Fault diagnosis method and device for power generation access device and electronic equipment

Abstract

The invention provides a fault diagnosis method and device for a power generation access device and electronic equipment, and the method comprises the steps: obtaining the operation data of the power generation access device, and enabling the operation data to at least comprise voltage and current; the operation data is analyzed through a data analysis model, the classification probability is obtained, the data analysis model is obtained through multiple sets of data according to machine learning training, and each set of data in the multiple sets of data comprises historical operation data and data containing data state labels; the data state label comprises a normal state label and a fault state label, and the classification probability represents the probability that the operation data is fault data; and under the condition that the classification probability is greater than a preset threshold value, determining that the operation data is fault data, and determining that the power generation access device has a fault. The problem of lack of intelligent fault prediction and diagnosis of the power generation access device in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a fault diagnosis method, device, computer-readable storage medium and electronic device for a power generation access device. Background Art

[0002] With the rapid development of renewable energy power generation in my country, the monitoring system of power generation access devices will play an increasingly important role in ensuring the safety of the power grid and improving energy utilization efficiency. In the future, the monitoring system will develop in the direction of intelligence, networking and integration. The monitoring system of power generation access devices plays an important role in the power industry. It helps to improve the operating efficiency of the power generation system, reduce operation and maintenance costs, and ensure the safety of the power grid. With the continuous advancement of technology, the monitoring system of power generation access devices will create more value for power companies and help the sustainable development of my country's power industry. With the continuous growth of energy demand and the increasing complexity of the power system, ensuring the stable operation and efficient maintenance of the power generation system has become an important issue in the power industry. Traditional monitoring methods for power generation access devices mainly rely on manual inspections and simple sensor technology.

[0003] However, there are the following limitations: manual inspections are inefficient, making it difficult to achieve real-time monitoring of power generation access devices, and are easily affected by subjective factors, resulting in missed detection or misjudgment. Simple sensor technology can only collect basic operating data, lacks in-depth analysis and processing of data, and is difficult to accurately diagnose equipment failures. Existing monitoring systems lack intelligent fault prediction and diagnosis functions, and often can only detect problems when serious equipment failures occur, resulting in increased maintenance costs and power supply interruptions. Summary of the invention

[0004] The main purpose of the present application is to provide a fault diagnosis method, device, computer-readable storage medium and electronic device for a power generation access device, so as to at least solve the problem of lack of intelligent fault prediction and diagnosis of power generation access devices in the prior art.

[0005] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a fault diagnosis method for a power generation access device is provided, comprising: obtaining operating data of the power generation access device, wherein the operating data at least includes voltage and current; analyzing the operating data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through machine learning training of multiple groups of data, each group of the multiple groups of data includes: historical operating data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operating data is fault data; when the classification probability is greater than a preset threshold, it is determined that the operating data is the fault data, and it is determined that the power generation access device has a fault.

[0006] Optionally, after acquiring the operating data of the power generation access device, the fault diagnosis method of the power generation access device also includes: removing outliers and noise in the operating data by a sliding window method to obtain denoised operating data; normalizing the denoised operating data to obtain normalized operating data; analyzing the normalized operating data by a principal component analysis method to obtain principal component characteristics of the normalized operating data.

[0007] Optionally, before analyzing the operating data through the data analysis model, the fault diagnosis method of the power generation access device also includes: describing the time series characteristics of the operating data through an autoregressive integrated moving average model, and determining the model parameters, and constructing the autoregressive integrated moving average model as an initial data analysis model; using a support vector machine algorithm to train the initial data analysis model through the historical operating data and the data containing the data status label to obtain the data analysis model.

[0008] Optionally, a support vector machine algorithm is used to train the initial data analysis model using the historical operation data and the data containing the data status label, including: extracting features of the historical operation data through the autoregressive integrated moving average model, using the features as input variables of the support vector machine algorithm, and using the data containing the data status label as output variables of the support vector machine algorithm; optimizing the parameters of the support vector machine algorithm through grid search and cross-validation methods, using a sequential minimum optimization algorithm to solve the objective function of the support vector machine algorithm, and obtaining a hyperplane with a maximum interval to complete the training of the initial data analysis model.

[0009] Optionally, after determining that the power generation access device fails, the fault diagnosis method of the power generation access device further includes: generating alarm information and sending the alarm information to the client, wherein the alarm information is used to notify the staff that the power generation access device is in a faulty state.

[0010] Optionally, after determining that the operating data is the fault data, the fault diagnosis method of the power generation access device also includes: classifying the fault data into types, and using cluster analysis to perform pattern recognition on the fault data to determine the cause and location of the fault; generating a fault diagnosis report based on the cause and location of the fault to guide staff to perform fault repairs.

[0011] Optionally, obtaining operation data of the power generation access device includes: obtaining the operation data of the power generation access device through a digital sensor, wherein the digital sensor is electrically connected to the power generation access device.

[0012] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a fault diagnosis device for a power generation access device is provided, comprising: an acquisition unit, used to acquire operating data of the power generation access device, wherein the operating data at least includes voltage and current; an analysis unit, used to analyze the operating data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through machine learning training of multiple groups of data, and each group of data in the multiple groups of data includes: historical operating data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operating data is fault data; a determination unit, used to determine that the operating data is the fault data and determine that a fault occurs in the power generation access device when the classification probability is greater than a preset threshold.

[0013] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned fault diagnosis methods for power generation access devices.

[0014] According to another aspect of the present application, an electronic device is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing a fault diagnosis method of a power generation access device described in any one of the above.

[0015] Applying the technical solution of the present application, first, obtain the operating data of the power generation access device, wherein the operating data includes at least voltage and current; analyze the operating data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through multiple groups of data according to machine learning training, and each group of data in the multiple groups of data includes: historical operating data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operating data is fault data; when the classification probability is greater than a preset threshold, determine that the operating data is fault data, and determine that the power generation access device has a fault. The present application uses machine learning and other functional algorithms to perform fault diagnosis and detection on the power generation access device, avoiding the problem of waste of manpower and material resources caused by manual inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings constituting part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0017] Figure 1 It is a hardware structure block diagram of a mobile terminal of a fault diagnosis method for a power generation access device according to an embodiment of the present invention;

[0018] Figure 2 is a flow chart of a fault diagnosis method for a power generation access device according to an embodiment of the present application;

[0019] Figure 3 is a schematic diagram of an SVM regression algorithm according to an embodiment of the present application;

[0020] Figure 4 It is a schematic diagram of a fault diagnosis device for a power generation access device according to an embodiment of the present application.

[0021] The above drawings include the following reference numerals:

[0022] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0023] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] As introduced in the background technology, the prior art lacks intelligent fault prediction and diagnosis of power generation access devices. To solve the above problems, the embodiments of the present application provide a fault diagnosis method, device, computer-readable storage medium and electronic device for a power generation access device.

[0027] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0028] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal for a fault diagnosis method of a power generation access device according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0029] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the fault diagnosis method of the power generation access device in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0030] In this embodiment, a fault diagnosis method for a power generation access device running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0031] Figure 2 FIG. 1 is a flow chart of a fault diagnosis method for a power generation access device according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0032] Step S201, obtaining operation data of the power generation access device, wherein the operation data at least includes voltage and current.

[0033] Specifically, by acquiring the operating data of the power generation access device, the operating status of the power generation access device can be monitored in real time, abnormal situations can be discovered in time, and the timeliness and accuracy of fault diagnosis can be improved. By monitoring and analyzing the operating data, it can be determined whether the power generation access device is operating normally and whether there are abnormal situations, thereby providing early warning of possible faults, discovering the wear and aging of the equipment in time, and formulating reasonable maintenance plans to ensure the safe and stable operation of the equipment.

[0034] Step S202: Analyze the operating data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through machine learning training of multiple groups of data, and each group of the multiple groups of data includes: historical operating data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operating data is fault data.

[0035] Specifically, by analyzing historical operating data through data analysis models, the probability that the current operating data is fault data can be obtained, equipment failures can be discovered in a timely manner, and corresponding maintenance measures can be taken, thereby reducing downtime and maintenance costs and improving equipment reliability and stability. The data analysis model obtained through machine learning training can more accurately identify fault conditions, improve the accuracy and efficiency of fault diagnosis, and make fault diagnosis more intelligent and automated.

[0036] Step S203: when the classification probability is greater than a preset threshold, it is determined that the operation data is fault data, and it is determined that a fault occurs in the power generation access device.

[0037] Specifically, by judging whether the data is fault data based on the classification probability, and determining that the power generation access device has a fault when the probability is greater than a preset threshold, the fault of the power generation access device can be identified quickly and accurately, and repair measures can be taken in time to avoid the expansion of the fault and cause more serious problems, thereby improving the reliability and stability of the equipment and ensuring the normal operation of the power generation system.

[0038] Through this embodiment, the operation data of the power generation access device can be obtained, wherein the operation data includes at least voltage and current; the operation data is analyzed by a data analysis model to obtain a classification probability, wherein the data analysis model is obtained by training multiple groups of data according to machine learning, and each group of data in the multiple groups of data includes: historical operation data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operation data is fault data; when the classification probability is greater than a preset threshold, the operation data is determined to be fault data, and it is determined that the power generation access device has a fault. Fault diagnosis and detection of power generation access devices are performed through functional algorithms such as machine learning to avoid the problem of waste of manpower and material resources caused by manual inspections.

[0039] Optionally, the operating data of the above-mentioned power generation access device may include but is not limited to: output voltage, current, power factor, frequency, winding temperature, bearing temperature, vibration frequency, amplitude and oil contamination data of the motor group.

[0040] Specifically, after obtaining the operating data of the power generation access device, the data model is analyzed using an intelligent algorithm based on machine learning to achieve real-time monitoring and abnormal diagnosis of the power generation access device. According to the operating characteristics and actual needs of the power generation access device, a reasonable alarm threshold (i.e., a preset threshold) is set so that an alarm can be issued in time when an abnormality occurs in the equipment. The real-time data is monitored through an intelligent algorithm. When it is detected that the key parameters exceed the alarm threshold, an alarm signal is immediately issued to notify the operation and maintenance personnel.

[0041] In the specific implementation process, after obtaining the operating data of the power generation access device, the above method also includes the following steps: removing outliers and noise in the operating data by a sliding window method to obtain denoised operating data; normalizing the denoised operating data to obtain normalized operating data; analyzing the normalized operating data by a principal component analysis method to obtain the principal component characteristics of the normalized operating data.

[0042] In this scheme, the sliding window method is used to remove outliers and noise in the operating data, which can improve the accuracy and reliability of the data, reduce the possibility of misdiagnosis, and thus more accurately diagnose the fault of the device. Normalizing the denoised operating data can eliminate the dimensional differences between different variables, making the data easier to compare and analyze. By analyzing the normalized operating data through the principal component analysis method, the main change direction in the data can be found and the main features can be identified, so as to better understand the structure and characteristics of the data and provide stronger support and basis for fault diagnosis.

[0043] Specifically, the collected data are preprocessed, including data cleaning, denoising, normalization, and principal component analysis dimensionality reduction, to improve data quality and provide a reliable data source for subsequent analysis. First, the sliding window method is used to remove obvious outliers and noise. Secondly, the Z-Score standardization method is used to normalize the data to make the data distribution more even. Finally, the principal component analysis is used to reduce the data dimensionality and extract the most representative features.

[0044] The specific steps to remove outliers are as follows: First, let the original data set be X. For each data point xi, if its deviation from the mean μ exceeds k times the standard deviation σ, it is regarded as an outlier and removed. The calculation formula is as follows: |xi-μ|>kσ, where μ is the mean of the data set X, σ is the standard deviation of the data set X, and k is the set threshold coefficient. The value of k is 2 or 3.

[0045] The specific steps of data normalization are as follows: standardize the data to a distribution with a mean of 0 and a standard deviation of 1. The calculation formula is as follows: Among them, zi is the standardized data point, xi is the original data point, μ is the mean of the original data set, and σ is the standard deviation of the original data set.

[0046] The specific steps of principal component analysis dimensionality reduction are as follows: let the original feature matrix be X, and use PCA to find a set of new eigenvectors Z to maximize the variance of the data on the new features. The calculation formula is as follows: Z = XV, where Z is the feature matrix after dimensionality reduction, X is the original feature matrix, V is the eigenvector matrix, and the principal components calculated by the PCA algorithm.

[0047] Specifically, the above-mentioned Z-Score standardization method (Z-Score Standardization) is a commonly used data standardization method, also known as the standard deviation standardization method. This method obtains standardized data by subtracting the mean of the original data and then dividing it by its standard deviation. The standardized data has the characteristics of a mean of 0 and a standard deviation of 1, so that different data are comparable.

[0048] Specifically, the above-mentioned principal component analysis (PCA) is a common dimensionality reduction technique, which can convert a high-dimensional data set into a low-dimensional feature subspace while retaining the maximum variance of the data set. The basic idea is to project the original data into a new coordinate system through linear transformation so that the projected data has the maximum variance. In the new coordinate system, the dimension of the data will be reduced, thereby achieving data dimensionality reduction.

[0049] In the specific implementation process, before analyzing the operating data through the data analysis model, the above method also includes the following steps: describing the time series characteristics of the operating data through the autoregressive integrated moving average model, and determining the model parameters, and constructing the autoregressive integrated moving average model as the initial data analysis model; using the support vector machine algorithm to train the initial data analysis model through historical operating data and data containing data status labels to obtain the data analysis model.

[0050] In this scheme, the time series characteristics of the operating data can be described through the autoregressive integrated moving average model, and the model parameters can be determined to build an initial data analysis model, identify trends and periodicities in the data, and thus better understand the regularity of the data. The initial data analysis model can be trained by the support vector machine algorithm to better identify patterns and features in the data, thereby building a more accurate data analysis model. The support vector machine algorithm can effectively process high-dimensional data and improve the accuracy and generalization ability of the model. This combined method can better analyze historical operating data, quickly and accurately identify faults and predict possible faults, detect problems early, take timely measures to repair and repair, improve the reliability and safety of equipment, and ensure stable operation of equipment.

[0051] Specifically, the above-mentioned autoregressive integrated moving average model (ARIMA) is a classic time series prediction model, which is often used to predict and model time series data. It combines the characteristics of the autoregressive (AR) model, the difference (I) model and the moving average (MA) model, and can handle non-stationary series and seasonal changes.

[0052] Specifically, the above-mentioned support vector machine (SVM) algorithm is a two-classification model. Its basic idea is to find an optimal hyperplane to separate samples of different categories. The goal of SVM is to find a maximum margin hyperplane so that the distance between the sample points of the two categories and the hyperplane is maximized. Support vector machines perform well in processing high-dimensional data and nonlinear data. By introducing kernel functions, low-dimensional nonlinear separable data can be mapped to high-dimensional space, thereby achieving linear separability.

[0053] Specifically, based on the preprocessed data, a data model is constructed, that is, a multiple combination analysis model of source and storage elements is constructed to achieve a comprehensive evaluation of the operating status of the power generation access device. The autoregressive integrated moving average model is selected to describe the time series characteristics of the data. In the model construction process, the parameters of the model (p, d, q) are determined by ACF and PACF diagrams, and the optimal model is selected by the Bayesian information criterion to ensure that the model can accurately capture the dynamic change trend of the data.

[0054] The formula and parameters for building the ARIMA model are as follows:

[0055] 1. Autoregression (AR) part: Assume that the time series is {X(t)}, and the p-order autoregression model is expressed as the following formula: X(t) = c + ∑Φi*X(ti) + ε(t), where c is the constant term, Φi is the autoregression coefficient, i is the lag order (i = 1, 2, ..., p), and ε(t) is the white noise sequence.

[0056] 2. Moving average (MA) part: The q-order moving average model is expressed as the following formula: X(t) = c + ε(t) + ∑θiε(ti), where, c is a constant term, θi is the moving average coefficient, i is the lag order (i = 1, 2, ..., q), and ε(t) is a white noise sequence.

[0057] 3. Difference (I) part: The d-order difference is used to make the time series smooth. The specific formula is as follows: ΔdX(t)=X(t)-X(td), where Δd represents the d-order difference operation, X(t) is the sequence value at the current time point, and X(td) is the sequence value d time units ago.

[0058] Figure 3 is a schematic diagram of the SVM regression algorithm according to an embodiment of the present application, such as Figure 3 As shown, the model training steps are as follows:

[0059] 1. Data preparation: The preprocessed data is divided into a feature set (X) and a label set (y). The feature set X contains the collected data of all sensors, while the label set y represents the normal or abnormal state of the corresponding data point.

[0060] 2. Data partitioning: Divide the dataset into training set and test set, with a ratio of 70%-30% or 80%-20%.

[0061] 3. Feature selection: Select the most influential features for classification from the feature set, using a feature selection algorithm, such as recursive feature elimination.

[0062] 4. Model training: Use the training set to train the SVM model. The goal of SVM is to find an optimal hyperplane that maximizes the interval between data points of different categories. Suppose the training set is {x_i, y_i}, i = 1,..., n, where x_i is the feature vector and y_i is the class label (-1 or 1). The objective function calculation formula of SVM is: The constraints are: subjectto:y_i*(w·x_i+b)≥1,foralli, where ||w|| 2 Represents the square of the Euclidean norm of vector w, w is the normal vector of the hyperplane, b is the bias term of the hyperplane, x_i is the i-th eigenvector, y_i is the class label of the i-th eigenvector, which takes a value of -1 or 1, and i traverses all sample points.

[0063] 5. Kernel function selection: If the data is not linearly separable, use the kernel function to map the data to a high-dimensional space to make it linearly separable. Common kernel functions include linear kernel, polynomial kernel and radial basis kernel function. The kernel function formula is: Among them, K is the kernel function, is the mapping function, x and x' are the data points.

[0064] 6. Parameter tuning: Use the cross-validation method to select the optimal penalty parameter C, control the weight of misclassification and kernel function parameters, such as σ for the radial basis kernel function.

[0065] 7. Model evaluation: Use the test set to evaluate the performance of the model. Common evaluation indicators include accuracy, precision, recall and F1 score.

[0066] 8. Abnormal diagnosis: Input the real-time collected data into the trained SVM model. If the model predicts an abnormality, the abnormal diagnosis process is triggered.

[0067] 9. Result output: Output abnormal diagnosis results, including abnormality type, possible causes and recommended maintenance measures.

[0068] Specifically, the above-mentioned ACF (Auto-Correlation Function) graph is a method for analyzing time series data. It is mainly used to detect whether there is autocorrelation in time series data. It shows the correlation between time series data and its lagged value. In the ACF graph, the horizontal axis represents the lag value (lag) and the vertical axis represents the correlation value (correlation). By observing the ACF graph, it can be determined whether the time series data has periodicity, trend or other characteristics.

[0069] Specifically, the above-mentioned PACF (Partial Autocorrelation Function) diagram is a graphical tool for examining the autocorrelation of time series data, which displays the autocorrelation coefficient of each lag order, where each coefficient represents the correlation between the current lag order and the current value after taking into account the previous lag effect. It is usually used to determine seasonal and trend patterns in time series data. In the PACF diagram, if the autocorrelation coefficient of a lag order is significantly greater than zero, it means that there is a significant correlation between the lag order and the current value, and there may be a seasonal or trend pattern.

[0070] Specifically, the above-mentioned Bayesian Information Criterion (BIC) is a model selection criterion used to compare the goodness of fit of different models, taking into account the goodness of fit of the model and the complexity of the model, and avoiding overfitting through a penalty term including the number of model parameters.

[0071] Specifically, the above-mentioned recursive feature elimination (RFE) is a method of recursively training a model and eliminating features that contribute less to model performance. It is usually used for feature selection, which can help reduce the dimension of the feature space and improve the generalization ability and prediction performance of the model.

[0072] Specifically, the above-mentioned linear kernel function is one of the simplest kernel functions, which can map data to a higher-dimensional space, so that the originally linearly inseparable data becomes linearly separable in the new space; the above-mentioned polynomial kernel function is a kernel function of kernel techniques used in support vector machines (SVM) and other machine learning algorithms, which maps input samples to a high-dimensional space, so that nonlinear problems can become linearly separable in a higher-dimensional space; the above-mentioned radial basis (Gaussian Radial Basis Function, RBF) kernel function is a commonly used kernel function, used in support vector machines and other machine learning algorithms, which can map input data to a high-dimensional feature space, so that the original nonlinear separable problem becomes linearly separable in this high-dimensional space.

[0073] Specifically, the above cross-validation method is a statistical technique for evaluating model performance and generalization ability by dividing the dataset into a training set and a test set and repeating this process multiple times to evaluate the stability and accuracy of the model.

[0074] Optionally, the cross-validation method may include but is not limited to: k-fold cross-validation, leave-one-out cross-validation, random cross-validation and other methods.

[0075] In some embodiments, a support vector machine algorithm is used to train the initial data analysis model using historical operating data and data containing data status labels. This can be specifically achieved through the following steps: extracting features of the historical operating data through an autoregressive integral sliding average model, using the features as input variables of the support vector machine algorithm, and using the data containing data status labels as output variables of the support vector machine algorithm; optimizing the parameters of the support vector machine algorithm through grid search and cross-validation methods, using a sequential minimum optimization algorithm to solve the objective function of the support vector machine algorithm, and obtaining a hyperplane with the maximum interval to complete the training of the initial data analysis model.

[0076] In this solution, historical operating data is trained through the support vector machine algorithm to identify patterns and regularities in the data, thereby better understanding the operating status and potential problems of the equipment. By optimizing the algorithm parameters and solving the objective function of the support vector machine algorithm, the generalization ability and accuracy of the model can be improved, so that equipment failures can be diagnosed and predicted more accurately, and equipment problems can be discovered in time and corresponding maintenance measures can be taken to reduce the impact of equipment failures on production and operations and improve the reliability and stability of the equipment.

[0077] Specifically, the support vector machine (SVM) algorithm is used to train the data model. First, the features extracted by the ARIMA model are used as input variables, and the equipment status label is used as the output variable. Then, the parameters of the SVM are optimized through grid search and cross-validation methods. During the training process, the sequential minimum optimization algorithm is used to solve the objective function and find the hyperplane with the maximum interval, thereby achieving effective classification of normal and abnormal states.

[0078] During the specific implementation process, after determining that the power generation access device has failed, the above method also includes the following steps: generating an alarm message and sending the alarm message to the client, wherein the alarm message is used to notify the staff that the power generation access device is in a faulty state.

[0079] In this solution, by generating and sending alarm information, it is helpful to timely discover the fault status of the power generation access device, so that repair measures can be taken in time, reducing power outage time and production losses, improving the fault diagnosis efficiency of the power generation access device, reducing fault handling time, ensuring the safety of staff and the normal operation of equipment, and timely repair of faults can also extend the service life of the equipment and reduce maintenance costs.

[0080] Specifically, the trained SVM model is applied to real-time data stream analysis. The model continuously receives data from sensors and calculates the classification probability of each data point in real time. When the classification probability of a data point is lower than the set threshold, the system will immediately trigger the alarm mechanism. The alarm system uses sound and light alarms, SMS notifications, and emails to ensure that operation and maintenance personnel can receive alarm information in a timely manner and take corresponding measures.

[0081] In the specific implementation process, after determining that the operating data is fault data, the above method also includes the following steps: classifying the fault data, and using cluster analysis to perform pattern recognition on the fault data to determine the cause and location of the fault; generating a fault diagnosis report based on the cause and location of the fault to guide the staff to perform fault repair.

[0082] In this solution, fault data is classified and pattern recognition is performed using cluster analysis, which can help staff quickly locate the cause and location of the fault, improve the efficiency and accuracy of fault diagnosis, and generate a fault diagnosis report to clearly understand the cause and location of the fault, so as to carry out targeted maintenance, reduce maintenance time and costs, and improve equipment reliability and operating efficiency.

[0083] Specifically, we conduct in-depth analysis of abnormal data, diagnose the causes of failures in power generation access devices, provide maintenance suggestions for operation and maintenance personnel, optimize and adjust the power generation access devices based on the monitoring and diagnosis results, improve the reliability and efficiency of the power generation system, and at the same time, collect feedback from operation and maintenance personnel to continuously optimize monitoring and diagnosis methods.

[0084] Specifically, by comparing historical data and combining the classification results of the SVM model, decision trees and logistic regression algorithms are used to segment fault types, and cluster analysis is used to perform pattern recognition on fault data to determine the possible cause and location of the fault. Finally, a detailed fault diagnosis report is generated to provide accurate maintenance guidance for operation and maintenance personnel.

[0085] In addition, by analyzing the monitoring data, the weak links in equipment operation can be identified and targeted improvement measures can be proposed. At the same time, a feedback mechanism is established to regularly collect usage feedback from operation and maintenance personnel to evaluate the performance of the monitoring and diagnosis system, and continuously adjust and optimize the algorithm model to improve the system's monitoring accuracy and fault prevention.

[0086] In some embodiments, obtaining the operating data of the power generation access device can be specifically achieved through the following steps: obtaining the operating data of the power generation access device through a digital sensor, wherein the digital sensor is electrically connected to the power generation access device.

[0087] In this solution, the operating data of the power generation access device is obtained through digital sensors, which can reflect the operating status of the equipment, help monitor the operating status of the equipment, detect any abnormal conditions, and promptly discover possible equipment failures. By analyzing the operating data, maintenance and repairs can be carried out in a timely manner to ensure the normal operation of the equipment and extend the life of the equipment.

[0088] Specifically, the steps for obtaining the operating data of the power generation access device are as follows:

[0089] 1. Determine the monitoring objectives: clarify the key operating parameters of the power generation access device, such as voltage, current, frequency, temperature and vibration, to provide a basis for the selection of subsequent monitoring hardware and data collection.

[0090] 2. Design detachable monitoring hardware: According to the monitoring objectives, design digital sensors with high precision and high reliability to ensure that the sensors can be connected on a large scale and are detachable for easy maintenance and upgrades.

[0091] 3. Deploy monitoring hardware: Install the designed digital sensors at key locations of the power generation access device to ensure that the sensors can accurately collect the required operating parameters.

[0092] 4. Data collection and transmission: Use digital sensors to collect the operating data of the power generation access device in real time, and send the data to the central processing system through the wireless transmission module.

[0093] Specifically, in order to ensure the comprehensiveness and accuracy of monitoring, the key thresholds and variation ranges of the above-mentioned key operating parameters are determined through in-depth research on the working principles and failure modes of the power generation access devices. The thresholds and ranges will serve as an important reference for the design and implementation of the monitoring system, so that signals can be captured in time when abnormalities occur in the equipment.

[0094] Optionally, the monitoring hardware may include but is not limited to: a high-precision current transformer, a voltage sensor, a thermocouple, an acceleration sensor, an oil sensor, etc.

[0095] Specifically, each sensor adopts a modular design with a standard interface and quick connection mechanism to facilitate quick installation and removal at different locations. At the same time, the sensor housing is made of high-temperature resistant and corrosion-resistant materials to ensure stable operation even in harsh environments. In addition, a self-calibration function is designed for the sensor to reduce measurement errors caused by long-term operation.

[0096] Specifically, during the deployment process, special installation tools and fixtures are used to ensure stable contact between sensors and monitored components. Anti-interference and anti-pull cables are used to transmit data to the central data acquisition system through wired or wireless networks. During the data acquisition process, a high-frequency sampling rate is used to ensure that slight changes in equipment operation can be captured. The TCP / IP (Transmission Control Protocol / Internet Protocol) protocol is used for data transmission to ensure data integrity and transmission stability. At the same time, in order to prevent data from being intercepted or tampered with during transmission, the AES encryption algorithm is used to encrypt the data.

[0097] The technical solution provided by the above embodiments of the present application solves the problem of lack of intelligent fault prediction and diagnosis of power generation access devices in the prior art, and has the following beneficial effects:

[0098] 1. Data preprocessing First, by removing outliers, the purity and reliability of the data set are ensured, and the interference of abnormal data on subsequent analysis and model training is avoided, thereby improving the accuracy and robustness of the monitoring system. Secondly, data normalization enables data of different dimensions and scales to be compared and analyzed on the same level, which not only helps the algorithm to better understand and process data, but also accelerates the convergence of the model. Finally, through principal component analysis (PCA) dimensionality reduction, the dimension of the data is effectively reduced, the noise and redundant information in the data are removed, the computational complexity is reduced, and the generalization ability of the model is improved, so that the monitoring system can more efficiently and accurately identify the operating status of the power generation access device.

[0099] 2. The constructed ARIMA model can capture linear relationships and nonlinear trends in time series data by combining autoregression, moving average and difference techniques, which is crucial for predicting and monitoring the future status of power generation access devices. The model can identify periodic patterns and potential abnormal fluctuations in the data, thereby providing early warning of possible equipment failures. In addition, by accurately determining the model parameters p, d, and q, it is possible to construct a model that best suits the current data characteristics, thereby improving the accuracy of the prediction, reducing the risk of overfitting, and ensuring the stability and reliability of the monitoring system in practical applications.

[0100] 3. The support vector machine intelligent algorithm has brought revolutionary improvements to the entire monitoring and diagnosis method. The SVM algorithm separates different categories of data by finding the optimal hyperplane. It has strong nonlinear processing and generalization capabilities, and can effectively identify normal and abnormal states from complex monitoring data. It significantly improves the fault diagnosis accuracy of the monitoring system and reduces false alarms and missed alarms. At the same time, the SVM algorithm can process high-dimensional data, which is particularly important for the monitoring of power generation access devices that contain a large amount of sensor data. It realizes real-time monitoring of power generation access devices and can perform early diagnosis and prediction of potential faults, thereby greatly improving the operating efficiency and safety of the power generation system.

[0101] As can be seen from the above, in the embodiment of the present application, the constructed ARIMA model can capture the linear relationship and nonlinear trend in the time series data by combining autoregression, moving average and difference technology. The model can identify the periodic patterns and potential abnormal fluctuations in the data, thereby early warning of possible equipment failures. In addition, by accurately determining the model parameters p, d, q, it is possible to construct a model that best suits the current data characteristics, which not only improves the accuracy of the prediction, but also reduces the risk of overfitting, ensuring the stability and reliability of the monitoring system in practical applications. The support vector machine intelligent algorithm separates different categories of data by finding the optimal hyperplane. Its powerful nonlinear processing ability and generalization ability enable it to effectively identify normal and abnormal states from complex monitoring data, greatly improving the operating efficiency and safety of the power generation system.

[0102] The embodiment of the present application also provides a fault diagnosis device for a power generation access device. It should be noted that the fault diagnosis device for a power generation access device in the embodiment of the present application can be used to execute the fault diagnosis method for a power generation access device provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0103] The following is an introduction to the fault diagnosis device of the power generation access device provided in the embodiment of the present application.

[0104] Figure 4 Schematic diagram of a fault diagnosis device for a power generation access device according to an embodiment of the present application. Figure 4 As shown, the device includes: an acquisition unit 41, an analysis unit 42 and a determination unit 43.

[0105] The acquisition unit 41 is used to acquire the operating data of the power generation access device, wherein the operating data at least includes voltage and current.

[0106] The analysis unit 42 is used to analyze the operation data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through multiple groups of data according to machine learning training, and each group of the multiple groups of data includes: historical operation data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operation data is fault data.

[0107] The determination unit 43 is used to determine that the operating data is fault data and that a fault occurs in the power generation access device when the classification probability is greater than a preset threshold.

[0108] Through this embodiment, the operation data of the power generation access device can be obtained, wherein the operation data includes at least voltage and current; the operation data is analyzed by a data analysis model to obtain a classification probability, wherein the data analysis model is obtained by training multiple groups of data according to machine learning, and each group of data in the multiple groups of data includes: historical operation data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operation data is fault data; when the classification probability is greater than a preset threshold, the operation data is determined to be fault data, and it is determined that the power generation access device has a fault. Fault diagnosis and detection of power generation access devices are performed through functional algorithms such as machine learning to avoid the problem of waste of manpower and material resources caused by manual inspections.

[0109] In the specific implementation process, after obtaining the operating data of the power generation access device, the above-mentioned device also includes a removal unit, a normalization unit and an analysis unit. The removal unit is used to remove outliers and noise in the operating data through a sliding window method to obtain denoised operating data; the normalization unit is used to normalize the denoised operating data to obtain normalized operating data; the analysis unit is used to analyze the normalized operating data through a principal component analysis method to obtain the principal component characteristics of the normalized operating data.

[0110] In this scheme, the sliding window method is used to remove outliers and noise in the operating data, which can improve the accuracy and reliability of the data, reduce the possibility of misdiagnosis, and thus more accurately diagnose the fault of the device. Normalizing the denoised operating data can eliminate the dimensional differences between different variables, making the data easier to compare and analyze. By analyzing the normalized operating data through the principal component analysis method, the main change direction in the data can be found and the main features can be identified, so as to better understand the structure and characteristics of the data and provide stronger support and basis for fault diagnosis.

[0111] In the specific implementation process, before analyzing the operating data through the data analysis model, the above-mentioned device also includes a construction unit and a training unit. The construction unit is used to describe the time series characteristics of the operating data through an autoregressive integrated moving average model, and determine the model parameters to construct the autoregressive integrated moving average model as the initial data analysis model; the training unit is used to use a support vector machine algorithm to train the initial data analysis model through historical operating data and data containing data status labels to obtain a data analysis model.

[0112] In this scheme, the time series characteristics of the operating data can be described through the autoregressive integrated moving average model, and the model parameters can be determined to build an initial data analysis model, identify trends and periodicities in the data, and thus better understand the regularity of the data. The initial data analysis model can be trained by the support vector machine algorithm to better identify patterns and features in the data, thereby building a more accurate data analysis model. The support vector machine algorithm can effectively process high-dimensional data and improve the accuracy and generalization ability of the model. This combined device can better analyze historical operating data, quickly and accurately identify faults and predict possible faults, detect problems early, take timely measures to repair and repair, improve the reliability and safety of the equipment, and ensure the stable operation of the equipment.

[0113] In some embodiments, the training unit includes a determination module and a solution module. The determination module is used to extract features of historical operating data through an autoregressive integrated moving average model, and use the features as input variables of a support vector machine algorithm, and use data containing data state labels as output variables of the support vector machine algorithm; the solution module is used to optimize the parameters of the support vector machine algorithm through a grid search and cross-validation device, and use a sequential minimum optimization algorithm to solve the objective function of the support vector machine algorithm to obtain a hyperplane with a maximum interval to complete the training of the initial data analysis model.

[0114] In this solution, historical operating data is trained through the support vector machine algorithm to identify patterns and regularities in the data, thereby better understanding the operating status and potential problems of the equipment. By optimizing the algorithm parameters and solving the objective function of the support vector machine algorithm, the generalization ability and accuracy of the model can be improved, so that equipment failures can be diagnosed and predicted more accurately, and equipment problems can be discovered in time and corresponding maintenance measures can be taken to reduce the impact of equipment failures on production and operations and improve the reliability and stability of the equipment.

[0115] During the specific implementation process, after determining that the power generation access device has failed, the above-mentioned device also includes a sending unit for generating an alarm message and sending the alarm message to the client, wherein the alarm message is used to notify the staff that the power generation access device is in a faulty state.

[0116] In this solution, by generating and sending alarm information, it is helpful to timely discover the fault status of the power generation access device, so that repair measures can be taken in time, reducing power outage time and production losses, improving the fault diagnosis efficiency of the power generation access device, reducing fault handling time, ensuring the safety of staff and the normal operation of equipment, and timely repair of faults can also extend the service life of the equipment and reduce maintenance costs.

[0117] In the specific implementation process, after determining that the operating data is fault data, the above-mentioned device also includes a determination unit and a generation unit. The determination unit is used to classify the fault data and use cluster analysis to perform pattern recognition on the fault data to determine the cause and location of the fault; the generation unit is used to generate a fault diagnosis report based on the cause and location of the fault to guide the staff to perform fault repair.

[0118] In this solution, fault data is classified and pattern recognition is performed using cluster analysis, which can help staff quickly locate the cause and location of the fault, improve the efficiency and accuracy of fault diagnosis, and generate a fault diagnosis report to clearly understand the cause and location of the fault, so as to carry out targeted maintenance, reduce maintenance time and costs, and improve equipment reliability and operating efficiency.

[0119] In some embodiments, the acquisition unit includes an acquisition module for acquiring operation data of the power generation access device through a digital sensor, wherein the digital sensor is electrically connected to the power generation access device.

[0120] In this solution, the operating data of the power generation access device is obtained through digital sensors, which can reflect the operating status of the equipment, help monitor the operating status of the equipment, detect any abnormal conditions, and promptly discover possible equipment failures. By analyzing the operating data, maintenance and repairs can be carried out in a timely manner to ensure the normal operation of the equipment and extend the life of the equipment.

[0121] The fault diagnosis device for the power generation access device includes a processor and a memory. The acquisition unit, analysis unit, and determination unit are stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions. The modules are all located in the same processor; or, the modules are located in different processors in any combination.

[0122] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problem of lack of intelligent fault prediction and diagnosis of power generation access devices in the prior art can be solved by adjusting kernel parameters.

[0123] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0124] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, a device where the computer-readable storage medium is located is controlled to execute a fault diagnosis method for a power generation access device.

[0125] Specifically, the fault diagnosis method of the power generation access device includes:

[0126] Step S201, obtaining operation data of the power generation access device, wherein the operation data at least includes voltage and current.

[0127] Specifically, by acquiring the operating data of the power generation access device, the operating status of the power generation access device can be monitored in real time, abnormal situations can be discovered in time, and the timeliness and accuracy of fault diagnosis can be improved. By monitoring and analyzing the operating data, it can be determined whether the power generation access device is operating normally and whether there are abnormal situations, thereby providing early warning of possible faults, discovering the wear and aging of the equipment in time, and formulating reasonable maintenance plans to ensure the safe and stable operation of the equipment.

[0128] Step S202: Analyze the operating data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through machine learning training of multiple groups of data, and each group of the multiple groups of data includes: historical operating data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operating data is fault data.

[0129] Specifically, the operation data is analyzed by a data analysis model to obtain a classification probability, wherein the data analysis model is obtained by machine learning training through multiple groups of data, and each group of the multiple groups of data includes: historical operation data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operation data is fault data.

[0130] Step S203: when the classification probability is greater than a preset threshold, it is determined that the operation data is fault data, and it is determined that a fault occurs in the power generation access device.

[0131] Specifically, by judging whether the data is fault data based on the classification probability, and determining that the power generation access device has a fault when the probability is greater than a preset threshold, the fault of the power generation access device can be identified quickly and accurately, and repair measures can be taken in time to avoid the expansion of the fault and cause more serious problems, thereby improving the reliability and stability of the equipment and ensuring the normal operation of the power generation system.

[0132] Optionally, after obtaining the operating data of the power generation access device, the fault diagnosis method of the power generation access device also includes: removing outliers and noise in the operating data by a sliding window method to obtain denoised operating data; normalizing the denoised operating data to obtain normalized operating data; analyzing the normalized operating data by a principal component analysis method to obtain the principal component characteristics of the normalized operating data.

[0133] Optionally, before analyzing the operating data through the data analysis model, the fault diagnosis method of the power generation access device also includes: describing the time series characteristics of the operating data through an autoregressive integrated moving average model, and determining the model parameters, and constructing the autoregressive integrated moving average model as an initial data analysis model; using a support vector machine algorithm to train the initial data analysis model through historical operating data and data containing data status labels to obtain a data analysis model.

[0134] Optionally, a support vector machine algorithm is used to train the initial data analysis model using historical operating data and data containing data status labels, including: extracting features of the historical operating data through an autoregressive integrated sliding average model, using the features as input variables of the support vector machine algorithm, and using the data containing data status labels as output variables of the support vector machine algorithm; optimizing the parameters of the support vector machine algorithm through grid search and cross-validation methods, using a sequential minimum optimization algorithm to solve the objective function of the support vector machine algorithm, and obtaining a hyperplane with the maximum interval to complete the training of the initial data analysis model.

[0135] Optionally, after determining that the power generation access device has failed, the fault diagnosis method for the power generation access device further includes: generating alarm information and sending the alarm information to the client, wherein the alarm information is used to notify the staff that the power generation access device is in a faulty state.

[0136] Optionally, after determining that the operating data is fault data, the fault diagnosis method of the power generation access device also includes: classifying the fault data, and using cluster analysis to perform pattern recognition on the fault data to determine the cause and location of the fault; generating a fault diagnosis report based on the cause and location of the fault to guide staff to perform fault repairs.

[0137] Optionally, obtaining the operating data of the power generation access device includes: obtaining the operating data of the power generation access device through a digital sensor, wherein the digital sensor is electrically connected to the power generation access device.

[0138] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a program stored in the memory and executable on the processor, wherein the processor implements at least the following steps when executing the program: obtaining operation data of a power generation access device, wherein the operation data at least includes voltage and current; analyzing the operation data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through machine learning training of multiple groups of data, and each group of data in the multiple groups of data includes: historical operation data and data including data status labels, the data status labels include normal status labels and fault status labels, and the classification probability indicates the probability that the operation data is fault data; when the classification probability is greater than a preset threshold, determining that the operation data is fault data, and determining that the power generation access device is faulty. The device in this article may be a server, a PC, a PAD, a mobile phone, etc.

[0139] Optionally, after obtaining the operating data of the power generation access device, the fault diagnosis method of the power generation access device also includes: removing outliers and noise in the operating data by a sliding window method to obtain denoised operating data; normalizing the denoised operating data to obtain normalized operating data; analyzing the normalized operating data by a principal component analysis method to obtain the principal component characteristics of the normalized operating data.

[0140] Optionally, before analyzing the operating data through the data analysis model, the fault diagnosis method of the power generation access device also includes: describing the time series characteristics of the operating data through an autoregressive integrated moving average model, and determining the model parameters, and constructing the autoregressive integrated moving average model as an initial data analysis model; using a support vector machine algorithm to train the initial data analysis model through historical operating data and data containing data status labels to obtain a data analysis model.

[0141] Optionally, a support vector machine algorithm is used to train the initial data analysis model using historical operating data and data containing data status labels, including: extracting features of the historical operating data through an autoregressive integrated sliding average model, using the features as input variables of the support vector machine algorithm, and using the data containing data status labels as output variables of the support vector machine algorithm; optimizing the parameters of the support vector machine algorithm through grid search and cross-validation methods, using a sequential minimum optimization algorithm to solve the objective function of the support vector machine algorithm, and obtaining a hyperplane with the maximum interval to complete the training of the initial data analysis model.

[0142] Optionally, after determining that the power generation access device has failed, the fault diagnosis method for the power generation access device further includes: generating alarm information and sending the alarm information to the client, wherein the alarm information is used to notify the staff that the power generation access device is in a faulty state.

[0143] Optionally, after determining that the operating data is fault data, the fault diagnosis method of the power generation access device also includes: classifying the fault data, and using cluster analysis to perform pattern recognition on the fault data to determine the cause and location of the fault; generating a fault diagnosis report based on the cause and location of the fault to guide staff to perform fault repairs.

[0144] Optionally, obtaining the operating data of the power generation access device includes: obtaining the operating data of the power generation access device through a digital sensor, wherein the digital sensor is electrically connected to the power generation access device.

[0145] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method in each embodiment of the present application: obtaining operating data of a power generation access device, wherein the operating data includes at least voltage and current; analyzing the operating data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through machine learning training based on multiple groups of data, and each group of data in the multiple groups of data includes: historical operating data and data containing data status labels, the data status labels include normal status labels and fault status labels, and the classification probability represents the probability that the operating data is fault data; when the classification probability is greater than a preset threshold, it is determined that the operating data is fault data, and it is determined that a fault has occurred in the power generation access device.

[0146] Optionally, after obtaining the operating data of the power generation access device, the fault diagnosis method of the power generation access device also includes: removing outliers and noise in the operating data by a sliding window method to obtain denoised operating data; normalizing the denoised operating data to obtain normalized operating data; analyzing the normalized operating data by a principal component analysis method to obtain the principal component characteristics of the normalized operating data.

[0147] Optionally, before analyzing the operating data through the data analysis model, the fault diagnosis method of the power generation access device also includes: describing the time series characteristics of the operating data through an autoregressive integrated moving average model, and determining the model parameters, and constructing the autoregressive integrated moving average model as an initial data analysis model; using a support vector machine algorithm to train the initial data analysis model through historical operating data and data containing data status labels to obtain a data analysis model.

[0148] Optionally, a support vector machine algorithm is used to train the initial data analysis model using historical operating data and data containing data status labels, including: extracting features of the historical operating data through an autoregressive integrated sliding average model, using the features as input variables of the support vector machine algorithm, and using the data containing data status labels as output variables of the support vector machine algorithm; optimizing the parameters of the support vector machine algorithm through grid search and cross-validation methods, using a sequential minimum optimization algorithm to solve the objective function of the support vector machine algorithm, and obtaining a hyperplane with the maximum interval to complete the training of the initial data analysis model.

[0149] Optionally, after determining that the power generation access device has failed, the fault diagnosis method for the power generation access device further includes: generating alarm information and sending the alarm information to the client, wherein the alarm information is used to notify the staff that the power generation access device is in a faulty state.

[0150] Optionally, after determining that the operating data is fault data, the fault diagnosis method of the power generation access device also includes: classifying the fault data, and using cluster analysis to perform pattern recognition on the fault data to determine the cause and location of the fault; generating a fault diagnosis report based on the cause and location of the fault to guide staff to perform fault repairs.

[0151] Optionally, obtaining the operating data of the power generation access device includes: obtaining the operating data of the power generation access device through a digital sensor, wherein the digital sensor is electrically connected to the power generation access device.

[0152] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0153] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0154] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0155] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0156] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0157] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0158] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0159] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0160] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0161] The above are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A fault diagnosis method for a power generation access device, characterized in that: include: Acquiring operation data of the power generation access device, wherein the operation data at least includes voltage and current; The operation data is analyzed by a data analysis model to obtain a classification probability, wherein the data analysis model is obtained by machine learning training using multiple sets of data, each set of data in the multiple sets of data includes: historical operation data and data including data state labels, the data state labels include normal state labels and fault state labels, and the classification probability indicates the probability that the operation data is fault data; When the classification probability is greater than a preset threshold, it is determined that the operating data is the fault data, and it is determined that a fault occurs in the power generation access device.

2. The fault diagnosis method for a power generation access device according to claim 1, characterized in that: After obtaining the operating data of the power generation access device, the method further includes: Removing outliers and noise from the operating data by a sliding window method to obtain denoised operating data; Normalizing the denoised operating data to obtain normalized operating data; The normalized operation data is analyzed by principal component analysis to obtain principal component characteristics of the normalized operation data.

3. The fault diagnosis method for a power generation access device according to claim 1, characterized in that: Before analyzing the operation data by using the data analysis model, the method further includes: Describing the time series characteristics of the operating data by an autoregressive integrated moving average model, determining model parameters, and constructing an autoregressive integrated moving average model as an initial data analysis model; The data analysis model is obtained by training the initial data analysis model with the historical operation data and the data containing the data status label using a support vector machine algorithm.

4. The fault diagnosis method for a power generation access device according to claim 3, characterized in that: The initial data analysis model is trained using the historical operation data and the data containing the data status label using a support vector machine algorithm, including: Extracting features of the historical operation data through the autoregressive integrated moving average model, using the features as input variables of the support vector machine algorithm, and using the data containing the data state label as output variables of the support vector machine algorithm; The parameters of the support vector machine algorithm are optimized by grid search and cross-validation methods, and the objective function of the support vector machine algorithm is solved using a sequential minimum optimization algorithm to obtain a hyperplane with a maximum interval to complete the training of the initial data analysis model.

5. The fault diagnosis method for a power generation access device according to claim 1, characterized in that: After determining that the power generation access device fails, the method further includes: Generate an alarm message and send the alarm message to the client, wherein the alarm message is used to notify the staff that the power generation access device is in a fault state.

6. The fault diagnosis method for a power generation access device according to claim 1, characterized in that: After determining that the operating data is the fault data, the method further includes: Classify the fault data into types, and use cluster analysis to perform pattern recognition on the fault data to determine the cause and location of the fault; A fault diagnosis report is generated according to the fault cause and the fault location to guide the staff to perform fault repair.

7. The fault diagnosis method for a power generation access device according to claim 1, characterized in that: Obtain operating data of power generation access devices, including: The operation data of the power generation access device is obtained through a digital sensor, wherein the digital sensor is electrically connected to the power generation access device.

8. A fault diagnosis device for a power generation access device, characterized in that: include: An acquisition unit, used to acquire operation data of the power generation access device, wherein the operation data at least includes voltage and current; an analysis unit, configured to analyze the operation data through a data analysis model to obtain a classification probability, wherein the data analysis model is obtained through multiple sets of data according to machine learning training, each set of the multiple sets of data includes: historical operation data and data including data state labels, the data state labels include normal state labels and fault state labels, and the classification probability indicates a probability that the operation data is fault data; A determination unit is used to determine that the operating data is the fault data and that a fault occurs in the power generation access device when the classification probability is greater than a preset threshold.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the fault diagnosis method for the power generation access device according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the fault diagnosis method of the power generation access device described in any one of claims 1 to 7.

Images