DC converter fault diagnosis method, device, equipment, medium and program product

Abstract

The present invention discloses a DC converter fault diagnosis method, device, equipment, medium and program product. The method comprises: collecting historical operation data and fault sample data of the DC converter, preprocessing the historical operation data and the fault sample data to obtain a data set, constructing a kernel function based on the data set, constructing an original model, and training and optimizing the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model comprises a support vector machine, collecting real-time operation data of the DC converter, preprocessing the real-time operation data, inputting the preprocessed real-time operation data into the fault diagnosis model for fault diagnosis, and obtaining the fault probability and fault type of the DC converter, thereby realizing monitoring of the operation state of the DC converter, ensuring the timeliness of fault diagnosis, effectively improving the accuracy of fault diagnosis, providing reliable diagnosis results for fault processing, and improving the response speed of fault processing.

Description

Technical Field

[0001] The present invention relates to the field of power electronics technology, and in particular to a direct current converter fault diagnosis method, device, equipment, medium and program product. Background Art

[0002] With the rapid development of power electronics technology, DC converters have been widely used in power systems, electric vehicles, data centers, etc. However, DC converters are easily affected by factors such as environment and load changes during operation, resulting in frequent failures.

[0003] Traditional fault diagnosis methods mostly rely on manual inspections and offline tests, which have problems such as long diagnosis cycles, low efficiency, large errors, and slow fault handling response speed. In addition, as DC converters develop towards high power density, high efficiency, and high reliability, higher requirements are placed on the real-time and accuracy of fault diagnosis. Summary of the invention

[0004] The main purpose of the present invention is to provide a DC converter fault diagnosis method, device, equipment, medium and program product, aiming to solve the technical problems of the prior art that the fault diagnosis cycle is long and the diagnosis error is large, resulting in low fault handling efficiency and slow fault handling response speed.

[0005] To achieve the above object, the present invention provides a DC converter fault diagnosis method, the method comprising the following steps:

[0006] Collecting historical operation data and fault sample data of the DC converter, wherein the historical operation data includes historical current data, historical voltage data and historical temperature data;

[0007] Preprocessing the historical operation data and the fault sample data to obtain a data set;

[0008] constructing a kernel function based on the data set;

[0009] Constructing an original model, and training and optimizing the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model includes a support vector machine;

[0010] Collecting real-time operating data of the DC converter and preprocessing the real-time operating data, wherein the real-time operating data includes current current data, current voltage data and current temperature data of the DC converter;

[0011] The preprocessed real-time operation data is input into the fault diagnosis model for fault diagnosis to obtain the fault probability and fault type of the DC converter.

[0012] Optionally, the data set includes a fault type and operating parameters of the DC converter, and the operating parameters include current operating parameters, voltage operating parameters, and temperature operating parameters; and constructing a kernel function based on the data set includes:

[0013] Generate initialization particle parameters based on the data set, the initialization particle parameters include: population number, position boundary, speed boundary and iteration number threshold, the population number is generated based on the number of fault types of the DC converter, the position boundary is generated based on the voltage threshold, current threshold and temperature threshold of the DC converter, and the speed boundary is generated based on the current change rate, voltage change rate and temperature change rate of the operating parameter;

[0014] Generate the initial position and initial velocity of the population particles according to the population quantity, the position boundary and the velocity boundary:

[0015]

[0016]

[0017] in, is the initial position of the particle at the initial time, is the initial velocity of the particle at the initial moment, is the minimum value of particle position, is the maximum particle position, is the minimum particle velocity, is the maximum particle velocity, is a random function;

[0018] Determine the historical optimal position and the global optimal position according to the operating parameters;

[0019] The position and velocity of each particle are updated based on the historical optimal position, the global optimal position, the initial position and the initial velocity until the number of updates reaches the iteration number threshold, or the updated particle meets the preset global optimal convergence condition;

[0020] The kernel parameters are determined according to the updated particles, and the kernel function is constructed based on the kernel parameters:

[0021]

[0022] in, is the kernel function, is the kernel parameter.

[0023] Optionally, determining the historical optimal position and the global optimal position according to the operating parameters includes:

[0024] Determine error information between actual operating parameters and theoretical operating parameters of the DC converter according to the operating parameters;

[0025] Construct a fitness function based on the error information:

[0026]

[0027] in, is the fitness value, is the actual operating parameter, are theoretical operating parameters;

[0028] The fitness value of each particle is calculated based on the fitness function, and the historical optimal position and the global optimal position are determined based on the fitness value.

[0029] Optionally, updating the position and velocity of each particle based on the historical optimal position, the global optimal position, the initial position and the initial velocity includes:

[0030] The speed of each particle is updated based on the historical optimal position, the global optimal position and the initial speed:

[0031]

[0032] in, and is a random number between [0, 1], The best position in history. is the global optimal position, is the inertia weight, and is the acceleration coefficient, For the The speed of the update, For the The location at the time of the update, For the Speed ​​after update;

[0033] Update the position of each particle based on the initial position:

[0034]

[0035] in, For the The updated position, For the The location at the time of the update, is the step length control quantity, is a random step length path;

[0036] The above random step length path is generated according to the following formula:

[0037]

[0038] in, and is a random variable generated based on a normal distribution, is the Levy distribution parameter, is the current position of the particle, For interval The particle position.

[0039] Optionally, the constructing the original model, and training and optimizing the original model based on the kernel function to obtain the fault diagnosis model includes:

[0040] Determine the operation characteristic information and the fault type of the DC converter according to the historical operation data and the fault sample data;

[0041] Determine a model input characteristic parameter and a model output target parameter based on the operation characteristic information and the fault mode, wherein the model input characteristic parameter includes a current parameter, a voltage parameter and a temperature parameter, and the model output target parameter includes a fault type and a fault probability;

[0042] Obtaining the DC converter fault judgment condition, wherein the fault judgment condition includes:

[0043]

[0044]

[0045]

[0046] in, When the current voltage exceeds the upper threshold, The current voltage is lower than the lower threshold. When the current exceeds the upper threshold, The current temperature exceeds the upper threshold. The current voltage operating trend does not match the expected voltage operating trend. The current operating trend does not match the expected current operating trend. The current temperature operating trend does not match the expected temperature operating trend;

[0047] An original model is constructed according to the fault judgment condition, the model input characteristic parameter and the model output target parameter, and the original model is trained and optimized based on the kernel function to obtain a fault diagnosis model.

[0048] Optionally, the preprocessing includes normalization processing and smoothing processing;

[0049] The normalization process includes:

[0050]

[0051] in, are operating parameters in historical operating data and real-time operating data, wherein the operating parameters include current parameters, voltage parameters and temperature parameters, is the minimum value of the operating parameter, is the maximum value of the operating parameter, is the normalized operating parameter;

[0052] The smoothing process comprises:

[0053]

[0054] in, is the exponential smoothing value at time point t, is the original operating parameter at time point t, is the exponential smoothing value at time point t-1, is a smoothing constant, and the value range of the smoothing constant is [0, 1].

[0055] In addition, to achieve the above-mentioned purpose, the present invention also proposes a DC converter fault diagnosis device, the DC converter fault diagnosis device comprising:

[0056] A historical data collection module, used to collect historical operation data and fault sample data of the DC converter, wherein the historical operation data includes historical current data, historical voltage data and historical temperature data;

[0057] A data processing module, used for preprocessing the historical operation data and the fault sample data to obtain a data set;

[0058] A kernel function construction module, used for constructing a kernel function based on the data set;

[0059] A model building module, used for building an original model, and training and optimizing the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model includes a support vector machine;

[0060] A real-time data acquisition module, used to acquire real-time operation data of the DC converter and pre-process the real-time operation data, wherein the real-time operation data includes current current data, current voltage data and current temperature data of the DC converter;

[0061] The fault diagnosis module is used to input the pre-processed real-time operation data into the fault diagnosis model for fault diagnosis to obtain the fault probability and fault type of the DC converter.

[0062] In addition, to achieve the above-mentioned objectives, the present application also proposes a DC converter fault diagnosis device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the DC converter fault diagnosis method as described above.

[0063] In addition, to achieve the above-mentioned purpose, the present application also proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the DC converter fault diagnosis method described above are implemented.

[0064] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the DC converter fault diagnosis method described above are implemented.

[0065] The present invention collects historical operation data and fault sample data of a DC converter, wherein the historical operation data includes historical current data, historical voltage data and historical temperature data, pre-processes the historical operation data and the fault sample data to obtain a data set, constructs a kernel function based on the data set, constructs an original model, and trains and optimizes the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model includes a support vector machine, collects real-time operation data of the DC converter, and pre-processes the real-time operation data, wherein the real-time operation data includes current current data, current voltage data and current temperature data of the DC converter. , the preprocessed real-time operation data is input into the fault diagnosis model for fault diagnosis, and the fault probability and fault type of the DC converter are obtained; because the present invention obtains a data set by preprocessing the historical operation data and the fault sample data, a kernel function is constructed based on the data set, and the original model is trained and optimized based on the kernel function, so as to construct an efficient and accurate fault diagnosis model, and the real-time operation data of the DC converter is analyzed by the fault diagnosis model to realize the monitoring of the operating state of the DC converter, ensure the timeliness of the fault diagnosis, effectively improve the accuracy of the fault diagnosis, provide reliable diagnosis results for fault processing, and improve the response speed of fault processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings herein are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the description, are used to explain the principles of the present application.

[0067] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0068] Figure 1 It is a structural schematic diagram of a DC converter fault diagnosis device in a hardware operating environment involved in an embodiment of the present invention;

[0069] Figure 2 It is a flow chart of a first embodiment of a DC converter fault diagnosis method of the present invention;

[0070] Figure 3 It is a flow chart of a second embodiment of a DC converter fault diagnosis method of the present invention;

[0071] Figure 4 It is a flow chart of a third embodiment of a DC converter fault diagnosis method of the present invention;

[0072] Figure 5 This is a diagram of the DC converter fault diagnosis system architecture according to an embodiment of the present invention;

[0073] Figure 6 This is a structural block diagram of the first embodiment of the DC converter fault diagnosis device of the present invention.

[0074] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0075] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0076] Reference Figure 1 , Figure 1 The present invention is a schematic diagram of the structure of a DC converter fault diagnosis device in the hardware operating environment involved in the embodiment of the present invention.

[0077] like Figure 1As shown, the DC converter fault diagnosis device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0078] Those skilled in the art will understand that Figure 1 The structure shown in does not constitute a limitation on the DC converter fault diagnosis device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0079] like Figure 1 As shown, the memory 1005 as a computer-readable storage medium may include an operating system, a network communication module, a user interface module, and a DC converter fault diagnosis program.

[0080] exist Figure 1 In the DC converter fault diagnosis device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the DC converter fault diagnosis device of the present invention can be set in the DC converter fault diagnosis device, and the DC converter fault diagnosis device calls the DC converter fault diagnosis program stored in the memory 1005 through the processor 1001, and executes the DC converter fault diagnosis method provided by the embodiment of the present invention.

[0081] The embodiment of the present invention provides a DC converter fault diagnosis method, referring to Figure 2 , Figure 2 FIG. 1 is a flow chart of a first embodiment of a DC converter fault diagnosis method according to the present invention.

[0082] In this embodiment, the DC converter fault diagnosis method includes the following steps:

[0083] Step S10: Collecting historical operation data and fault sample data of the DC converter.

[0084] It should be understood that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or a terminal electronic device capable of realizing the above functions, etc. The following takes a DC converter fault diagnosis device (hereinafter referred to as a diagnosis device) as an example to illustrate this embodiment and the following embodiments.

[0085] It should be noted that the above historical operation data may be operation parameters of the DC converter during its historical operation, and the above historical operation data include historical current data, historical voltage data and historical temperature data.

[0086] It is understandable that the diagnostic device can collect the input voltage U of the DC converter through the current sensor, voltage sensor, temperature sensor and other sensor devices installed on the DC converter. in 、Input current I in , output voltage U out , output current I out , Temperature T c And other operating parameters.

[0087] In some embodiments, the diagnostic device can obtain the historical operation log and fault log of the DC converter, obtain the historical operation data of the DC converter based on the historical operation log, align the data based on the timestamps of the historical operation log and the fault log, thereby obtaining the fault label corresponding to the historical operation data, and generate fault sample data based on the fault label.

[0088] Step S20: pre-processing the historical operation data and the fault sample data to obtain a data set.

[0089] It should be noted that the preprocessing may include data cleaning, normalization processing, smoothing processing, etc. The above data set may include a training set and a test set.

[0090] In some embodiments, the diagnostic device may pre-process the collected historical voltage, current, temperature and other data, including data cleaning, normalization, smoothing or other necessary operations to improve the data quality and the training effect of the model.

[0091] Step S30: constructing a kernel function based on the data set.

[0092] It should be noted that the choice of kernel function has an important impact on the performance of the model. The kernel function is used to allow the model to work efficiently even in nonlinear situations. The role of the kernel function is to calculate the similarity between two sample points in the original input space, which can analyze the correlation between the operating data and the fault samples, thereby realizing the similarity calculation between the operating data and accurately matching the fault label for the operating data of the input model. For example, you can choose the radial basis function kernel (RBF Kernel), polynomial kernel, etc.

[0093] Step S40: constructing an original model, and training and optimizing the original model based on the kernel function to obtain a fault diagnosis model.

[0094] It should be noted that the original model includes a support vector machine (SVM) or other classification models, and this embodiment is described by taking the support vector machine as an example.

[0095] In some embodiments, the diagnostic device may divide the data set into a training set and a test set, optimize the optimal parameters based on the training set to obtain the optimal parameters, construct a kernel function based on the optimal parameters and the test set, train and optimize the original model based on the kernel function, and obtain a fault diagnosis model.

[0096] Step S50: collecting real-time operating data of the DC converter and preprocessing the real-time operating data.

[0097] It should be understood that the diagnostic device can collect the input voltage U of the DC converter in real time through the current sensor, voltage sensor, temperature sensor and other sensor devices installed on the DC converter. in 、Input current I in , output voltage U out , output current I out , Temperature T c And other operating parameters, so as to realize real-time monitoring of the DC converter.

[0098] It should be noted that preprocessing may include data cleaning, normalization processing, smoothing processing, etc. This embodiment preprocesses real-time operation data to improve fault diagnosis efficiency and diagnosis accuracy, and prevents invalid data and messy data from interfering with fault diagnosis results.

[0099] Step S60: inputting the preprocessed real-time operation data into the fault diagnosis model for fault diagnosis to obtain the fault probability and fault type of the DC converter.

[0100] It should be noted that the diagnostic device can determine whether the DC converter has a fault based on the fault probability, for example, by determining whether the fault probability exceeds a preset probability threshold. If it exceeds, it is determined that there is a fault risk; if it does not exceed, it is determined that there is no fault risk. The above fault types are used to determine the fault mode, faulty components and fault conditions of the DC converter. The diagnostic device can accurately locate the fault of the DC converter based on the fault type.

[0101] In some embodiments, the diagnostic device performs fault processing according to the fault diagnosis results output by the fault diagnosis model. After detecting that a fault occurs in the DC converter, the diagnostic device can automatically take corresponding fault processing measures according to the type and severity of the fault.

[0102] For example, in the case of a short circuit, the diagnostic equipment can immediately cut off the power supply to prevent further damage; in the case of an overload fault, normal operation can be restored by adjusting the load or adding cooling measures. At the same time, the diagnostic equipment can also generate detailed fault reports and maintenance suggestions to provide maintenance personnel with convenient fault location and repair guidance. The fault report should include detailed information such as the fault type, occurrence time, fault parameters, as well as recommended maintenance steps and precautions.

[0103] Furthermore, in order to improve data processing efficiency, fault diagnosis accuracy and response speed, the preprocessing includes normalization processing and smoothing processing;

[0104] It should be noted that the diagnostic equipment can pre-process the collected historical voltage, current, temperature and other data, including normalization, smoothing or other necessary operations, to improve the data quality and the training effect of the model.

[0105] It is understandable that the diagnostic device can normalize the collected voltage, current and temperature data to ensure that each feature is in the value range of [0, 1]. This algorithm adopts the minimum-maximum normalization method, and the formula is as follows: The normalization process includes:

[0106]

[0107] in, are operating parameters in historical operating data and real-time operating data, wherein the operating parameters include current parameters, voltage parameters and temperature parameters, is the minimum value of the operating parameter, is the maximum value of the operating parameter, is the normalized operating parameter;

[0108] It should be understood that the diagnostic device can smooth the collected voltage, current and temperature data to eliminate the impact of abnormal values ​​or noise. Considering that the collected data has no obvious trend change, this embodiment can use a single exponential smoothing method to calculate the weighted average of the current actual value of the voltage, current and temperature data points in the specified window and the previous exponential smoothing value. The smoothing process includes:

[0109]

[0110] in, is the exponential smoothing value at time point t, is the original operating parameter at time point t, is the exponential smoothing value at time point t-1, is a smoothing constant, and the value range of the smoothing constant is [0, 1].

[0111] This embodiment collects historical operation data and fault sample data of the DC converter, wherein the historical operation data includes historical current data, historical voltage data, and historical temperature data, pre-processes the historical operation data and the fault sample data to obtain a data set, constructs a kernel function based on the data set, constructs an original model, and trains and optimizes the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model includes a support vector machine, collects real-time operation data of the DC converter, and pre-processes the real-time operation data, wherein the real-time operation data includes current current data, current voltage data, and current temperature data of the DC converter. , the preprocessed real-time operation data is input into the fault diagnosis model for fault diagnosis, and the fault probability and fault type of the DC converter are obtained; since this embodiment obtains a data set by preprocessing the historical operation data and the fault sample data, a kernel function is constructed based on the data set, and the original model is trained and optimized based on the kernel function, so as to construct an efficient and accurate fault diagnosis model, the real-time operation data of the DC converter is analyzed by the fault diagnosis model, and the operating status of the DC converter is monitored, so as to ensure the timeliness of fault diagnosis, effectively improve the accuracy of fault diagnosis, provide reliable diagnosis results for fault processing, and improve the response speed of fault processing.

[0112] refer to Figure 3 , Figure 3 FIG. 1 is a flow chart of a second embodiment of a DC converter fault diagnosis method according to the present invention.

[0113] Based on the above first embodiment, in this embodiment, the data set includes the fault type and operating parameters of the DC converter, and the operating parameters include current operating parameters, voltage operating parameters and temperature operating parameters;

[0114] The step S30 further includes:

[0115] Step S301: generating initialization particle parameters based on the data set.

[0116] It should be noted that the initialization particle parameters include: population size, position boundary, speed boundary and iteration number threshold, the population size is generated based on the number of fault types of the DC converter, the position boundary is generated based on the voltage threshold, current threshold and temperature threshold of the DC converter, and the speed boundary is generated based on the current change rate, voltage change rate and temperature change rate of the operating parameters.

[0117] In some embodiments, initialization particle parameters may include but are not limited to:

[0118] The population size (N), which represents the number of different failure modes or initial guess values;

[0119] Maximum number of iterations (T), which indicates the maximum number of cycles that the algorithm runs;

[0120] Inertia weight (ω), which controls the influence of the current velocity of the particle on the velocity of the next step;

[0121] The acceleration coefficients (c1) and (c2) represent the acceleration of the particle approaching its historical optimal position and the global optimal position, respectively;

[0122] Position boundary, which corresponds to the physical range of the converter's voltage, current and other parameters;

[0123] Speed ​​boundary, speed boundary corresponds to the rate at which the parameter changes.

[0124] It can be understood that the present embodiment can adopt the particle swarm algorithm to solve the optimal parameters, and the optimal parameters can be the optimal parameter combination [C, g] for constructing the kernel function, where C can be the model error tolerance parameter, and g corresponds to the kernel parameter in the kernel function, that is, the kernel parameter can be obtained based on g, thereby constructing the kernel function.

[0125] It should be noted that this embodiment can adopt an improved particle swarm algorithm, and the flight process of the particle is the search process of the individual. The flight speed of the particle can be dynamically adjusted according to the historical optimal position of the particle and the historical optimal position of the population (the speed of the particle represents the speed of movement, and the position represents the direction of movement); the speed and position of the particle are continuously updated until the global optimal solution that meets the termination condition is found.

[0126] Step S302: generating the initial position and initial velocity of the population particles according to the population quantity, the position boundary and the velocity boundary.

[0127] It can be understood that this embodiment calculates the initial position and initial velocity based on the following formula to initialize the particle population:

[0128]

[0129]

[0130] in, is the initial position of the particle at the initial time, is the initial velocity of the particle at the initial moment, is the minimum value of particle position, is the maximum particle position, is the minimum particle velocity, is the maximum particle velocity, is a random function.

[0131] Step S303: Determine the historical optimal position and the global optimal position according to the operating parameters.

[0132] In some embodiments, the diagnostic device may compare the fitness value of the particle's position in the current iteration with the fitness value of the particle's previously recorded historical best position (Personal Best Position, pBest). If the current fitness value is better than (or equal to) the fitness value of pBest, pBest is updated to the current position and the corresponding fitness value is saved. Otherwise, the original pBest is kept unchanged, thereby determining the historical best position.

[0133] In some embodiments, the diagnostic device may select the one with the best fitness value from the pBest of all particles as the new global best position (Global Best Position, gBest), compare the pBest fitness values ​​of all particles, select the best one, update gBest to the position of this best pBest, and save the corresponding fitness value, thereby determining the global optimal position.

[0134] Further, in order to accurately determine the historical optimal position and the global optimal position of the particle, the above step S303 may include:

[0135] Step S3031: Determine error information between actual operating parameters and theoretical operating parameters of the DC converter according to the operating parameters.

[0136] It should be noted that the actual operating parameters may be operating parameters of the DC converter during actual operation, such as actual voltage, actual current, actual temperature value, etc. The theoretical operating parameters may be theoretical operating parameters of the DC converter, such as theoretical voltage, theoretical current, theoretical temperature value, etc. The error information may be the error between the actual voltage, actual current or actual temperature value and the theoretical voltage, theoretical current or theoretical temperature value.

[0137] Step S3032: constructing a fitness function based on the error information.

[0138] It is understandable that the present embodiment can construct a fitness function based on the following formula:

[0139]

[0140] in, is the fitness value, are the actual operating parameters. are theoretical operating parameters.

[0141] Step S3033: Calculate the fitness value of each particle based on the fitness function, and determine the historical optimal position and the global optimal position based on the fitness value.

[0142] It can be understood that this embodiment can calculate the fitness value of each particle and record the historical optimal position and the global optimal position of each particle.

[0143] Step S304: updating the position and velocity of each particle based on the historical optimal position, the global optimal position, the initial position and the initial velocity, until the number of updates reaches the iteration number threshold, or the updated particle meets the preset global optimal convergence condition.

[0144] It is understandable that the present embodiment can update the speed and position of each particle based on the following formula:

[0145]

[0146]

[0147] in, and is a random number between [0, 1], The best position in history. is the global optimal position, is the inertia weight, and is the acceleration coefficient, For the The speed of the update, For the The location at the time of the update, For the The updated position, For the The speed after the update.

[0148] Furthermore, in order to expand the particle search range and increase the diversity of the particle swarm, thereby better completing the search for the global optimal solution, the above step S304 may include:

[0149] Step S3041: updating the speed of each particle based on the historical optimal position, the global optimal position and the initial speed.

[0150] It is understandable that in this embodiment, the velocity of each particle can be updated based on the following formula:

[0151]

[0152] in, and is a random number between [0, 1], The best position in history. is the global optimal position, is the inertia weight, and is the acceleration coefficient, For the The speed of the update, For the The location at the time of the update, For the The speed after the update.

[0153] In some embodiments, the diagnostic device can further improve the convergence speed and global search capability of the algorithm by adjusting the inertia weight. The inertia weight can be adjusted dynamically. , weight adjustment refers to the following formula:

[0154]

[0155] in, is the maximum value of the inertia weight, is the minimum value of the inertia weight, is the maximum number of iterations, Is the number of updates.

[0156] Step S3042: updating the position of each particle based on the initial position.

[0157] It should be noted that this embodiment can expand the search range of particles and increase the diversity of particle swarms by introducing Levy flight, and can also improve the search efficiency and particle vitality of particles in a fuzzy state, so that the local optimal solution can be jumped out during the solution, and the search for the global optimal solution can be better completed.

[0158] It can be understood that after the levy flight is introduced in this embodiment, the particle position update formula is as follows:

[0159]

[0160] in, After introducing levy flight for the i-th particle, The updated position, For the The location at the time of the update, is the step size control or scaling factor, is a random step length path;

[0161] The above random step length path is generated according to the following formula:

[0162]

[0163] and The normal distribution of refers to the following formula:

[0164]

[0165]

[0166] in, and is a random variable generated based on a normal distribution, is the Levy distribution parameter, is the current position of the particle, For interval The particle position, and Used to generate and The variance of .

[0167] Step S305: determining kernel parameters according to the updated particles, and constructing a kernel function based on the kernel parameters.

[0168] It can be understood that, in this embodiment, by repeating step S303 to step S304 until the number of updates reaches the iteration number threshold, or the updated particles meet the preset global optimal convergence condition, the optimal parameter combination [C, g] is output, and the kernel function is constructed based on the following formula:

[0169]

[0170] in, is the kernel function, is the kernel parameter, which corresponds to g in the optimal parameter combination.

[0171] It should be understood that this embodiment uses the optimal parameters [C, g] obtained by optimizing the improved particle swarm algorithm to train the original model. During the training process, the original model will find an optimal hyperplane to distinguish fault data of different fault types.

[0172] This embodiment generates initialized particle parameters based on the data set, generates initial positions and initial speeds of population particles according to the population size, the position boundary, and the speed boundary, determines a historical optimal position and a global optimal position according to the operating parameters, updates the position and speed of each particle based on the historical optimal position, the global optimal position, the initial position, and the initial speed until the number of updates reaches the iteration number threshold, or the updated particles meet the preset global optimal convergence condition, determines kernel parameters according to the updated particles, and constructs a kernel function based on the kernel parameters, thereby accurately obtaining the optimal parameters required for building a model and improving the prediction accuracy and generalization ability of the model.

[0173] refer to Figure 4 , Figure 4 FIG. 4 is a flow chart of a third embodiment of a DC converter fault diagnosis method according to the present invention.

[0174] Based on the above embodiments, in this embodiment, step S40 further includes:

[0175] Step S401: determining the operation characteristic information and the fault type of the DC converter according to the historical operation data and the fault sample data.

[0176] It should be noted that the operation characteristic information may include current operation characteristic information, voltage operation characteristic information, power operation characteristic information, temperature operation characteristic information, etc. of the DC converter. The above-mentioned fault type may include a fault mode of the DC converter.

[0177] Step S402: Determine model input characteristic parameters and model output target parameters based on the operation characteristic information and the failure mode.

[0178] It should be noted that the model input characteristic parameters include current parameters, voltage parameters and temperature parameters, and the model output target parameters include fault type and fault probability.

[0179] It is understandable that the diagnostic equipment can determine the input characteristics (such as current, voltage, power, etc.) and output targets (such as fault type, fault probability, etc.) of the model based on the operating characteristics and fault modes of the DC converter. Then, the model is trained and optimized using historical operating data and fault samples to improve the model's prediction accuracy and generalization ability.

[0180] Step S403: obtaining the DC converter fault judgment condition.

[0181] It should be noted that the above fault judgment conditions include:

[0182]

[0183]

[0184]

[0185] in, When the current voltage exceeds the upper threshold, The current voltage is lower than the lower threshold. When the current exceeds the upper threshold, The current temperature exceeds the upper threshold. The current voltage operating trend does not match the expected voltage operating trend. The current operating trend does not match the expected current operating trend. The current temperature operating trend does not match the expected temperature operating trend;

[0186] Step S404: constructing an original model according to the fault judgment condition, the model input characteristic parameters and the model output target parameters, and training and optimizing the original model based on the kernel function to obtain a fault diagnosis model.

[0187] It is understandable that the present embodiment may adopt a support vector machine (SVM) as the basic framework of the fault diagnosis model. SVM is a machine learning algorithm based on statistical learning theory, and has the advantages of strong generalization ability and low computational complexity.

[0188] In the specific implementation, when constructing the prediction model, the diagnostic equipment first determines the model's input features (such as current, voltage, temperature, etc.) and output targets (such as fault type, fault probability, etc.) based on the operating characteristics and fault modes of the DC converter. Then, the SVM model is trained and optimized using historical operating data and fault samples to improve the model's prediction accuracy and generalization ability.

[0189] In some embodiments, the diagnostic device may perform a performance evaluation on the fault diagnosis model to ensure the diagnostic accuracy of the model. The diagnostic device may use accuracy, recall, precision, and F1 score. ) Four points were used to evaluate the prediction model.

[0190] Among them, accuracy is one of the commonly used indicators in model evaluation, which is used to measure the accuracy of model prediction. The calculation formula is as follows:

[0191]

[0192] The recall rate measures the model's ability to identify positive samples, also known as sensitivity or recall rate. The calculation formula is as follows:

[0193]

[0194] Precision measures the accuracy of the model in predicting positive samples. The calculation formula is:

[0195]

[0196] The F1 score is an indicator that comprehensively considers precision and recall and is used to evaluate the performance of the model. It is the harmonic mean of precision and recall, and is calculated as follows:

[0197]

[0198] in, is the number of true positive samples (those predicted by the model to be positive and actually positive), is the number of true negative examples (those predicted by the model to be negative and actually negative), is the number of false positives (those predicted by the model to be positive but actually negative), is the number of false negatives (samples predicted by the model to be negative but actually positive).

[0199] Finally, the trained model is verified using the validation data to evaluate the model’s predictive ability, and the model is adjusted based on the evaluation results. The best state prediction model is selected based on the validation results.

[0200] In some embodiments, the diagnostic device inputs the real-time collected operating parameters into the trained fault diagnosis model to perform state prediction and fault diagnosis. The model calculates the predicted value of the output target based on the change trend and characteristics of the input parameters and compares it with the set threshold or standard. If the predicted value exceeds the normal range or meets specific conditions, it is determined that a fault has occurred.

[0201] In some embodiments, the diagnostic device uses the fault diagnosis model to predict the voltage, current and temperature data collected in real time to obtain the prediction result. , indicating the probability or confidence of failure. According to the preset threshold , to judge the reliability of the prediction results, if Greater than or equal to , the prediction result is considered reliable. The expression is as follows:

[0202]

[0203] In some embodiments, the fault judgment conditions may include:

[0204] Input voltage upper threshold: U in_th_high =K ref1 U in ;

[0205] Output voltage upper threshold: U out_th_high =K ref4 U out ;

[0206] When judging, it is collectively referred to as the voltage upper threshold U th_high ;

[0207] Input voltage lower threshold: U in_th_low = K ref2 U in ;

[0208] Output voltage lower threshold: U out_th_low = K ref5 U out ;

[0209] When judging, it is collectively referred to as the voltage lower threshold U th_low ;

[0210] Input current upper threshold: I in_th_high = K ref3 I in ;

[0211] Output current upper threshold: I out_th_high = K ref6 I out ;

[0212] When judging, it is collectively referred to as the upper current threshold I th_high ;

[0213] Converter temperature upper threshold: T c_th_high = K ref7 T c ;

[0214] Among them, K ref1 ~Kre7 They are threshold coefficients of voltage, current and temperature, respectively, and their values ​​are set according to actual conditions (for example, according to parameter attribute settings of a DC converter).

[0215] Then collect voltage and current in real time and set trends:

[0216] Current voltage: U in , U out , collectively referred to as U when judging;

[0217] Current: I in ,I out , collectively referred to as I when judging;

[0218] Current temperature: T c ;

[0219] Voltage change trend: U in_trend , U out_trend , collectively referred to as U trend ;

[0220] Current change trend: I in_trend ,I out_trend , collectively referred to as I when judging trend ;

[0221] Temperature change trend: T c_trend ;

[0222] Finally, determine whether there is a fault or abnormality:

[0223]

[0224]

[0225]

[0226] In some embodiments, after detecting a fault in the DC converter, the diagnostic device can automatically take appropriate fault handling measures according to the type and severity of the fault. For example, for a short circuit fault, the diagnostic device can immediately cut off the power supply to prevent damage from expanding; for an overload fault, normal operation can be restored by adjusting the load or increasing heat dissipation measures. At the same time, the diagnostic device can also generate detailed fault reports and maintenance recommendations to provide maintenance personnel with convenient fault location and repair guidance. The fault report should include detailed information such as the fault type, occurrence time, fault parameters, and recommended maintenance steps and precautions.

[0227] This embodiment determines the operating characteristic information and fault type of the DC converter according to the historical operating data and the fault sample data, determines the model input characteristic parameters and the model output target parameters based on the operating characteristic information and the fault mode, obtains the fault judgment condition of the DC converter, constructs the original model according to the fault judgment condition, the model input characteristic parameters and the model output target parameters, and trains and optimizes the original model based on the kernel function to obtain a fault diagnosis model, thereby effectively improving the generalization ability of fault diagnosis, ensuring the accuracy of fault diagnosis while improving the fault diagnosis speed, combining machine learning algorithms and predictive models for fault diagnosis, reducing the possibility of manual intervention and erroneous judgment, and improving the intelligence level and automation level of fault diagnosis.

[0228] In addition, the embodiment of the present invention also provides a DC converter fault diagnosis system, referring to Figure 5 , Figure 5 : is an architecture diagram of a DC converter fault diagnosis system. The DC converter fault diagnosis system includes a data preprocessing module, a prediction model module, a fault diagnosis module and a fault processing module. The data preprocessing module is used to collect DC converter operation data (including historical operation data and real-time operation data) through voltage sensors, current sensors and temperature sensors, and preprocess the operation data. The prediction model module is used to build, train and arrange models. The fault diagnosis module is used to input the preprocessed data into the model for fault diagnosis. The fault processing module is used to process the fault of the DC converter based on the fault diagnosis results and provide decision support.

[0229] The DC converter fault diagnosis system provided by the present application adopts the DC converter fault diagnosis method in the above embodiment, which can solve the technical problem of DC converter fault diagnosis. Compared with the prior art, the beneficial effects of the DC converter fault diagnosis system provided by the present application are the same as the beneficial effects of the DC converter fault diagnosis method provided by the above embodiment, and other technical features in the DC converter fault diagnosis system are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0230] In addition, an embodiment of the present invention further proposes a computer-readable storage medium, on which a DC converter fault diagnosis program is stored. When the DC converter fault diagnosis program is executed by a processor, the steps of the DC converter fault diagnosis method described above are implemented.

[0231] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0232] The computer-readable storage medium may be included in the DC converter fault diagnosis device; or may exist independently without being assembled into the DC converter fault diagnosis device.

[0233] In addition, an embodiment of the present invention further provides a computer program product, including a DC converter fault diagnosis program, which implements the steps of the DC converter fault diagnosis method described above when executed by a processor.

[0234] The specific implementation of the computer program product of the present invention is basically the same as the various embodiments of the DC converter fault diagnosis method described above, and will not be repeated here.

[0235] Reference Figure 6 , Figure 6 This is a structural block diagram of the first embodiment of the DC converter fault diagnosis device of the present invention.

[0236] like Figure 6 As shown, the DC converter fault diagnosis device proposed in the embodiment of the present invention includes:

[0237] A historical data collection module 10 is used to collect historical operation data and fault sample data of the DC converter, wherein the historical operation data includes historical current data, historical voltage data and historical temperature data;

[0238] A data processing module 20 is used to pre-process the historical operation data and the fault sample data to obtain a data set;

[0239] A kernel function construction module 30, used to construct a kernel function based on the data set;

[0240] A model building module 40 is used to build an original model, and train and optimize the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model includes a support vector machine;

[0241] A real-time data acquisition module 50, used to acquire real-time operation data of the DC converter and pre-process the real-time operation data, wherein the real-time operation data includes current current data, current voltage data and current temperature data of the DC converter;

[0242] The fault diagnosis module 60 is used to input the pre-processed real-time operation data into the fault diagnosis model for fault diagnosis, so as to obtain the fault probability and fault type of the DC converter.

[0243] Further, the data set includes fault types and operating parameters of the DC converter, and the operating parameters include current operating parameters, voltage operating parameters and temperature operating parameters; the kernel function construction module 30 is also used to generate initialization particle parameters based on the data set, and the initialization particle parameters include: population number, position boundary, speed boundary and iteration number threshold, the population number is generated based on the number of fault types of the DC converter, the position boundary is generated based on the voltage threshold, current threshold and temperature threshold of the DC converter, and the speed boundary is generated based on the current change rate, voltage change rate and temperature change rate of the operating parameters;

[0244] Generate the initial position and initial velocity of the population particles according to the population quantity, the position boundary and the velocity boundary:

[0245]

[0246]

[0247] in, is the initial position of the particle at the initial time, is the initial velocity of the particle at the initial moment, is the minimum value of particle position, is the maximum particle position, is the minimum particle velocity, is the maximum particle velocity, is a random function;

[0248] Determine the historical optimal position and the global optimal position according to the operating parameters;

[0249] The position and velocity of each particle are updated based on the historical optimal position, the global optimal position, the initial position and the initial velocity until the number of updates reaches the iteration number threshold, or the updated particle meets the preset global optimal convergence condition;

[0250] The kernel parameters are determined according to the updated particles, and the kernel function is constructed based on the kernel parameters:

[0251]

[0252] in, is the kernel function, is the kernel parameter.

[0253] Furthermore, the kernel function building module 30 is also used to determine error information between actual operating parameters and theoretical operating parameters of the DC converter according to the operating parameters;

[0254] Construct a fitness function based on the error information:

[0255]

[0256] in, is the fitness value, are the actual operating parameters. are theoretical operating parameters;

[0257] The fitness value of each particle is calculated based on the fitness function, and the historical optimal position and the global optimal position are determined based on the fitness value.

[0258] Furthermore, the kernel function construction module 30 is also used to update the speed of each particle based on the historical optimal position, the global optimal position and the initial speed:

[0259]

[0260] in, and is a random number between [0, 1], The best position in history. is the global optimal position, is the inertia weight, and is the acceleration coefficient, For the The speed of the update, For the The location at the time of the update, For the Speed ​​after update;

[0261] Update the position of each particle based on the initial position:

[0262]

[0263] in, For the The updated position, For the The location at the time of the update, is the step length control quantity, is a random step length path;

[0264] The above random step length path is generated according to the following formula:

[0265]

[0266] in, and is a random variable generated based on a normal distribution, is the Levy distribution parameter, is the current position of the particle, For interval The particle position.

[0267] Furthermore, the model building module 40 is also used to determine the operation characteristic information and the fault type of the DC converter according to the historical operation data and the fault sample data;

[0268] Determine a model input characteristic parameter and a model output target parameter based on the operation characteristic information and the fault mode, wherein the model input characteristic parameter includes a current parameter, a voltage parameter and a temperature parameter, and the model output target parameter includes a fault type and a fault probability;

[0269] Obtaining the DC converter fault judgment condition, wherein the fault judgment condition includes:

[0270]

[0271]

[0272]

[0273] in, When the current voltage exceeds the upper threshold, The current voltage is lower than the lower threshold. When the current exceeds the upper threshold, The current temperature exceeds the upper threshold. The current voltage operating trend does not match the expected voltage operating trend. The current operating trend does not match the expected current operating trend. The current temperature operating trend does not match the expected temperature operating trend;

[0274] An original model is constructed according to the fault judgment condition, the model input characteristic parameter and the model output target parameter, and the original model is trained and optimized based on the kernel function to obtain a fault diagnosis model.

[0275] Furthermore, the preprocessing includes normalization processing and smoothing processing;

[0276] The normalization process includes:

[0277]

[0278] in, are operating parameters in historical operating data and real-time operating data, wherein the operating parameters include current parameters, voltage parameters and temperature parameters, is the minimum value of the operating parameter, is the maximum value of the operating parameter, is the normalized operating parameter;

[0279] The smoothing process comprises:

[0280]

[0281] in, is the exponential smoothing value at time point t, is the original operating parameter at time point t, is the exponential smoothing value at time point t-1, is a smoothing constant, and the value range of the smoothing constant is [0, 1].

[0282] This embodiment collects historical operation data and fault sample data of the DC converter, wherein the historical operation data includes historical current data, historical voltage data, and historical temperature data, pre-processes the historical operation data and the fault sample data to obtain a data set, constructs a kernel function based on the data set, constructs an original model, and trains and optimizes the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model includes a support vector machine, collects real-time operation data of the DC converter, and pre-processes the real-time operation data, wherein the real-time operation data includes current current data, current voltage data, and current temperature data of the DC converter. , the preprocessed real-time operation data is input into the fault diagnosis model for fault diagnosis, and the fault probability and fault type of the DC converter are obtained; since this embodiment obtains a data set by preprocessing the historical operation data and the fault sample data, a kernel function is constructed based on the data set, and the original model is trained and optimized based on the kernel function, so as to construct an efficient and accurate fault diagnosis model, the real-time operation data of the DC converter is analyzed by the fault diagnosis model, and the operating status of the DC converter is monitored, so as to ensure the timeliness of fault diagnosis, effectively improve the accuracy of fault diagnosis, provide reliable diagnosis results for fault processing, and improve the response speed of fault processing.

[0283] The DC converter fault diagnosis device provided by the present application adopts the DC converter fault diagnosis method in the above embodiment, which can solve the technical problem of DC converter fault diagnosis. Compared with the prior art, the beneficial effects of the DC converter fault diagnosis device provided by the present application are the same as the beneficial effects of the DC converter fault diagnosis method provided by the above embodiment, and other technical features in the DC converter fault diagnosis device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0284] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present invention. In specific applications, technicians in this field can make settings as needed, and the present invention does not limit this.

[0285] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0286] In addition, for technical details not fully described in this embodiment, reference can be made to the DC converter fault diagnosis method provided in any embodiment of the present invention, and will not be repeated here.

[0287] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0288] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0289] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0290] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A DC converter fault diagnosis method, characterized in that: The DC converter fault diagnosis method comprises: Collecting historical operation data and fault sample data of the DC converter, wherein the historical operation data includes historical current data, historical voltage data and historical temperature data; Preprocessing the historical operation data and the fault sample data to obtain a data set; constructing a kernel function based on the data set; Constructing an original model, and training and optimizing the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model includes a support vector machine; Collecting real-time operation data of the DC converter and preprocessing the real-time operation data, wherein the real-time operation data includes current current data, current voltage data and current temperature data of the DC converter; Inputting the preprocessed real-time operation data into the fault diagnosis model for fault diagnosis to obtain the fault probability and fault type of the DC converter; The constructing of the original model, and training and optimizing the original model based on the kernel function to obtain a fault diagnosis model includes: Determine the operation characteristic information and the fault type of the DC converter according to the historical operation data and the fault sample data; Determine a model input characteristic parameter and a model output target parameter based on the operation characteristic information and the fault mode, wherein the model input characteristic parameter includes a current parameter, a voltage parameter and a temperature parameter, and the model output target parameter includes a fault type and a fault probability; Obtaining the DC converter fault judgment condition, wherein the fault judgment condition includes: in, When the current voltage exceeds the upper threshold, The current voltage is lower than the lower threshold. When the current exceeds the upper threshold, The current temperature exceeds the upper threshold. The current voltage operating trend does not match the expected voltage operating trend. The current operating trend does not match the expected current operating trend. The current temperature operating trend does not match the expected temperature operating trend; An original model is constructed according to the fault judgment condition, the model input characteristic parameter and the model output target parameter, and the original model is trained and optimized based on the kernel function to obtain a fault diagnosis model.

2. The DC converter fault diagnosis method according to claim 1, characterized in that: The data set includes fault types and operating parameters of the DC converter, and the operating parameters include current operating parameters, voltage operating parameters and temperature operating parameters; The step of constructing a kernel function based on the data set includes: Generate initialization particle parameters based on the data set, the initialization particle parameters include: population number, position boundary, speed boundary and iteration number threshold, the population number is generated based on the number of fault types of the DC converter, the position boundary is generated based on the voltage threshold, current threshold and temperature threshold of the DC converter, and the speed boundary is generated based on the current change rate, voltage change rate and temperature change rate of the operating parameter; Generate the initial position and initial velocity of the population particles according to the population quantity, the position boundary and the velocity boundary: in, is the initial position of the particle at the initial time, is the initial velocity of the particle at the initial moment, is the minimum value of particle position, is the maximum particle position, is the minimum particle velocity, is the maximum particle velocity, is a random function; Determine the historical optimal position and the global optimal position according to the operating parameters; The position and velocity of each particle are updated based on the historical optimal position, the global optimal position, the initial position and the initial velocity until the number of updates reaches the iteration number threshold, or the updated particle meets the preset global optimal convergence condition; The kernel parameters are determined according to the updated particles, and the kernel function is constructed based on the kernel parameters: in, is the kernel function, is the kernel parameter.

3. The DC converter fault diagnosis method according to claim 2, characterized in that: Determining the historical optimal position and the global optimal position according to the operating parameters includes: Determine error information between actual operating parameters and theoretical operating parameters of the DC converter according to the operating parameters; Construct a fitness function based on the error information: in, is the fitness value, are the actual operating parameters. are theoretical operating parameters; The fitness value of each particle is calculated based on the fitness function, and the historical optimal position and the global optimal position are determined based on the fitness value.

4. The DC converter fault diagnosis method according to claim 3, characterized in that: The updating of the position and velocity of each particle based on the historical optimal position, the global optimal position, the initial position and the initial velocity comprises: The speed of each particle is updated based on the historical optimal position, the global optimal position and the initial speed: in, and is a random number between [0, 1], The best position in history. is the global optimal position, is the inertia weight, and is the acceleration coefficient, For the The speed of the update, For the The location at the time of the update, For the Speed ​​after update; Update the position of each particle based on the initial position: in, For the The updated position, For the The location at the time of the update, is the step length control quantity, is a random step length path; The above random step length path is generated according to the following formula: in, and is a random variable generated based on a normal distribution, is the Levy distribution parameter, is the current position of the particle, For interval The particle position.

5. The DC converter fault diagnosis method according to any one of claims 1 to 4, characterized in that: The preprocessing includes normalization processing and smoothing processing; The normalization process includes: in, are operating parameters in historical operating data and real-time operating data, wherein the operating parameters include current parameters, voltage parameters and temperature parameters, is the minimum value of the operating parameter, is the maximum value of the operating parameter, is the normalized operating parameter; The smoothing process comprises: in, is the exponential smoothing value at time point t, is the original operating parameter at time point t, is the exponential smoothing value at time point t-1, is a smoothing constant, and the value range of the smoothing constant is [0, 1].

6. A DC converter fault diagnosis device, characterized in that: The DC converter fault diagnosis device comprises: A historical data collection module, used to collect historical operation data and fault sample data of the DC converter, wherein the historical operation data includes historical current data, historical voltage data and historical temperature data; A data processing module, used for preprocessing the historical operation data and the fault sample data to obtain a data set; A kernel function construction module, used for constructing a kernel function based on the data set; A model building module, used for building an original model, and training and optimizing the original model based on the kernel function to obtain a fault diagnosis model, wherein the original model includes a support vector machine; A real-time data acquisition module, used to acquire real-time operation data of the DC converter and pre-process the real-time operation data, wherein the real-time operation data includes current current data, current voltage data and current temperature data of the DC converter; A fault diagnosis module, used for inputting the pre-processed real-time operation data into the fault diagnosis model for fault diagnosis, and obtaining the fault probability and fault type of the DC converter; The model building module is further used to determine the operation characteristic information and fault type of the DC converter according to the historical operation data and the fault sample data; determine the model input characteristic parameters and the model output target parameters based on the operation characteristic information and the fault mode, the model input characteristic parameters include current parameters, voltage parameters and temperature parameters, and the model output target parameters include fault type and fault probability; obtain the fault judgment condition of the DC converter, the fault judgment condition includes: in, When the current voltage exceeds the upper threshold, The current voltage is lower than the lower threshold. When the current exceeds the upper threshold, The current temperature exceeds the upper threshold. The current voltage operating trend does not match the expected voltage operating trend. The current operating trend does not match the expected current operating trend. The current temperature operating trend does not match the expected temperature operating trend; An original model is constructed according to the fault judgment condition, the model input characteristic parameter and the model output target parameter, and the original model is trained and optimized based on the kernel function to obtain a fault diagnosis model.

7. A DC converter fault diagnosis device, characterized in that: The DC converter fault diagnosis device includes: a memory, a processor, and a DC converter fault diagnosis program stored in the memory and executable on the processor, wherein the DC converter fault diagnosis program is configured to implement the DC converter fault diagnosis method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a DC converter fault diagnosis program, and when the DC converter fault diagnosis program is executed by the processor, the DC converter fault diagnosis method according to any one of claims 1 to 5 is implemented.

9. A computer program product, characterized in that The computer program product comprises a DC converter fault diagnosis program, and when the DC converter fault diagnosis program is executed by a processor, the steps of the DC converter fault diagnosis method as claimed in any one of claims 1 to 5 are implemented.

Images