Generator stator winding temperature early warning method, system, device and medium

Abstract

The invention relates to a generator stator winding temperature early warning method, system and device and a medium, and the method comprises the steps: obtaining a working condition parameter of a generator at a certain moment as an independent variable, and obtaining a maximum temperature value of the end part of a stator winding after the certain moment as a dependent variable; and based on the independent variables and the dependent variables, obtaining a plurality of Spearman correlation coefficients, and selecting the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the maximum numerical value as a model training sample and a test sample. And according to the training sample and the test sample, constructing a support vector machine temperature prediction model, and obtaining a target temperature prediction model. And according to the real-time operation parameters of the generator, a temperature prediction value is obtained based on the target temperature prediction model, and a temperature early warning result is obtained based on the temperature prediction value. The temperature monitoring accuracy is high, potential fault risks can be found in time, and the fault occurrence probability is reduced. The problem that the early warning accuracy of the temperature of the generator stator winding is low due to low monitoring accuracy in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the technical field of generator temperature monitoring, and in particular to a generator stator winding temperature early warning method, system, equipment and medium. Background Art

[0002] Steam turbine generator is one of the important power generation equipment in the power system. Its stable and efficient operation is crucial to the safe operation and power supply of the power system. The stator winding end temperature is one of the important parameters that need to be monitored and managed during the operation of the steam turbine generator. The generator will generate a lot of heat during operation, especially when the load changes greatly or the working conditions change suddenly. The stator winding end temperature is prone to abnormal increase, even exceeding the design temperature range, which may cause equipment damage or accidents, seriously affecting the safe and stable operation of the generator.

[0003] Existing temperature monitoring methods mainly rely on sensors to collect temperature data and display and alarm in real time through monitoring systems. However, the existing generator stator winding temperature monitoring methods have low monitoring accuracy, resulting in low accuracy of generator stator winding temperature warning, which is difficult to meet the needs of timely monitoring and warning of the generator operating status. Summary of the invention

[0004] The embodiments of the present application provide a method, system, device and medium for warning the temperature of a generator stator winding, so as to at least solve the problem of low monitoring accuracy in the related art, resulting in low accuracy of warning the temperature of a generator stator winding.

[0005] In a first aspect, an embodiment of the present application provides a method for applying to a stator winding end of a steam turbine generator, the method comprising:

[0006] In the historical operation data of the stator winding end of the generator, the operating parameters of the generator at a certain moment are obtained as independent variables, and the maximum temperature value of the stator winding end after the certain moment is obtained as a dependent variable;

[0007] Based on the independent variable and the dependent variable, a plurality of Spearman correlation coefficients are obtained, and among the plurality of Spearman correlation coefficients, the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the largest value are selected as model training samples and test samples;

[0008] Constructing a support vector machine temperature prediction model according to the training samples and the test samples, evaluating the constructed temperature prediction model according to the evaluation index, and obtaining a target temperature prediction model;

[0009] According to the real-time operating parameters of the generator and based on the target temperature prediction model, a temperature prediction value is obtained, and based on the temperature prediction value, a temperature warning result is obtained.

[0010] In one embodiment, constructing a support vector machine temperature prediction model according to the training sample and the test sample, evaluating the constructed temperature prediction model according to an evaluation index, and obtaining a target temperature prediction model includes:

[0011] Determine the feature weights of each category in the training sample by covariance analysis, obtain feature values ​​based on the feature weights, and perform feature scaling according to the difference in the feature value ranges;

[0012] The support vector machine temperature prediction model is trained by the feature-scaled data to obtain the hyperplane fitting data;

[0013] Using the test set to test the trained support vector machine temperature prediction model to obtain the test results;

[0014] Evaluate the performance of the temperature prediction model according to an evaluation index to obtain an evaluation result, wherein the evaluation index includes a mean square error;

[0015] The model parameters are adjusted according to the evaluation results, and the model is repeatedly trained until the target prediction result is obtained to obtain the target temperature prediction model.

[0016] In one embodiment, the step of training the support vector machine temperature prediction model with data in the training set to obtain hyperplane fitting data includes:

[0017] The support vector machine temperature prediction model adopts Gaussian kernel function, polynomial kernel function and Sigmoid kernel function, and trains the Gaussian kernel function, the polynomial kernel function and the Sigmoid kernel function through the data in the training set to obtain the optimal hyperplane fitting data, and nonlinearly maps multiple data parameters related to the stator wire bar temperature in a high-dimensional space.

[0018] In one embodiment, the method further comprises:

[0019] Monitor the loss function value during the training of the support vector machine temperature prediction model;

[0020] The learning rate is automatically adjusted according to the change of the loss function value through an adaptive learning rate adjustment strategy, wherein the adaptive learning rate adjustment strategy includes an adaptive gradient algorithm, a root mean square propagation algorithm and an adaptive moment estimation algorithm.

[0021] In one embodiment, the method further comprises:

[0022] The training process of the support vector machine temperature prediction model is optimized by the Lagrange multiplier method, which includes constraint conditions and Lagrangian functions.

[0023] In one embodiment, obtaining a plurality of Spearman correlation coefficients based on the independent variable and the dependent variable includes:

[0024] According to the total number of data pairs n of the independent variable and the dependent variable, and the difference d between the size and rank values ​​of the independent variable and the dependent variable in the i-th data pair among the n data i , get the Spearman correlation coefficient;

[0025] Among them, the Spearman correlation coefficient is a positive value, indicating that the changing patterns of the independent variable and the dependent variable are positively correlated, and the Spearman correlation coefficient is a negative value, indicating that the changing patterns of the independent variable and the dependent variable are negatively correlated. The larger the absolute value of the Spearman correlation coefficient, the stronger the correlation between the independent variable and the dependent variable. The smaller the absolute value of the Spearman correlation coefficient, the weaker the correlation between the independent variable and the dependent variable. The Spearman correlation coefficient is equal to zero, indicating that there is no correlation between the independent variable and the dependent variable.

[0026] In one embodiment, the independent variables include stator current, stator voltage, active power, reactive power, load change rate, cooling water inlet temperature and cooling water inlet flow rate, hydrogen pressure, hydrogen temperature and hydrogen flow rate parameters.

[0027] In a second aspect, an embodiment of the present application provides a generator stator winding temperature warning system, the system comprising an independent variable and dependent variable acquisition module, a training sample acquisition, a test sample acquisition module, a target temperature prediction model acquisition module and a temperature warning module, wherein:

[0028] The module for obtaining independent variables and dependent variables is used to obtain the operating parameters of the generator at a certain moment as the independent variable from the historical operating data of the stator winding end of the generator, and obtain the maximum temperature value of the stator winding end after the certain moment as the dependent variable;

[0029] The module for obtaining training samples and test samples is used to obtain multiple Spearman correlation coefficients based on the independent variable and the dependent variable, and select the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the largest value among the multiple Spearman correlation coefficients as the model training sample and the test sample;

[0030] The module for acquiring a target temperature prediction model is used to construct a support vector machine temperature prediction model according to the training sample and the test sample, evaluate the constructed temperature prediction model according to the evaluation index, and acquire a target temperature prediction model;

[0031] The temperature warning module is used to obtain a temperature prediction value according to the real-time operating parameters of the generator and based on the target temperature prediction model, and to obtain a temperature warning result based on the temperature prediction value.

[0032] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a generator stator winding temperature warning method as described in the first aspect above is implemented.

[0033] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a generator stator winding temperature warning method as described in the first aspect above.

[0034] The embodiments of the present application provide a generator stator winding temperature warning method, system, device and medium which have at least the following technical effects.

[0035] In the historical operating data of the stator winding end of the generator, the operating parameters of the generator at a certain moment are obtained as independent variables, and the maximum temperature value of the stator winding end after a certain moment is obtained as the dependent variable. Based on the independent variables and dependent variables, multiple Spearman correlation coefficients are obtained. Among the multiple Spearman correlation coefficients, the independent variables and dependent variables corresponding to the Spearman correlation coefficient with the largest value are selected as model training samples and test samples. According to the training samples and test samples, a support vector machine temperature prediction model is constructed, and the constructed temperature prediction model is evaluated according to the evaluation index to obtain the target temperature prediction model. According to the real-time operating parameters of the generator, based on the target temperature prediction model, the temperature prediction value is obtained, and based on the temperature prediction value, the temperature warning result is obtained. The temperature monitoring accuracy of this application is high, and potential fault risks can be discovered in time, thereby reducing the probability of faults. It solves the problem that the monitoring accuracy in the related technology is not high, resulting in low accuracy of the temperature warning of the generator stator winding.

[0036] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0038] Figure 1 is a flow chart of a generator stator winding temperature early warning method according to an embodiment of the present application;

[0039] Figure 2 is a flowchart of step S103 according to an exemplary embodiment;

[0040] Figure 3 is a block diagram of a system for early warning of stator winding temperature of a generator according to an exemplary embodiment

[0041] Figure 4 is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0043] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.

[0044] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0045] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0046] In a first aspect, an embodiment of the present application provides a generator stator winding temperature early warning method, which is applied to the end of a steam turbine generator stator winding. Figure 1 It is a flow chart of the generator stator winding temperature warning, such as Figure 1 As shown, a generator stator winding temperature early warning method includes:

[0047] Step S101, in the historical operation data of the stator winding end of the generator, the operating parameters of the generator at a certain moment are obtained as independent variables, and the maximum temperature value of the stator winding end after a certain moment is obtained as a dependent variable.

[0048] Step S102: Based on the independent variable and the dependent variable, a plurality of Spearman correlation coefficients are obtained, and among the plurality of Spearman correlation coefficients, the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the largest value are selected as model training samples and test samples.

[0049] Step S103: construct a support vector machine temperature prediction model based on the training samples and the test samples, evaluate the constructed temperature prediction model based on the evaluation index, and obtain a target temperature prediction model.

[0050] Step S103: construct a support vector machine temperature prediction model based on the training samples and the test samples, evaluate the constructed temperature prediction model based on the evaluation index, and obtain a target temperature prediction model.

[0051] Step S104: According to the real-time operating parameters of the generator and based on the target temperature prediction model, a temperature prediction value is obtained, and based on the temperature prediction value, a temperature warning result is obtained.

[0052] In summary, in the historical operating data of the stator winding end of the generator, the operating parameters of the generator at a certain moment are obtained as independent variables, and the maximum temperature value of the stator winding end after a certain moment is obtained as the dependent variable. Based on the independent variables and dependent variables, multiple Spearman correlation coefficients are obtained. Among the multiple Spearman correlation coefficients, the independent variables and dependent variables corresponding to the Spearman correlation coefficient with the largest value are selected as model training samples and test samples. According to the training samples and test samples, a support vector machine temperature prediction model is constructed, and the constructed temperature prediction model is evaluated according to the evaluation index to obtain the target temperature prediction model. According to the real-time operating parameters of the generator, based on the target temperature prediction model, the temperature prediction value is obtained, and based on the temperature prediction value, the temperature warning result is obtained. The temperature monitoring accuracy of this application is high, and potential fault risks can be discovered in time, thereby reducing the probability of faults. It solves the problem that the monitoring accuracy in the related technology is not high, resulting in low accuracy of the temperature warning of the generator stator winding.

[0053] In one embodiment, step S101, in the historical operation data of the stator winding end of the generator, obtain the operating parameters of the generator at a certain moment as the independent variable, and obtain the maximum temperature value of the stator winding end after a certain moment as the dependent variable. Specifically includes:

[0054] From the historical operating data of the generator stator winding end, select the various parameter values ​​of the motor at a certain time (denoted as t) as the independent variable x i .x i Specifically, it includes but is not limited to stator current, stator voltage, active power, reactive power, load change rate, cooling water inlet temperature and flow rate, hydrogen pressure, hydrogen temperature, hydrogen flow rate and other parameters. Get the maximum temperature value of the stator winding end after time t as the dependent variable y. Use the information at the past time point to predict a certain result in the future. In this example, we use the data at time t to predict the maximum temperature value of the stator winding end after time t.

[0055] By using historical data to train the model, the accuracy of predicting future maximum stator winding end temperatures can be improved.

[0056] In one embodiment, step S102, based on the independent variable and the dependent variable, obtain multiple Spearman correlation coefficients, and select the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the largest value as the model training sample and the test sample. Specifically includes:

[0057] According to the total number of data pairs of independent variables and dependent variables n, and the difference d between the size and rank of the independent variable and the dependent variable in the i-th data pair among n data i , get the Spearman correlation coefficient;

[0058] Among them, the Spearman correlation coefficient is a positive value, indicating that the changing patterns of the independent variable and the dependent variable are positively correlated. The Spearman correlation coefficient is a negative value, indicating that the changing patterns of the independent variable and the dependent variable are negatively correlated. The larger the absolute value of the Spearman correlation coefficient, the stronger the correlation between the independent variable and the dependent variable. The smaller the absolute value of the Spearman correlation coefficient, the weaker the correlation between the independent variable and the dependent variable. The Spearman correlation coefficient is equal to zero, indicating that there is no correlation between the independent variable and the dependent variable.

[0059] Optionally, the Spearman correlation coefficient is a nonparametric statistic used to measure the monotonic relationship between two variables. Unlike the Pearson correlation coefficient, it is based on the rank of the variable rather than the actual value. Therefore, the Spearman correlation coefficient can better handle nonlinear relationships and non-normally distributed data. Step 2 uses the Spearman correlation coefficient method to select parameters with strong correlation to form the model training sample. The Spearman correlation coefficient is a method used to characterize the strength of the correlation between two sets of parameters (independent variables and dependent variables). The calculation formula is as follows:

[0060]

[0061] Where ρ is the Spearman correlation coefficient, which ranges from -1 to 1. A positive value indicates that the two parameters are positively correlated, and a negative value indicates that the two parameters are negatively correlated. The larger the absolute value of ρ, the stronger the correlation between the two parameters. The smaller the absolute value, the weaker the correlation between the two parameters. ρ = 0 indicates that there is no correlation between the two parameters. n represents the total number of data logarithms of the two parameters, and d represents the total number of data logarithms of the two parameters. i Represents the difference in the size and rank values ​​of the two parameters in the i-th data pair among the n data; the Spearman correlation coefficient of the pre-selected independent variables in step 1 and the maximum temperature value of the stator winding end are calculated respectively, and the parameters with the larger absolute value of ρ are selected to form the model training sample data set and the test sample data set, among which 80% of the sample data are input into the support vector machine temperature prediction model for training, and 20% of the sample data are used to evaluate the model prediction effect.

[0062] Through step S102, parameters with a strong correlation with the temperature of the stator winding end are selected as input quantities of the model, which reduces input information redundancy and model complexity and improves the working efficiency of the model.

[0063] Figure 2 is a flowchart of step S103 according to an exemplary embodiment. Figure 2 As shown, step S103, construct a support vector machine temperature prediction model according to the training samples and the test samples, evaluate the constructed temperature prediction model according to the evaluation index, and obtain the target temperature prediction model.

[0064] Step S1031 : determining the feature weights of each category in the training sample through covariance analysis, obtaining feature values ​​based on the feature weights, and performing feature scaling according to the differences in feature value ranges.

[0065] Optionally, data about the variables to be predicted and the predictor variables are imported, and covariance analysis is used to determine the weights of each feature in the training sample. Covariance analysis can determine the linear relationship between features and determine the degree of influence of each feature on the predictor variable (the maximum temperature value at the end of the stator winding). Based on the results of the covariance analysis, a weight is assigned to each feature. The weight reflects the importance of the feature to the predictor variable. The feature value, that is, the actual observed value of each feature, is obtained according to the feature weight. If the range of the feature value is very different, feature scaling is performed to scale the feature value to the same range to avoid excessive influence of certain features on the model prediction results. Feature scaling can be achieved by standardization or normalization. Step S1031 determines the feature weights through covariance analysis, and can identify which features are most important to the prediction results. It helps feature selection and reduces the complexity of the model. Feature scaling can improve the performance of the model, especially when using gradient-based optimization methods such as distance or gradient descent. The scaled features can accelerate the convergence speed of the model and reduce the risk of overfitting.

[0066] Step S1032: training a support vector machine temperature prediction model using the feature scaled data to obtain hyperplane fitting data.

[0067] Optionally, the support vector machine temperature prediction model uses Gaussian kernel function, polynomial kernel function and Sigmoid kernel function, trains Gaussian kernel function, polynomial kernel function and Sigmoid kernel function through the data in the training set, obtains the optimal hyperplane fitting data, and nonlinearly maps multiple data parameters related to the stator bar temperature in a high-dimensional space.

[0068] The Gaussian kernel function is determined by the following formula:

[0069] K(x,y)=exp[-xy 2 (2σ 2 )]

[0070] Where σ is a constant;

[0071] The polynomial kernel function is determined by the following formula:

[0072] K(x,y)=( <xy>+R) d ,R≥0,d≤0

[0073] Where: R is a constant, d is a power, <xy>is a polynomial involving x and y;

[0074] Sigmoid kernel function: determined by the following formula:

[0075] K(x,y)=tanh(g <xy>+h) d ,g>0,h<0

[0076] Where: tanh is the hyperbolic tangent function, g is the coefficient constant, and h is a constant.

[0077] Specifically, the data processed by feature scaling is used as the training set. Gaussian kernel function, polynomial kernel function and Sigmoid kernel function are selected as the kernel function of the SVM model. These kernel functions are trained using the data in the training set to find the optimal hyperplane to fit the data. The kernel function is used to nonlinearly map multiple data parameters related to the stator wire bar temperature in a high-dimensional space in order to better separate the data. Step S1032 By using different kernel functions, the support vector machine model can process nonlinearly separable data and improve prediction accuracy. The kernel function allows the model to automatically perform feature mapping without explicitly increasing the feature dimension, which helps to reduce the complexity of the model. By finding the optimal hyperplane in a high-dimensional space, the support vector machine model can better generalize to unseen data.

[0078] Step S1033: Use the test set to test the trained support vector machine temperature prediction model to obtain the test results.

[0079] Optionally, 20% of the sample data that has not participated in the model training before is used as a test set, and the trained support vector machine temperature prediction model is tested using the test set. The prediction results of the model on the test set are recorded, and these results are used to evaluate the performance of the model. Step S1033 can evaluate the performance of the model through the test set to understand the performance of the model on real-world data. The test set checks whether the model is overfitted to the training data. If the model performs well on the training set but performs poorly on the test set, there may be an overfitting problem.

[0080] Step S1034: Evaluate the performance of the temperature prediction model according to evaluation indicators to obtain evaluation results, where the evaluation indicators include mean square error.

[0081] Optionally, the mean square error (MSE) is selected as the evaluation indicator. The mean square error is calculated using the difference between the predicted result and the actual value on the test set. The mean square error value is recorded as the evaluation result to evaluate the performance of the model. Step S1034 provides a quantitative measure through the mean square error to directly compare the performance of different models or the performance of the same model under different parameter settings. By evaluating the performance of the model, it can be determined which aspects need to be improved, such as adjusting model parameters, adding or reducing features, using different kernel functions, etc.

[0082] Step S1035: adjust the model parameters according to the evaluation results, repeat the model training, until the target prediction result is obtained, and obtain the target temperature prediction model.

[0083] Optionally, the performance of the model is analyzed according to the evaluation results (e.g., mean square error) obtained in step S1034. Based on the evaluation results, the parameters of the model are adjusted to optimize the performance. The model is retrained using the adjusted parameters. The above process is repeated until the performance of the model reaches a predetermined target level. In response to reaching the target prediction result, the final target temperature prediction model is saved and obtained. Step S1035 Through continuous parameter adjustment and model training, the prediction performance of the model can be gradually improved, so that the model can more accurately predict the stator bar temperature.

[0084] In one embodiment, the generator stator winding temperature warning further includes:

[0085] Monitor the loss function value during the training of the support vector machine temperature prediction model;

[0086] The learning rate is automatically adjusted according to the change of the loss function value through an adaptive learning rate adjustment strategy, wherein the adaptive learning rate adjustment strategy includes an adaptive gradient algorithm, a root mean square propagation algorithm and an adaptive moment estimation algorithm.

[0087] Optionally, define a loss function to measure the gap between the model's predictions and the actual labels, and monitor changes in the loss function value during training. The adaptive learning rate adjustment strategy automatically adjusts the learning rate according to changes in the loss function value to accelerate the convergence of the model and improve the generalization ability of the model. Adaptive learning rate adjustment strategies include adaptive gradient algorithms, root mean square propagation algorithms, and adaptive moment estimation algorithms. The adaptive learning rate adjustment strategy can help the model converge to the optimal solution faster, especially in non-convex optimization problems. By dynamically adjusting the learning rate, overfitting and underfitting problems can be avoided, thereby improving the model's generalization ability to new data.

[0088] In one embodiment, the generator stator winding temperature warning further includes:

[0089] The training process of the support vector machine temperature prediction model is optimized by the Lagrange multiplier method, which includes constraints and Lagrangian functions.

[0090] Optionally, the dual problem of the original optimization problem is constructed using the Lagrange multiplier method, which includes constraints and the Lagrangian function, which is shown as follows:

[0091]

[0092] In the formula, λ j is a Lagrange multiplier, and its optimality necessary condition is:

[0093]

[0094] Specifically, the Lagrange multiplier method is used to transform the original optimization problem into its dual problem, and a dual problem containing constraints and Lagrange functions is constructed. The Lagrange function is defined, which contains Lagrange multipliers. The non-negativity, dual feasibility and complementary slackness conditions of Lagrange multipliers are satisfied. By constructing the dual problem, the solution process of the optimization problem can be simplified, especially when dealing with nonlinearly separable data. The dual problem is usually easier to solve, especially when using the kernel trick, which can handle nonlinearly separable data in high-dimensional space. By optimizing the training process, the predictive performance of the model can be improved and the risk of overfitting can be reduced.

[0095] In one embodiment, step S104, according to the real-time operating parameters of the generator and based on the target temperature prediction model, obtains a temperature prediction value, and based on the temperature prediction value, obtains a temperature warning result. Specifically includes:

[0096] Optionally, various parameter values ​​of the generator during real-time operation are obtained. The real-time operating parameters include various key parameters when the motor is running, such as current, voltage, speed, etc. These parameters are the basis for the model to predict the temperature. Use the trained target temperature prediction model to predict the real-time operating parameters to obtain a temperature prediction value. Based on the temperature prediction value, determine whether it exceeds the warning threshold and generate a temperature warning result. When the temperature prediction value exceeds the threshold, a warning signal is triggered. By establishing a temperature prediction model based on a machine learning algorithm, the maximum temperature of the stator winding end when the motor is running can be predicted more accurately, and a warning can be issued based on the predicted value so that operation and maintenance personnel can take corresponding measures to reduce the occurrence of failures and improve the operating reliability of the motor.

[0097] In summary, in the historical operating data of the stator winding end of the generator, the operating parameters of the generator at a certain moment are obtained as independent variables, and the maximum temperature value of the stator winding end after a certain moment is obtained as the dependent variable. Based on the independent variables and dependent variables, multiple Spearman correlation coefficients are obtained. Among the multiple Spearman correlation coefficients, the independent variables and dependent variables corresponding to the Spearman correlation coefficient with the largest value are selected as model training samples and test samples. According to the training samples and test samples, a support vector machine temperature prediction model is constructed, and the constructed temperature prediction model is evaluated according to the evaluation index to obtain the target temperature prediction model. According to the real-time operating parameters of the generator, based on the target temperature prediction model, the temperature prediction value is obtained, and based on the temperature prediction value, the temperature warning result is obtained. The temperature monitoring accuracy of this application is high, and potential fault risks can be discovered in time, thereby reducing the probability of faults. It solves the problem that the monitoring accuracy in the related technology is not high, resulting in low accuracy of the temperature warning of the generator stator winding.

[0098] Principle of the invention: In view of the shortcomings of traditional monitoring technologies for the temperature of stator winding ends, the present invention proposes a temperature early warning method for the stator winding ends of a steam turbine generator. By establishing a temperature prediction model based on a machine learning algorithm, the maximum temperature of the stator winding ends when the motor is running can be accurately predicted, and an early warning can be issued based on the predicted value so that operation and maintenance personnel can take corresponding measures to reduce the occurrence rate of failures, improve the operating reliability of the motor, and ensure the safe and stable operation of the power system. It has important practical application value.

[0099] In a second aspect, an embodiment of the present application provides a system for early warning of temperature of a generator stator winding. Figure 3 FIG. 1 is a block diagram of a system for early warning of stator winding temperature of a generator according to an exemplary embodiment. Figure 3 As shown, the system includes a module 310 for obtaining independent variables and dependent variables, a module 320 for obtaining training samples and test samples, a module 330 for obtaining a target temperature prediction model, and a temperature warning module 340, wherein:

[0100] The module 310 for obtaining independent variables and dependent variables is used to obtain the operating parameters of the generator at a certain moment as the independent variable and obtain the maximum temperature value of the stator winding end after a certain moment as the dependent variable from the historical operating data of the stator winding end of the generator;

[0101] The module 320 for obtaining training samples and test samples is used to obtain multiple Spearman correlation coefficients based on the independent variable and the dependent variable, and select the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the largest value as the model training sample and the test sample.

[0102] The module 330 for obtaining a target temperature prediction model is used to construct a support vector machine temperature prediction model according to the training samples and the test samples, evaluate the constructed temperature prediction model according to the evaluation index, and obtain the target temperature prediction model;

[0103] The temperature warning module 340 is used to obtain a temperature prediction value according to the real-time operating parameters of the generator and based on the target temperature prediction model, and obtain a temperature warning result based on the temperature prediction value.

[0104] In summary, the embodiment of the present application provides a kind of generator stator winding temperature early warning system, through the independent variable and dependent variable module 310, the acquisition of training samples, the test sample module 320, the acquisition of the target temperature prediction model module 330 and the temperature early warning module 340, to establish a temperature prediction model based on a machine learning algorithm, which can accurately predict the maximum temperature of the stator winding end when the motor is running, and give an early warning according to the predicted value, so that the operation and maintenance personnel can take corresponding measures to reduce the failure rate, improve the operation reliability of the motor, and ensure the safe and stable operation of the power system, which has important practical application value. Specifically, in the historical operation data of the stator winding end of the generator, the operating parameters of the generator at a certain moment are obtained as independent variables, and the maximum temperature value of the stator winding end after a certain moment is obtained as a dependent variable. Based on the independent variable and the dependent variable, multiple Spearman correlation coefficients are obtained, and the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the largest value are selected from the multiple Spearman correlation coefficients as the model training sample and the test sample. According to the training sample and the test sample, a support vector machine temperature prediction model is constructed, and the constructed temperature prediction model is evaluated according to the evaluation index to obtain the target temperature prediction model. According to the real-time operating parameters of the generator, based on the target temperature prediction model, the temperature prediction value is obtained, and based on the temperature prediction value, the temperature warning result is obtained. The temperature monitoring accuracy of this application is high, and potential fault risks can be discovered in time, thereby reducing the probability of faults. The problem of low monitoring accuracy in related technologies, which leads to low accuracy of temperature warning of the generator stator winding, is solved.

[0105] It should be noted that the system for early warning of the temperature of the stator winding of a generator provided in this embodiment is used to implement the above-mentioned implementation mode, and the descriptions that have been made will not be repeated. As used above, the terms "module", "unit", "subunit", etc. can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the above embodiment is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0106] In a third aspect, an embodiment of the present application provides an electronic device, Figure 4 FIG. 1 is a block diagram of an electronic device according to an exemplary embodiment. Figure 4 As shown, the electronic device may include a processor 41 and a memory 42 storing computer program instructions.

[0107] Specifically, the processor 41 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0108] Among them, the memory 42 may include a large-capacity memory for data or instructions. By way of example and not limitation, the memory 42 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 42 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 42 may be inside or outside the data processing device. In a specific embodiment, the memory 42 is a non-volatile memory. In a specific embodiment, the memory 42 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, PROM for short), an erasable PROM (Erasable ProgrammableRead-Only Memory, EPROM for short), an electrically erasable PROM (Electrically Erasable ProgrammableRead-Only Memory, EEPROM for short), an electrically alterable ROM (Electrically Alterable Read-Only Memory, EAROM for short) or a flash memory (FLASH) or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), wherein the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0109] The memory 42 may be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 41 .

[0110] The processor 41 implements any one of the generator stator winding temperature early warning methods in the above embodiments by reading and executing computer program instructions stored in the memory 42 .

[0111] In one embodiment, a device for early warning of stator winding temperature of a generator may further include a communication interface 43 and a bus 40. Figure 4 As shown, the processor 41, the memory 42, and the communication interface 43 are connected via a bus 40 and communicate with each other.

[0112] The communication interface 43 is used to implement communication between the modules, devices, units and / or equipment in the embodiment of the present application. The communication port 43 can also implement data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.

[0113] The bus 40 includes hardware, software or both, and couples the components of a generator stator winding temperature warning device to each other. The bus 40 includes but is not limited to at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example and not limitation, bus 40 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of the above. Where appropriate, bus 40 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0114] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, a generator stator winding temperature warning method provided in the first aspect is implemented.

[0115] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0116] In a possible implementation, the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps of a generator stator winding temperature warning method provided in the first aspect.

[0117] Among them, the program code for executing the present invention can be written in any combination of one or more programming languages, and the program code can be executed completely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or completely on the remote device.

[0118] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.< / xy> < / xy> < / xy>

Claims

1. A generator stator winding temperature early warning method, characterized in that: Applied to the stator winding end of a steam turbine generator, the method comprises: In the historical operation data of the stator winding end of the generator, the operating parameters of the generator at a certain moment are obtained as independent variables, and the maximum temperature value of the stator winding end after the certain moment is obtained as a dependent variable; Based on the independent variable and the dependent variable, a plurality of Spearman correlation coefficients are obtained, and among the plurality of Spearman correlation coefficients, the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the largest value are selected as model training samples and test samples; Constructing a support vector machine temperature prediction model according to the training samples and the test samples, evaluating the constructed temperature prediction model according to the evaluation index, and obtaining a target temperature prediction model; According to the real-time operating parameters of the generator and based on the target temperature prediction model, a temperature prediction value is obtained, and based on the temperature prediction value, a temperature warning result is obtained.

2. The method according to claim 1, characterized in that The step of constructing a support vector machine temperature prediction model according to the training sample and the test sample, evaluating the constructed temperature prediction model according to an evaluation index, and obtaining a target temperature prediction model includes: Determine the feature weights of each category in the training sample by covariance analysis, obtain feature values ​​based on the feature weights, and perform feature scaling according to the difference in the feature value ranges; The support vector machine temperature prediction model is trained by the feature-scaled data to obtain the hyperplane fitting data; Using the test set to test the trained support vector machine temperature prediction model to obtain the test results; Evaluate the performance of the temperature prediction model according to an evaluation index to obtain an evaluation result, wherein the evaluation index includes a mean square error; The model parameters are adjusted according to the evaluation results, and the model is repeatedly trained until the target prediction result is obtained to obtain the target temperature prediction model.

3. The method according to claim 2, characterized in that The method of training the support vector machine temperature prediction model with the data in the training set to obtain the hyperplane fitting data includes: The support vector machine temperature prediction model adopts Gaussian kernel function, polynomial kernel function and Sigmoid kernel function, and trains the Gaussian kernel function, the polynomial kernel function and the Sigmoid kernel function through the data in the training set to obtain the optimal hyperplane fitting data, and nonlinearly maps multiple data parameters related to the stator wire bar temperature in a high-dimensional space.

4. The method according to claim 2, characterized in that: The method further comprises: Monitor the loss function value during the training of the support vector machine temperature prediction model; The learning rate is automatically adjusted according to the change of the loss function value through an adaptive learning rate adjustment strategy, wherein the adaptive learning rate adjustment strategy includes an adaptive gradient algorithm, a root mean square propagation algorithm and an adaptive moment estimation algorithm.

5. The method according to claim 2, characterized in that: The method further comprises: The training process of the support vector machine temperature prediction model is optimized by the Lagrange multiplier method, which includes constraint conditions and Lagrangian functions.

6. The method according to claim 1, characterized in that The step of obtaining a plurality of Spearman correlation coefficients based on the independent variable and the dependent variable includes: According to the total number of data pairs n of the independent variable and the dependent variable, and the difference d between the size and rank values ​​of the independent variable and the dependent variable in the i-th data pair among the n data i , get the Spearman correlation coefficient; Among them, the Spearman correlation coefficient is a positive value, indicating that the changing patterns of the independent variable and the dependent variable are positively correlated, and the Spearman correlation coefficient is a negative value, indicating that the changing patterns of the independent variable and the dependent variable are negatively correlated. The larger the absolute value of the Spearman correlation coefficient, the stronger the correlation between the independent variable and the dependent variable. The smaller the absolute value of the Spearman correlation coefficient, the weaker the correlation between the independent variable and the dependent variable. The Spearman correlation coefficient is equal to zero, indicating that there is no correlation between the independent variable and the dependent variable.

7. The method according to claim 1, characterized in that The independent variables include stator current, stator voltage, active power, reactive power, load change rate, cooling water inlet temperature and cooling water inlet flow rate, hydrogen pressure, hydrogen temperature and hydrogen flow rate parameters.

8. A generator stator winding temperature early warning system, characterized in that: The system includes a module for obtaining independent variables and dependent variables, a module for obtaining training samples and test samples, a module for obtaining a target temperature prediction model, and a temperature early warning module, wherein: The module for obtaining independent variables and dependent variables is used to obtain the operating parameters of the generator at a certain moment as the independent variable from the historical operating data of the stator winding end of the generator, and obtain the maximum temperature value of the stator winding end after the certain moment as the dependent variable; The module for obtaining training samples and test samples is used to obtain multiple Spearman correlation coefficients based on the independent variable and the dependent variable, and select the independent variable and the dependent variable corresponding to the Spearman correlation coefficient with the largest value among the multiple Spearman correlation coefficients as the model training sample and the test sample; The target temperature prediction model acquisition module is used to construct a support vector machine temperature prediction model according to the training sample and the test sample, evaluate the constructed temperature prediction model according to the evaluation index, and acquire the target temperature prediction model; The temperature warning module is used to obtain a temperature prediction value according to the real-time operating parameters of the generator and based on the target temperature prediction model, and to obtain a temperature warning result based on the temperature prediction value.

9. An electronic device, characterized in that: The invention comprises a memory and a processor, a computer program stored in the memory and executable on the processor, and the processor implements a generator stator winding temperature early warning method as claimed in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a generator stator winding temperature early warning method according to any one of claims 1 to 7 is implemented.

Images