined with the fault features extracted in this paper and input into the fault classifier to form a diagnosis result.

[0060] In a possible embodiment, the simulation parameters of the grid-connected three-phase inverter are set as follows: Three-phase power grid 220V50Hz, sampling frequency 100K, switching frequency 10K, R=0.3Ω, L=8e-3H, U DC =400V Preferred measurement conditions: Experimental purpose: to form a three-dimensional graph (signal-to-noise ratio, time window length, robustness of fault signature to measurement noise) Experimental steps: ① Collect the failure modes through simulation of Data (the corresponding coefficient matrix has been calculated) ② Enumerate different time windows and calculate the corresponding time window

[0061] ③ Enumerate different signal-to-noise ratios, take different time windows as known conditions, and calculate the corresponding time windows and signal-to-noise ratios.

[0062] ④ According to (14) and (15), calculate the corresponding window and signal-to-noise ratio

[0063] ⑤ Taking time window as axis, the signal-to-noise ratio is Axis, degree Axis, forming a three-dimensional graph Working condition robustness test To verify that the extracted fault features are robust, change the grid voltage from 220V to (for example, 230V, 240V...300V) and collect the As a test set, the average fault diagnosis accuracy of several experiments is calculated.

[0064] Under the condition of grid voltage fluctuation, the diagnostic method based on the fault features extracted in this paper is robust to the fault diagnosis of the system.

[0065] Change the DC voltage from 400V to (for example, 410V, 420V...500V) and collect the As the test set, the average fault diagnosis accuracy of several experiments is calculated. Under the condition of DC voltage fluctuation, the diagnosis method based on the fault features extracted in this paper is robust to the fault diagnosis of the system.

[0066] Sensor noise robustness experiment In order to verify the robustness of the extracted fault features to sensor noise under the recommended measurement conditions, Gaussian white noise with a certain signal-to-noise ratio is superimposed on the test set, and the average fault diagnosis accuracy of several experiments is calculated. Under the recommended measurement conditions, the diagnostic method based on the fault features extracted in this paper is robust to sensor noise for the fault diagnosis of the system.

[0067] In order to implement the above-mentioned embodiment, the present application also proposes a robust diagnosis device for open-circuit fault of a grid-connected three-phase inverter. Figure 5 The present invention provides a schematic diagram of a robust diagnostic device for open-circuit faults of a three-phase grid-connected inverter. Figure 5 As shown, the device comprises: A first calculation module 510, for determining fault characteristics of a robust operating condition and calculating the detectability of various faults during operation of the grid-connected three-phase inverter under different measurement conditions; A condition screening module 520, configured to determine the robustness of the fault to measurement noise according to the detectability of the fault, and to screen and obtain recommended measurement conditions according to the robustness; A first acquisition module 530 is used to obtain training fault features for model training based on the recommended measurement conditions combined with historical operation data of the grid-connected three-phase inverter under different fault states; A training module 540 is used to train a model according to the training fault features to obtain a fault classifier; A second acquisition module 550 is used to collect real-time operation data in the grid-connected three-phase inverter, and extract real-time fault features for fault detection based on the real-time operation data; The classification module 560 is used to input the real-time fault feature into the fault classifier to obtain a diagnosis result.

[0068] In order to implement the above embodiments, the present application also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments.

[0069] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0070] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0071] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0072] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0073] The present application is expected to provide an implementation scheme for users to selectively block the use or access of personal information data. That is, the present disclosure is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, the risk can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0074] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they contradict each other.

[0075] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0076] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0077] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0078] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0079] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0080] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0081] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A robust diagnosis method for open circuit fault of a grid-connected three-phase inverter, characterized in that: The following steps are involved: Determine the fault characteristics of the robust operating condition, and calculate the detectability of various faults during the operation of the grid-connected three-phase inverter under different measurement conditions, wherein the robust operating condition is a condition where the fault characteristics are not affected under different inverter output powers; Determining the robustness of the fault to measurement noise according to the detectability of the fault, and screening and obtaining recommended measurement conditions according to the robustness; According to the recommended measurement conditions and in combination with historical operation data of the grid-connected three-phase inverter under different fault conditions, training fault features for model training are obtained; Training a model according to the training fault features to obtain a fault classifier; Collecting real-time operating data from the grid-connected three-phase inverter, and extracting real-time fault features for fault detection based on the real-time operating data; The real-time fault features are input into the fault classifier to obtain a diagnosis result.

2. The method according to claim 1, characterized in that The calculation formula of the detectability of the fault is: in, and For fault conditions, For failure and The relative detectability of and is the total set of fault characteristics, Indications and faults The corresponding probability distribution of the features, Indications and faults The probability distribution of the corresponding features, where , is the kernel function, for The mth sample in for The nth sample in for The mth sample in for The nth sample in for The sum of the samples in for The sum of the samples in .

3. The method according to claim 2, characterized in that The calculation formula of the robustness of the fault to measurement noise is: in, is the robustness of the fault to measurement noise, , , Indicates a fault in the presence of noise The probability distribution of the corresponding feature, Indicates a fault in the presence of noise The probability distribution of the corresponding feature.

4. The method according to claim 3, characterized in that The step of screening and obtaining the recommended measurement conditions according to the robustness level includes: Obtain the noise robustness corresponding to different time windows at each signal-to-noise ratio, and determine the signal-to-noise ratio range in which the robustness corresponding to different time windows is greater than a preset threshold; The time window corresponding to the maximum signal-to-noise ratio range that meets the robustness threshold is used as the recommended measurement condition.

5. The method according to claim 4, characterized in that The training fault features are obtained according to the recommended measurement conditions combined with the historical operation data of the grid-connected three-phase inverter under different fault states, including: Fault features are extracted from the historical operation data, and the training fault features are generated in combination with the time window in the recommended measurement condition.

6. A robust diagnostic device for open-circuit fault of a grid-connected three-phase inverter, characterized in that: include: A first calculation module is used to determine the fault characteristics of the robust operating condition and calculate the detectability of various faults during the operation of the grid-connected three-phase inverter under different measurement conditions; A condition screening module, used to determine the robustness of the fault to measurement noise according to the detectability of the fault, and to screen and obtain recommended measurement conditions according to the robustness; A first acquisition module is used to obtain training fault features for model training based on the recommended measurement conditions combined with historical operation data of the grid-connected three-phase inverter under different fault states; A training module, used for training a model according to the training fault features to obtain a fault classifier; A second acquisition module is used to collect real-time operation data in the grid-connected three-phase inverter, and extract real-time fault features for fault detection based on the real-time operation data; The classification module is used to input the real-time fault characteristics into the fault classifier to obtain a diagnosis result.

7. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 5 when executed by a processor.

9. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 5 when being executed by a processor.

Images