Pellet fan fault diagnosis method

Abstract

The invention discloses a pellet fan fault diagnosis method, and aims to solve the problems of low efficiency and insufficient accuracy caused by the fact that traditional fan fault diagnosis depends on manual inspection and experience judgment. The method comprises the following steps: firstly collecting fan operation parameters to form a data set, after data cleaning and normalization processing, carrying out data enhancement by adopting random sampling and Gaussian noise addition, carrying out feature extraction by utilizing a ConvNeXt network, dividing the data set, calculating a fault category feature prototype vector, and predicting a result through distance measurement and classification decision. An L2 regular term is added in a prototype network model loss function, model parameters are adjusted through back propagation, the model is deployed after performance evaluation of a test set is stable, preprocessing and enhancing steps are repeated on real-time data, and if a prediction result is not consistent or new data exist, the model is retrained. According to the method, the diagnosis accuracy and efficiency are improved through multi-technology cooperation, the over-fitting of the model is effectively prevented, the fault feature capturing capability is improved, and the advantages of function perfection and technical perspectiveness are shown.

Description

Technical Field

[0001] The present invention relates to the technical field of blower fault diagnosis, and more specifically, to a method for diagnosing a blower fault. Background Art

[0002] Pelletizing fans play a vital role in steel production, and their performance and reliability are directly related to the stable operation of the entire steel plant. In the pelletizing production line, the pelletizing fan is responsible for providing the necessary oxygen and fuel for pellet roasting, and its efficient and stable operation plays a decisive role in the quality and output of pellets. However, since the pelletizing fan has been operating in harsh working environments (such as high temperature, high humidity, and high dust) for a long time, its internal bearings, blades and other key components may degrade in performance or even fail due to wear and aging. If these faults are not discovered and handled in time, they will not only lead to a decrease in production efficiency, but may also cause safety accidents, which will have a serious impact on the economic benefits and production safety of the enterprise.

[0003] Faced with this challenge, the traditional method of pellet blower fault diagnosis mainly relies on manual inspection and empirical judgment. This method has defects such as low efficiency and insufficient accuracy when dealing with small fault sample problems. Therefore, it is particularly important to use a few-sample method for training. Through this method, even when there are fewer fault parameters, a fault diagnosis model can be effectively trained to deal with small fault sample problems. In order to further improve the accuracy and efficiency of fault diagnosis, data enhancement technology can be used to increase the diversity of training data. This method simulates different fault conditions and generates more training samples, thereby improving the model's ability to identify various potential faults. At the same time, through the L2 regularization technology, the problem of model overfitting that may occur during small sample training can be effectively prevented to ensure the stability and reliability of the model.

[0004] These technological innovations and process improvements not only improve the stability and reliability of pelletizing blowers, but also promote the technological progress and industrial upgrading of the entire pelletizing material preparation industry. By adopting methods such as data enhancement, L2 regularization and prototype network models for fault diagnosis, the maintenance and fault diagnosis of pelletizing blowers have become more efficient and accurate, providing strong technical support for the intelligent manufacturing and automation of the steel industry, and providing a solid technical foundation for the sustainable development of the steel industry. Summary of the invention

[0005] The present invention provides a pelletizing fan fault diagnosis method to solve the problems of low efficiency and insufficient accuracy caused by reliance on manual inspection and experience judgment in the existing traditional pelletizing fan fault diagnosis.

[0006] According to one aspect of the present invention, a pelletizing blower fault diagnosis method is provided, comprising the following steps:

[0007] Step S1, collecting historical operating parameters of the pelletizing fan to form a data set, removing outliers and noise interference in the data set through data cleaning, and then normalizing the data;

[0008] Step S2, using a combination of random sampling and Gaussian noise addition to perform data enhancement on the data set;

[0009] Step S3, after feature extraction through convolutional neural network, the data set is divided into training set, validation set and test set, the training set is divided into support set and query set, the feature prototype vector of each fault category in the support set is calculated, and then the distance between the query sample and the prototype is calculated through distance metric, and the result is predicted through classification decision;

[0010] Step S4, adding an L2 regularization term to the loss function of the prototype network model, determining the L2 regularization coefficient to balance the model's fitting ability and generalization ability, adjusting the model parameters through back propagation, regularly calculating the validation set loss, and determining whether the training termination condition is met;

[0011] Step S5: input the test set data into the trained model for prediction. According to the prediction results and the true labels, the accuracy, recall rate, and F1 value are calculated and evaluated. If the performance is stable, the model is deployed.

[0012] Step S6, repeat steps S1 and S2 to process and enhance the real-time data, predict the real-time feature vector through the prototype network model constructed in S3, and output the category with the highest probability as the predicted fault type; if it is found that the prediction result of the model is obviously inconsistent with the actual situation and new data is collected, return to step S3 to re-divide the data set and re-train the model.

[0013] Preferably, based on the above scheme, in step S1, the specific steps are:

[0014] Step S11, the collected parameters include: the vibration sensor provides vibration signal data, the temperature sensor collects temperature data, and the current sensor records current parameters under known abnormal conditions in the normal state; the box plot method is used to process the collected data set, and the data points outside the upper and lower limits of the box plot are regarded as abnormal values;

[0015] Step S12, according to the characteristics of the abnormal value, select low-pass filtering, high-pass filtering and median filtering to perform noise reduction processing;

[0016] Step S13, replacing the marked outliers by Lagrange interpolation method;

[0017] Step S14, convert the one-dimensional vibration signal into a two-dimensional time-frequency diagram through short-time Fourier transform, use Z-score normalization to unify the temperature and current data into the same scale range as additional channels to be spliced ​​with the vibration time-frequency diagram, and Z-score normalization converts the data into a distribution with a mean of 0 and a standard deviation of 1. The calculation formula is:

[0018]

[0019] Where x is the original temperature and current data value, μ is the mean of the data, σ is the standard deviation, and x normalized It is the temperature and current data value after normalization and then unified scale;

[0020] Among them, in step S14, a time-frequency diagram of size N×M can be obtained through short-time Fourier transform, where N is the resolution of the frequency axis and M is the resolution of the time axis; the temperature and current data can be expanded into a matrix of the same size as the vibration time-frequency diagram after normalization, each as a channel, and the shape of the input data is N×M×C, where C is the number of channels.

[0021] Preferably, based on the above scheme, in step S2, the specific steps are:

[0022] S21, set the minimum and maximum sampling duration limits for the time series data set, randomly generate a sampling duration, and randomly select a starting point within the total duration range, extract the data segment of the corresponding duration starting from this point, and obtain a new sequence with random length and similar characteristics;

[0023] S22, sampling the feature dimension data to determine the total number of feature dimensions of the data, randomly selecting a portion of the number of feature dimensions, and randomly selecting feature dimensions therefrom to form a new feature subset;

[0024] S23, determine the mean and standard deviation of the Gaussian noise, independently generate a random noise value that obeys the Gaussian distribution of the mean and standard deviation for each data point in the original data, and add it to the original data point, and obtain new enhanced data after point-by-point operation.

[0025] Preferably, based on the above scheme, in step S3, the specific steps are:

[0026] S311, using a ConvNeXt network of a standard convolution module to perform architecture design, extracting features of the time-frequency data set; using the N×M×C time-frequency graph after data enhancement as an input layer;

[0027] S312, the first convolution layer selects a 3×3 convolution kernel, sets 32 convolution kernels to extract different basic features, adopts the "same" padding method and a step size of 1, uses the ReLU activation function, and the second and third convolution layers have the same parameters as the first convolution layer. The output feature map shape is N×M×32;

[0028] S313, add pooling windows after the first and second convolutional layers respectively, use a 2×2 maximum pooling window with a step size of 2; perform a pooling operation on the feature map output by the convolutional layer, reduce the size of the feature map by half, and the shape of the output feature map becomes

[0029] The second convolution layer also uses a 3×3 convolution kernel, sets 64 convolution kernels, the padding method and step size remain unchanged, the activation function is still ReLU, and the input is the output of the previous layer. The output feature map shape is,

[0030] Use a 2×2 maximum pooling window with a step size of 2; perform a pooling operation on the feature map output by the convolutional layer, reduce the size of the feature map by half, and the shape of the output feature map becomes

[0031] The third convolution layer still uses a 3×3 convolution kernel, sets 128 convolution kernels, "same" padding and step size to 1, the activation function is ReLU, and the input is the output of the previous layer. The output feature map shape is

[0032] S314, in the two-dimensional convolution, the calculation formula for each element (i, j, k) in the output feature map (where i and j are spatial position indexes and k is a channel index) is:

[0033]

[0034] Among them, I i+m，j+n，l Represents the element value at the corresponding position in the input feature map, K m，n，l，k Represents the weight value of the corresponding position in the convolution kernel;

[0035] S315, add BatchNormalization after each convolution layer to speed up training and improve model stability;

[0036] S316, after a series of convolutional layers and pooling layers, the feature map of shape N×M×C is obtained, where N1 and M1 are the spatial sizes of the feature map, and C is the number of channels; the feature map is flattened into a one-dimensional vector with a length of N1×M1×C;

[0037] S317, the first fully connected layer has L neurons, which are linearly transformed by a weight matrix W1 (size is N1×M1×C×L) and a bias vector b1 (length is L) to obtain the output of the fully connected layer; the formula is:

[0038] z=f(W1x+b1);

[0039] Among them, x is the flattened feature vector, and f is the ReLU function;

[0040] S318, the second fully connected layer has K neurons) (K is the number of fault classes), and a weight matrix W2 (size is N1×M1×C×K) and a bias vector b2 (length is K) are linearly transformed to obtain the output of the fully connected layer; the formula is:

[0041] y=f(W2z+b2);

[0042] Where z is the output of the first fully connected layer, and f is the Softmax function;

[0043] S32 divides the data set: divide the feature vector obtained in S31 into 70% training set, 20% validation set, and 10% test set; use the random division method to ensure that each category has a certain proportion distribution in different data sets; divide the training set into 75% support set and 25% query set;

[0044] The S321 training set is used for the model to learn data patterns and rules, adjust parameters to establish mapping relationships; the validation set is used to evaluate performance during training, help adjust hyperparameters, and determine overfitting or underfitting; the test set is used to evaluate the final performance after training and measure generalization ability; the support set (part of the training set) is used to calculate the category prototype; the query set is used to test the model's ability to classify new samples, and to make classification decisions based on the distance to the category prototype obtained based on the support set.

[0045] S33 Computing Prototype:

[0046] Computation prototype initialization phase: divide the samples in the support set according to the fault category; mark the K fault categories as C1, C2, ..., C K ; For each fault category C k (k=1,2,…,K), represents the kth fault category, K is the total number of fault categories, and a small number of initial samples are randomly selected to form the initial sample set:

[0047]

[0048] Where n is the number of initial samples; calculate the initial prototype for each fault category That is, the average value of the initial sample feature vector of this category; the calculation formula is:

[0049]

[0050] n: The number of initial samples selected for each fault category during the initialization phase

[0051] x k，i : The feature vector of the i-th sample in the k-th fault category in the initial sample set

[0052] Prototype calculation batch update phase: As new samples are added, the prototype is gradually updated in batch mode; the current number of iterations is t, and the newly added batch of samples is

[0053]

[0054] A batch of samples newly added to the kth fault category at the th iteration

[0055] y k，m : is the feature vector of the mth sample newly added to the sample set of the kth fault category at the tth iteration;

[0056] m: The number of samples newly added to the kth fault category at the tth iteration, that is, the sample set The number of samples in ;

[0057] The existing prototype is Calculate the total number of samples processed;

[0058]

[0059] Its i is the number of new samples added in the i-th iteration; the prototype is updated after iteration The calculation formula is:

[0060]

[0061] S34 distance metric:

[0062] The faults are measured and classified by the nearest neighbor classification (K-NN); the sample feature vector y of the query set is calculated using the Euclidean distance method k，n (y k，n represents the feature vector of the nth sample in the kth fault category in the query set) and the prototype set The distance between two feature vectors y = (y1, y2, ..., y n ) and x=(x1,x2,…,x n), (x and y are any two feature vectors in the query set and prototype set respectively, and n is the dimension of the feature vector)

[0063] The Euclidean distance calculation formula is:

[0064]

[0065] n is the dimension of the feature vector;

[0066] S35 Decision Classification:

[0067] According to the calculated distance, find the k training samples closest to it, which are called y k，n The k nearest neighbors of ; count the frequency of occurrence of each category in these k nearest neighbor samples;

[0068] The distance is negatively exponentially converted into a probability by the softmax function; for a sample x, the probability of belonging to the i-th fault category is calculated as:

[0069]

[0070] Where N is the total number of fault categories;

[0071] d(x,c i ): represents the relationship between sample x and the i-th fault category prototype c i The distance between

[0072] c i : represents the prototype vector of the i-th fault category;

[0073] y k，n The class with the highest probability is considered to best represent the characteristics of the new sample.

[0074] Preferably, based on the above scheme, in step S4, the specific steps are:

[0075] S41, use functions such as cross entropy loss as the basic loss function.

[0076] The true category of the sample is y, and the predicted category is The cross entropy loss function is

[0077]

[0078] If the sample belongs to the jth fault category, then the true category vector y = (0, 0, ..., 1, ..., 0), where the jth position is 1 and the rest are 0

[0079] S42, calculate the L2 regularization term based on the model parameters. The formula of the L2 regularization term is:

[0080] λ∑ w ||w|| 2 , where w is the parameter of the model and λ is the regularization coefficient;

[0081] S45, add the L2 regularization term to the basic loss function to obtain a new loss function:

[0082]

[0083] S43 back propagation;

[0084] S431 calculates the gradient of the loss function with respect to the model parameters:

[0085] Calculate the predicted probability of the sample belonging to each fault category, the distance function, and the gradient of the L2 regularization term to the model parameters, so as to use the optimizer to update the model parameters and improve the performance of the model and the accuracy of fault diagnosis;

[0086] Use the chain rule to gradually calculate the gradient of the loss function with respect to the model parameters. For each parameter, calculate

[0087] First, calculate the gradient of the cross entropy loss function to the predicted probability

[0088] Among them, L is the cross entropy loss function, y i is the true category label of the sample, is the predicted probability that the sample belongs to the i-th fault category;

[0089] According to the calculation formula of predicted probability, where d(x,c i ) is the distance between the sample and the prototype of the i-th category, and N is the total number of fault categories;

[0090] calculate where d(x,c i ) is the distance between the sample and the prototype of the i-th category, and N is the total number of fault categories:

[0091] Combined with the distance function d(x,c i ) with respect to the derivative of the parameters, we get the gradient of the predicted probability of each fault category with respect to the model parameters:

[0092]

[0093] The formula for the L2 regularization term is λ∑ m＝1 ∥w∥ 2 , where w is the parameter of the model and λ is the regularization coefficient: The gradient of the L2 regularization term with respect to the parameter is calculated as 2λw:

[0094] Finally, the total gradient is

[0095] S432, use the optimizer to update the model parameters:

[0096] S4321, initialize Adam optimizer parameters:

[0097] Set the learning rate α, usually between 0.001 and 0.01, the exponential decay rate β1 of the first-order moment estimate is 0.9; the exponential decay rate β2 of the second-order moment estimate is 0.999; set a very small constant ∈ to 1e-8 to prevent the denominator from being zero;

[0098] For each model parameter w;

[0099] Initialize the first-order moment estimate m0=0 and the second-order moment estimate v0=0;

[0100] At iteration t:

[0101] Compute the first moment estimate:

[0102] m t First-order moment estimate, representing the exponentially weighted average of the gradient;

[0103] Compute the second moment estimate:

[0104] v t The second-order moment estimate represents the exponentially weighted average of the squared gradient, which is used to estimate the variance of the gradient and help adjust the step size of the parameter update;

[0105] Compute the corrected first moment estimate:

[0106] Compute the corrected second moment estimate:

[0107] Update parameters:

[0108] By repeating the above steps, the Adam optimizer is used to update the model parameters in each iteration, so that the model is gradually optimized during the training process;

[0109] S4322, regularly calculate the validation set loss

[0110] During model training, the loss value is calculated on the validation set in each training cycle. The loss function is the cross entropy loss plus the L2 regularization term, that is, L total =L+λ∑ w ∥w∥ 2 , where L is the cross entropy loss function, λ is the regularization coefficient, and w is the model parameter:

[0111] For each sample in the validation set, input it into the model for forward propagation, calculate the prediction result, and then calculate the loss value based on the true label. Finally, the loss values ​​of all samples are averaged to get the average loss of the validation set.

[0112] Stop training when the change in validation set loss is less than 5% over several consecutive training cycles. If the loss is unstable, continue training and adjust model parameters.

[0113] Preferably, based on the above scheme, in step S5, the specific steps are:

[0114] S5, model evaluation and testing further includes: S51 using the test set to evaluate the trained model; S52 analyzing the diagnosis results of the model to check whether there is any misclassification or missed diagnosis; S53 deploying the model.

[0115] Preferably, based on the above solution, in step S5,

[0116] S51, using the test set to evaluate the trained model further includes: calculating the harmonic mean of F1 and recall, where the F1 value is the harmonic mean of precision and recall, which comprehensively considers the precision and recall of the model;

[0117] Input the samples of the test set into the trained model and calculate the performance indicators of the model on the test set, such as accuracy, recall, F1 value and other parameters;

[0118] Accuracy calculation formula:

[0119]

[0120] Recall calculation formula:

[0121]

[0122] The accuracy formula is:

[0123]

[0124] F1 formula:

[0125]

[0126] According to the evaluation results on the validation set, adjust the model's hyperparameters, such as learning rate, regularization coefficient, and number of neighbors to find the best hyperparameter combination;

[0127] S52, analyzing the diagnosis results of the model to check whether there is any misclassification or missed diagnosis further includes: traversing each sample in the test set, comparing the predicted category of the model with the actual category; if the predicted category is inconsistent with the actual category, it is a misclassification; recording the features of the misclassified sample and the actual category and the predicted category to analyze the cause; for missed diagnosis, checking whether there is a situation where there is actually a fault but the model has not detected the fault;

[0128] If it is found that there are many misclassifications or missed diagnoses, the following three adjustment methods are used: adjust the model hyperparameters, such as increasing the regularization strength, adjusting the learning rate, changing the number of neighbors, etc., retrain the model and evaluate it; perform data enhancement, such as adding more noise, rotating, flipping and other operations to expand the data set, and train the model again to improve the generalization ability of the model; analyze the feature extraction process, which may require improving the feature extraction method or adding new features to better distinguish different fault categories;

[0129] If the performance is stable, the model is deployed and loaded into the local server.

[0130] Preferably, based on the above scheme, in step S52, the specific steps are:

[0131] S6 fault diagnosis and analysis:

[0132] Repeat steps S1 and S2 to process and enhance the real-time data. Use the prototype network model built in S3 to predict the real-time feature vector and output the category with the highest probability as the predicted fault type. If it is found that the prediction result of the model is obviously inconsistent with the actual situation or new data is collected, return to step S3 to re-divide the data set and re-train the model to improve its accuracy and adaptability.

[0133] Compared with the prior art, the pelletizing fan fault diagnosis method of the present invention has the following advantages:

[0134] 1. Diagnostic accuracy and efficiency: The fault diagnosis method of the present invention not only improves the accuracy and efficiency of pellet blower fault diagnosis through the synergistic effect of data enhancement, L2 regularization and prototype network model, but also optimizes the fault detection process through intelligent diagnosis process. The fault diagnosis method of the present invention analyzes the operating status of the pellet blower from different angles, data enhancement enriches the training data, L2 regularization ensures the stability of the model, and the prototype network model accurately captures the fault characteristics. At the same time, intelligent diagnosis ensures the consistency and reliability of the results.

[0135] 2. Prevent model overfitting: The present invention effectively avoids the model overfitting phenomenon caused by small sample data through L2 regularization, ensures that the fault diagnosis model can still accurately identify faults under limited sample data, improves the generalization ability of the model, ensures the reliability of pellet blower fault diagnosis, and prevents misjudgment and missed judgment problems caused by overfitting.

[0136] 3. Improve the ability to capture fault characteristics: The present invention uses the prototype network model to accurately capture the characteristic mode of pellet blower faults. Each diagnosis can timely discover potential fault hazards, prevent the fault from further deteriorating, and reduce the downtime and maintenance costs caused by the fault. At the same time, the application of the prototype network model makes fault diagnosis more intelligent, providing strong support for fast and accurate fault handling.

[0137] 4. Functional improvement and technical foresight: Although the fault diagnosis method of the present invention is in the stage of continuous development and improvement, it has already demonstrated great potential. Data enhancement provides a way to obtain more diverse data in the future, and the continuous optimization of L2 regularization and prototype network models also reserves development space for adapting to pelletizing fan fault diagnosis under different working conditions. This design idea reflects the comprehensive consideration of the foresight and sustainability of fault diagnosis technology, ensuring that the method can adapt to the higher requirements and development trends of fan fault diagnosis in future pelletizing production. BRIEF DESCRIPTION OF THE DRAWINGS

[0138] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0139] In the figure:

[0140] Figure 1 is a fault diagnosis flow chart of the present invention; DETAILED DESCRIPTION

[0141] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0142] It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections.

[0143] In order to simplify the drawings, only the parts related to the present invention are schematically shown in each figure, and they do not represent the actual structure of the product. In addition, in order to simplify the drawings and facilitate understanding, in some figures, only one of the parts with the same structure or function is schematically drawn or marked. In this article, "one" not only means "only one", but also means "more than one".

[0144] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0145] In the embodiments shown in the drawings, the indications of directions (such as up, down, left, right, front and back) used to explain the structure and movement of the various components of the present invention are not absolute but relative. These descriptions are appropriate when these components are in the positions shown in the drawings. If the descriptions of the positions of these components change, the indications of these directions also change accordingly.

[0146] In addition, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0147] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings and other implementation methods can be obtained based on these drawings without creative work.

[0148] See also Figure 1 The present invention provides a method for diagnosing fan faults, comprising the following steps:

[0149] S1, data collection and preprocessing: collect data sets of various operating parameters of the fan, remove outliers and noise interference through data cleaning, and then normalize the data;

[0150] The data sets of various operating parameters of the wind turbine collected in step S1 include: collecting vibration signal data provided by the vibration sensor under known abnormal conditions in normal conditions, collecting temperature data from the temperature sensor, and recording current data from the current sensor.

[0151] Step S11, the detailed steps of data cleaning are: using the box plot method to process the collected data set, treating the data points outside the upper and lower limits of the box plot as outliers, and determining the center position and dispersion of the data by calculating the lower quartile (Q1), median (Q2) and upper quartile (Q3) of the data set. According to the 1.5 times interquartile range (IQR) rule, points below Q1-1.5×IQR or above Q3+1.5×IQR are marked as outliers.

[0152] Step S12: according to the characteristics of the abnormal point, a low-pass filter is used to remove high-frequency components, a high-pass filter is used to remove low-frequency drift or trend items, and a median filter is used to remove the interference of impulse noise.

[0153] Step S13, replacing the marked outlier by Lagrange interpolation method; selecting n normal data points around the outlier as interpolation nodes;

[0154] Construct the Lagrange interpolation polynomial. For the given n interpolation nodes (x0, y0), (x1, y1), ..., (x n ,y n ), the Lagrange interpolation polynomial L(x) is defined as:

[0155]

[0156] Among them, L i (x) is the Lagrangian basis function, defined as:

[0157]

[0158] Calculate the estimated value of the outlier: For the point x corresponding to the outlier, substitute it into the Lagrange interpolation polynomial L(x) for calculation, and calculate each Lagrange basis function L in turn i (x) and then the corresponding y i Multiply them together and add up all the products to get an estimate of the outlier.

[0159] The calculated estimated values ​​of the outliers are used to replace the outliers in the original data, thus making the data more reasonable for subsequent fault diagnosis analysis.

[0160] Step S14, converting the one-dimensional vibration signal into a two-dimensional time-frequency diagram through short-time Fourier transform, the size of the time-frequency diagram is N×M, where N is the resolution of the frequency axis and M is the resolution of the time axis.

[0161] Use Z-score normalization to unify the temperature and current data to the same scale range as additional channels to be spliced ​​with the vibration time-frequency diagram. Z-score normalization converts the data into a distribution with a mean of 0 and a standard deviation of 1. The calculation formula is:

[0162]

[0163] Where x is the original temperature and current data value, μ is the mean of the data, σ is the standard deviation, and x normalized It is the temperature and current data value after normalization and then unified scale.

[0164] Among them, in step S14, a time-frequency diagram of size N×M can be obtained through short-time Fourier transform, where N is the resolution of the frequency axis and M is the resolution of the time axis; the temperature and current data can be expanded into a matrix of the same size as the vibration time-frequency diagram after normalization, each as a channel, and the shape of the input data is N×M×C, where C is the number of channels.

[0165] After normalization, the temperature and current data can be expanded into a matrix of the same size as the vibration time-frequency diagram, each as a channel, and the shape of the input data is N×M×C, where C is the number of channels (vibration time-frequency diagram plus temperature and current channels).

[0166] Through the above data collection and preprocessing steps, high-quality and reliable data can be provided for the subsequent fan fault diagnosis model, thereby improving the accuracy and generalization ability of the model.

[0167] S2 data enhancement: A combination of random sampling and Gaussian noise addition is used to enhance the data set, increase data diversity, and improve the generalization ability of the model;

[0168] The specific S2 steps include:

[0169] S21, set the minimum and maximum sampling duration limits for the time series data set, randomly generate a sampling duration, and randomly select a starting point within the total duration range, extract the data segment of the corresponding duration starting from this point, and obtain a new sequence with random length and similar characteristics;

[0170] S22, sampling the feature dimension data to determine the total number of feature dimensions of the data, randomly selecting a portion of the number of feature dimensions and randomly selecting feature dimensions therefrom to form a new feature subset;

[0171] The detailed steps of step S21 are as follows: the total duration of the original time series data is T, and a minimum sampling duration t-min and a maximum sampling duration t-max are set. A sampling duration t is randomly generated, satisfying t-min≤t≤t-max. A starting time point s is randomly selected so that s+t does not exceed the total duration T. Data of a continuous time period starting from time point s and having a length of t is extracted as the sampling result. In this way, a new time series segment is obtained, whose length is random and has similar characteristics to the original time series but is not completely the same.

[0172] S23, Gaussian noise addition includes: determining the mean and standard deviation of Gaussian noise, independently generating random noise values ​​that obey the Gaussian distribution of the mean and standard deviation dui on each data point in the original data, and adding them to the original data points, and obtaining new enhanced data after point-by-point operation. This combination method can effectively increase the diversity and robustness of data and improve the generalization ability of the model when facing different data situations.

[0173] S3 builds a prototype network model: After feature extraction through a convolutional neural network, the data set is divided into a training set, a validation set, and a test set. The training set is divided into a support set and a query set. The feature prototype vector of each fault category in the support set is calculated. The distance between the query sample and the prototype is calculated through a distance metric, and the result is predicted through classification decision.

[0174] Through the feature extraction method of convolutional neural network, the enhanced data set is input into the feature extractor to extract high-dimensional feature vectors; for each fault category, the average value of its feature vector is calculated as the prototype of the fault category; for the query sample, the distance between its feature vector and the prototype of each fault category is calculated, and the calculated distance is converted into probability distribution using the softmax function, and the category with the highest probability is selected as the predicted category;

[0175] The detailed steps of S3 step include:

[0176] S311, using a ConvNeXt network of a standard convolution module to perform architecture design, extracting features of the time-frequency data set; using the N×M×C time-frequency graph after data enhancement as an input layer;

[0177] S312, the first convolution layer selects a 3×3 convolution kernel, sets 32 convolution kernels to extract different basic features, adopts the "same" padding method and a step size of 1, uses the ReLU activation function, and the second and third convolution layers have the same parameters as the first convolution layer. The output feature map shape is N×M×32;

[0178] S313, add pooling windows after the first and second convolutional layers respectively, use a maximum pooling window of 2×2, and a step size of 2. Perform a pooling operation on the feature map output by the convolutional layer, reduce the size of the feature map by half, and the shape of the output feature map becomes

[0179] The second convolution layer also uses a 3×3 convolution kernel, sets 64 convolution kernels, the padding method and step size remain unchanged, the activation function is still ReLU, and the input is the output of the previous layer. The output feature map shape is,

[0180] A maximum pooling window of 2×2 is used with a step size of 2. Pooling is performed on the feature map output by the convolutional layer, reducing the size of the feature map by half, and the shape of the output feature map becomes

[0181] The third convolution layer still uses a 3×3 convolution kernel, sets 128 convolution kernels, "same" padding and step size to 1, the activation function is ReLU, and the input is the output of the previous layer. The output feature map shape is

[0182] S314, in the two-dimensional convolution, the calculation formula for each element (i, j, k) in the output feature map (where i and j are spatial position indexes and k is a channel index) is:

[0183]

[0184] Among them, I i+m，j+n，l Represents the element value at the corresponding position in the input feature map, K m，n，l，k Represents the weight value of the corresponding position in the convolution kernel.

[0185] S315, add BatchNormalization after each convolution layer to speed up training and improve model stability;

[0186] S316, after a series of convolutional layers and pooling layers, the feature map of shape N×M×C is obtained, where N1 and M1 are the spatial dimensions of the feature map, and C is the number of channels. The feature map is flattened into a one-dimensional vector with a length of N1×M1×C;

[0187] S317, the first fully connected layer has L neurons, which are linearly transformed by a weight matrix W1 (size is N1×M1×C×L) and a bias vector b1 (length is L) to obtain the output of the fully connected layer; the formula is:

[0188] z=f(W1x+b1);

[0189] Among them, x is the flattened feature vector, and f is the ReLU function;

[0190] S318, the second fully connected layer has K neurons) (K is the number of fault classes), and a weight matrix W2 (size is N1×M1×C×K) and a bias vector b2 (length is K) are linearly transformed to obtain the output of the fully connected layer; the formula is:

[0191] y=f(W2z+b2);

[0192] Where z is the output of the first fully connected layer and f is the Softmax function.

[0193] S32 divides the data set: divide the feature vector obtained in S31 into 70% training set, 20% validation set, and 10% test set ratios; use the random division method to ensure that each category has a certain proportion distribution in different data sets; divide the training set into 75% support set and 25% query set.

[0194] The S321 training set is used for the model to learn data patterns and rules, adjust parameters to establish mapping relationships; the validation set is used to evaluate performance during training, help adjust hyperparameters, and determine overfitting or underfitting; the test set is used to evaluate the final performance after training and measure generalization ability; the support set (part of the training set) is used to calculate the category prototype; the query set is used to test the model's ability to classify new samples, and to make classification decisions based on the distance to the category prototype obtained based on the support set.

[0195] S33 Computing Prototype:

[0196] Initialization phase of computing prototypes: Divide the samples in the support set according to the fault categories. Label the K fault categories as C1, C2, …, C K For each fault category C k , randomly select a small number of initial samples to form an initial sample set:

[0197]

[0198] Where n is the number of initial samples. Calculate the initial prototype for each fault category That is, the average value of the initial sample feature vector of this category; the calculation formula is:

[0199]

[0200] Prototype calculation batch update phase: As new samples are added, the prototype is gradually updated using batch processing. The current iteration number is t, and the newly added batch of samples is

[0201]

[0202] The existing prototype is Calculate the total number of samples processed

[0203]

[0204] Its i is the number of new samples added in the i-th iteration. Update the prototype after iteration The calculation formula is:

[0205]

[0206] S34 distance metric

[0207] The faults are measured and classified by the nearest neighbor classification (K-NN); the sample feature vector y of the query set is calculated using the Euclidean distance method k，n With prototype set The distance between two feature vectors x = (x1, x2, ..., x n ) and y=(y1,y2,…,y n ), the Euclidean distance calculation formula is:

[0208]

[0209] n is the dimension of the feature vector.

[0210] S35 Decision Classification

[0211] According to the calculated distance, find the k training samples closest to it, which are called y k，n The k nearest neighbors of . Count the frequency of occurrence of each category in these k nearest neighbor samples.

[0212] The distance is negatively exponentially converted to a probability by the softmax function. For a sample x, the probability of belonging to the i-th fault category is calculated as

[0213]

[0214] Where N is the total number of fault categories

[0215] y k，n The class with the highest probability is considered to best represent the characteristics of the new sample.

[0216] S4 model training:

[0217] Add an L2 regularization term to the loss function of the prototype network model and determine the L2 regularization coefficient to balance the model's fitting ability and generalization ability;

[0218] Step S4 further includes: S41 calculating a basic loss function; S42 adding an L2 regularization term;

[0219] S41 calculates the basic loss function further including:

[0220] Use cross entropy loss and other functions as the basic loss function. The true category of the sample is y, and the predicted category is The cross entropy loss function is

[0221]

[0222] If the sample belongs to the jth fault category, then the true category vector y = 0, 0, ..., 1, ..., 0), where the jth position is 1 and the rest are 0

[0223] S42 adding the L2 regularization term further includes: adding to the new loss function through the L2 regularization term formula:

[0224] The L2 regularization term is calculated based on the model parameters. The formula for the L2 regularization term is λ∑ w ∥w∥ 2 , where w is the parameter of the model and λ is the regularization coefficient; the L2 regularization term is added to the basic loss function to obtain a new loss function:

[0225]

[0226] S43 Back Propagation

[0227] S431 calculates the gradient of the loss function with respect to the model parameters

[0228] Calculate the predicted probability of the sample belonging to each fault category, the distance function, and the gradient of the L2 regular term to the model parameters, so as to use the optimizer to update the model parameters and improve the performance of the model and the accuracy of fault diagnosis. Use the chain rule to gradually calculate the gradient of the loss function to the model parameters. For each parameter, calculate

[0229] First, calculate the gradient of the cross entropy loss function to the predicted probability

[0230] Where L is the cross entropy loss function, y i is the true category label of the sample, is the predicted probability that the sample belongs to the i-th fault category.

[0231] Then according to the calculation formula of predicted probability, where d(x,c i ) is the distance between the sample and the prototype of the i-th category, and N is the total number of fault categories.

[0232] Further calculation where d(x,c i ) is the distance between the sample and the prototype of the i-th category, and N is the total number of fault categories.

[0233] Combined with the distance function d(x,c i ) with respect to the derivative of the parameters, and obtain the gradient of the predicted probability of each fault category with respect to the model parameters

[0234]

[0235] The formula for the L2 regularization term is λ∑m ＝1 ∥w∥ 2 , where w is the parameter of the model and λ is the regularization coefficient. The gradient of the L2 regularization term with respect to the parameter is calculated as 2λw.

[0236] Finally, the total gradient is

[0237] S432 Update model parameters using optimizer

[0238] S4321 Initialize Adam optimizer parameters

[0239] Set the learning rate α, which is usually between 0.001 and 0.01. The exponential decay rate β1 of the first-order moment estimate is 0.9; the exponential decay rate β2 of the second-order moment estimate is 0.999; set a very small constant ∈ to 1e-8 to prevent the denominator from being zero.

[0240] For each model parameter w:

[0241] Initialize the first-order moment estimate m0=0 and the second-order moment estimate v0=0.

[0242] At iteration t:

[0243] Compute the first moment estimate:

[0244] m t The first-order moment estimate represents the exponentially weighted average of the gradient. It reflects the overall trend of past gradients and helps stabilize the direction of parameter updates during training.

[0245] Compute the second moment estimate:

[0246] v t The second-order moment estimate represents the exponentially weighted average of the square of the gradient. It is used to estimate the variance of the gradient and help adjust the step size of the parameter update.

[0247] Compute the corrected first moment estimate:

[0248] Compute the corrected second moment estimate:

[0249] Update parameters:

[0250] By repeating the above steps, the Adam optimizer is used to update the model parameters in each iteration, so that the model is gradually optimized during the training process.

[0251] S4322 Regularly calculate validation set loss

[0252] During model training, the loss value is calculated on the validation set in each training cycle. The loss function is the cross entropy loss plus the L2 regularization term, that is, L total =L+λ∑ w ||w|| 2 , where L is the cross entropy loss function, λ is the regularization coefficient, and w is the model parameter.

[0253] For each sample in the validation set, it is input into the model for forward propagation, the prediction result is calculated, and then the loss value is calculated based on the true label. Finally, the loss values ​​of all samples are averaged to obtain the average loss of the validation set.

[0254] Stop training when the change in validation set loss is less than 5% over several consecutive training cycles. If the loss is unstable, continue training and adjust model parameters.

[0255] S5 model testing and evaluation

[0256] Input the test set data into the trained model for prediction. Based on the prediction results and the true labels, calculate the accuracy, recall, F1 value, etc. for evaluation. These indicators can measure the performance of the model on the test set. If the performance is stable, the model is deployed.

[0257] S5 model evaluation and testing further includes: S51 using the test set to evaluate the trained model, S52 analyzing the diagnosis results of the model to check whether there is any misclassification or missed diagnosis; S53 deploying the model.

[0258] S51 further evaluates the trained model using the test set, including: calculating the harmonic mean of F1 and recall. The F1 value is the harmonic mean of precision and recall, which comprehensively considers the precision and recall of the model;

[0259] Input the samples of the test set into the trained model and calculate the performance indicators of the model on the test set, such as accuracy, recall, F1 value and other parameters.

[0260] Accuracy calculation formula:

[0261]

[0262] Recall calculation formula:

[0263]

[0264] The accuracy formula is:

[0265]

[0266] F1 formula:

[0267]

[0268] According to the evaluation results on the validation set, adjust the model's hyperparameters, such as learning rate, regularization coefficient, and number of neighbors to find the best hyperparameter combination;

[0269] S52 analyzes the diagnosis results of the model to check whether there is any misclassification or missed diagnosis, further comprising: traversing each sample in the test set, comparing the predicted category of the model with the actual category; if the predicted category is inconsistent with the actual category, it is a misclassification; recording the features of the misclassified sample and the actual category and the predicted category to analyze the cause; for missed diagnosis, checking whether there is a situation where there is actually a fault but the model has not detected the fault;

[0270] If many misclassifications or missed diagnoses are found, the following three adjustment methods are used: adjust the model hyperparameters, such as increasing the regularization strength, adjusting the learning rate, changing the number of neighbors, etc., retrain the model and evaluate it; perform data enhancement, such as adding more noise, rotating, flipping and other operations to expand the data set, and train the model again to improve the generalization ability of the model; analyze the feature extraction process, it may be necessary to improve the feature extraction method or add new features to better distinguish different fault categories.

[0271] If the performance is stable, the model is deployed and loaded into the local server.

[0272] S6 fault diagnosis and analysis

[0273] Repeat steps S1 and S2 to process and enhance the real-time data, predict the real-time feature vector through the prototype network model built in S3, and output the category with the highest probability as the predicted fault type. If it is found that the prediction result of the model is obviously inconsistent with the actual situation, or new data is collected, return to step S3 to re-divide the data set and re-train the model to improve its accuracy and adaptability.

[0274] Finally, the method of the present application is only a preferred implementation scheme and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A pelletizing fan fault diagnosis method, characterized in that: The following steps are involved: Step S1, collecting historical operating parameters of the pelletizing fan to form a data set, removing outliers and noise interference in the data set through data cleaning, and then normalizing the data; Step S2, using a combination of random sampling and Gaussian noise addition to perform data enhancement on the data set; Step S3, after feature extraction through convolutional neural network, the data set is divided into training set, validation set and test set, the training set is divided into support set and query set, the feature prototype vector of each fault category in the support set is calculated, and then the distance between the query sample and the prototype is calculated through distance metric, and the result is predicted through classification decision; Step S4, adding an L2 regularization term to the loss function of the prototype network model, determining the L2 regularization coefficient to balance the fitting ability and generalization ability of the prototype network model, adjusting the model parameters through back propagation, regularly calculating the validation set loss, and judging whether the training termination condition is met; Step S5, input the test set data into the trained balanced prototype network model for prediction, calculate the accuracy, recall rate, and F1 value based on the prediction results and the true labels, and evaluate them. If the performance is stable, the model is deployed; Step S6, repeating steps S1 and S2 to process and enhance the real-time data, predicting the real-time feature vector through the prototype network model constructed in S3, and outputting the category with the highest probability as the predicted fault type; If it is found that the prediction results of the model are obviously inconsistent with the actual situation and new data is collected, return to step S3 to re-divide the data set and retrain the model.

2. A pelletizing blower fault diagnosis method according to claim 1, characterized in that: In step S1, the specific steps are: Step S11, the collected parameters include: the vibration sensor provides vibration signal data, the temperature sensor collects temperature data, and the current sensor records current parameters under known abnormal conditions in the normal state; the box plot method is used to process the collected data set, and the data points outside the upper and lower limits of the box plot are regarded as abnormal values; Step S12, according to the characteristics of the abnormal value, select low-pass filtering, high-pass filtering and median filtering to perform noise reduction processing; Step S13, replacing the marked outliers by Lagrange interpolation method; Step S14, convert the one-dimensional vibration signal into a two-dimensional time-frequency diagram through short-time Fourier transform, use Z-score normalization to unify the temperature and current data into the same scale range as additional channels to be spliced ​​with the vibration time-frequency diagram, and Z-score normalization converts the data into a distribution with a mean of 0 and a standard deviation of 1. The calculation formula is: Where x is the original temperature and current data value, μ is the mean of the data, σ is the standard deviation, and x normalized It is the temperature and current data value after normalization and then unified scale; Among them, in step S14, a time-frequency diagram of size N×M is obtained through short-time Fourier transform, where N is the resolution of the frequency axis and M is the resolution of the time axis; the temperature and current data can be expanded into a matrix of the same size as the vibration time-frequency diagram after normalization, each as a channel, and the shape of the input data is N×M×C, where C is the number of channels.

3. A pelletizing blower fault diagnosis method according to claim 1, characterized in that: In step S2, the specific steps are: S21, set the minimum and maximum sampling duration limits for the time series data set, randomly generate a sampling duration, and randomly select a starting point within the total duration range, extract the data segment of the corresponding duration starting from this point, and obtain a new sequence with random length and similar characteristics; S22, sampling the feature dimension data to determine the total number of feature dimensions of the data, randomly selecting a portion of the number of feature dimensions, and randomly selecting feature dimensions therefrom to form a new feature subset; S23, determine the mean and standard deviation of the Gaussian noise, independently generate a random noise value that obeys the Gaussian distribution of the mean and standard deviation for each data point in the original data, and add it to the original data point, and obtain new enhanced data after point-by-point operation.

4. A pelletizing blower fault diagnosis method according to claim 1, characterized in that: In step S3, the specific steps are: S311, using a ConvNeXt network of a standard convolution module to perform architecture design, extracting features of the time-frequency data set; using the N×M×C time-frequency graph after data enhancement as an input layer; S312, the first convolution layer selects a 3×3 convolution kernel, sets 32 convolution kernels to extract different basic features, adopts the "same" padding method and a step size of 1, uses the ReLU activation function, and the second and third convolution layers have the same parameters as the first convolution layer. The output feature map shape is N×M×32; S313, add pooling windows after the first and second convolutional layers respectively, use a 2×2 maximum pooling window with a step size of 2; perform a pooling operation on the feature map output by the convolutional layer, reduce the size of the feature map by half, and the shape of the output feature map becomes The second convolution layer also uses a 3×3 convolution kernel, sets 64 convolution kernels, the padding method and step size remain unchanged, the activation function is still ReLU, and the input is the output of the previous layer. The output feature map shape is, Use a 2×2 maximum pooling window with a step size of 2; perform a pooling operation on the feature map output by the convolutional layer, reduce the size of the feature map by half, and the shape of the output feature map becomes The third convolution layer still uses a 3×3 convolution kernel, sets 128 convolution kernels, "same" padding and step size to 1, the activation function is ReLU, and the input is the output of the previous layer. The output feature map shape is S314, in the two-dimensional convolution, the calculation formula for each element (i, j, k) in the output feature map (where i and j are spatial position indexes and k is a channel index) is: Among them, I i+m，j+n，l Represents the element value at the corresponding position in the input feature map, K m，n，l，k Represents the weight value of the corresponding position in the convolution kernel; S315, add BatchNormalization after each convolution layer to speed up training and improve model stability; S316, after a series of convolutional layers and pooling layers, the feature map of shape N×M×C is obtained, where N1 and M1 are the spatial sizes of the feature map, and C is the number of channels; the feature map is flattened into a one-dimensional vector with a length of N1×M1×C; S317, the first fully connected layer has L neurons, which are linearly transformed by a weight matrix W1 (size is N1×M1×C×L) and a bias vector b1 (length is L) to obtain the output of the fully connected layer; the formula is: z=f(W1x+b1); Among them, x is the flattened feature vector, and f is the ReLU function; S318, the second fully connected layer has K neurons) (K is the number of fault classes), and a weight matrix W2 (size is N1×M1×C×K) and a bias vector b2 (length is K) are linearly transformed to obtain the output of the fully connected layer; the formula is: y=f(W2z+b2); Where z is the output of the first fully connected layer, and f is the Softmax function; S32, divide the feature vector obtained in S31 into 70% training set, 20% validation set, and 10% test set ratios; use the random division method to ensure that each category has a certain proportion distribution in different data sets; divide the training set into 75% support set and 25% query set; S321, the training set is used for the model to learn data patterns and rules, adjust parameters to establish mapping relationships; the validation set is used to evaluate performance during training, help adjust hyperparameters, and determine overfitting or underfitting; the test set is used to evaluate the final performance after training and measure generalization ability; the support set is used to calculate the category prototype; the query set is used to test the model's ability to classify new samples, and to make classification decisions based on the distance to the category prototype obtained based on the support set. S33, calculation prototype: Computation prototype initialization phase: divide the samples in the support set according to the fault category; mark the K fault categories as C1, C2, ..., C K ; For each fault category C k (k=1,2,…,K), represents the kth fault category, K is the total number of fault categories, and a small number of initial samples are randomly selected to form the initial sample set: Where n is the number of initial samples; calculate the initial prototype of each fault category That is, the average value of the initial sample feature vector of this category; the calculation formula is: n represents the number of initial samples selected for each fault category in the initialization phase; x k，i The feature vector of the i-th sample in the k-th fault category in the initial sample set is calculated. Prototype batch update phase: As new samples are continuously added, the prototype is gradually updated in a batch processing manner; the current number of iterations is t, and the newly added batch of samples is: represents a batch of samples newly added to the kth fault category at the th iteration; y k，m represents the feature vector of the mth sample newly added to the sample set of the kth fault category at the tth iteration; m represents the number of samples newly added to the kth fault category at the tth iteration, that is, the sample set The number of samples in ; The existing prototype is Calculate the total number of samples processed; m i is the number of new samples added in the i-th iteration; the prototype is updated after iteration The calculation formula is: S34 distance metric: The faults are measured and classified by the nearest neighbor classification (K-NN); the sample feature vector y of the query set is calculated using the Euclidean distance method k，n With prototype set The distance between two feature vectors y = (y1, y2, ..., y n ) and x=(x1,x2,…,x n ), where x and y are any two feature vectors in the query set and prototype set, n is the dimension of the feature vector, and y is the k，n Represents the feature vector of the nth sample in the kth fault category in the query set; The Euclidean distance calculation formula is: n represents the dimension of the feature vector; S35 Decision Classification: According to the calculated distance, find the k training samples closest to it, which are called y k，n The k nearest neighbors of ; count the frequency of occurrence of each category in these k nearest neighbor samples; Take the negative exponent of the distance and convert it into probability through the softmax function; for sample x, the probability calculation formula of belonging to the i-th fault category is: Where N is the total number of fault categories; d(x,c i ) represents the relationship between sample x and the i-th fault category prototype c i The distance between c i Represents the prototype vector of the i-th fault category; y k，n The class with the highest probability is considered to best represent the characteristics of the new sample.

5. A pelletizing blower fault diagnosis method as claimed in claim 4, characterized in that: In step S4, the specific steps are: S41, using functions such as cross entropy loss as the basic loss function; The true category of the sample is y, and the predicted category is The cross entropy loss function is: If the sample belongs to the jth fault category, then the true category vector y = (0, 0, ..., 1, ..., 0), where the jth position is 1 and the rest of the positions are 0; S42, calculate the L2 regularization term based on the model parameters. The formula of the L2 regularization term is: λ∑ w ||w|| 2 , where w is the parameter of the model and λ is the regularization coefficient; S45, add the L2 regularization term to the basic loss function to obtain a new loss function: L total =L+λ∑ w ||w|| 2 ; S43 back propagation; S431 calculates the gradient of the loss function with respect to the model parameters: Calculate the predicted probability of the sample belonging to each fault category, the distance function, and the gradient of the L2 regularization term to the model parameters, so as to use the optimizer to update the model parameters and improve the performance of the model and the accuracy of fault diagnosis; Use the chain rule to gradually calculate the gradient of the loss function with respect to the model parameters. For each parameter, calculate First, calculate the gradient of the cross entropy loss function with respect to the predicted probability Among them, L is the cross entropy loss function, y i is the true category label of the sample, is the predicted probability that the sample belongs to the i-th fault category; According to the calculation formula of predicted probability, where d(x,c i ) is the distance between the sample and the prototype of the i-th category, and N is the total number of fault categories; calculate where d(x,c i ) is the distance between the sample and the prototype of the i-th category, and N is the total number of fault categories: Combined with the distance function d(x,c i ) with respect to the derivative of the parameters, we get the gradient of the predicted probability of each fault category with respect to the model parameters: The formula for the L2 regularization term is λ∑ m＝1 ||w|| 2 , where w is the parameter of the model and λ is the regularization coefficient: The gradient of the L2 regularization term with respect to the parameter is calculated as 2λw: Finally, the total gradient is S432, use the optimizer to update the model parameters: S4321, initialize Adam optimizer parameters: Set the learning rate α, usually between 0.001 and 0.01, the exponential decay rate β1 of the first-order moment estimate is 0.9; the exponential decay rate β2 of the second-order moment estimate is 0.999; set a very small constant ∈ to 1e-8 to prevent the denominator from being zero; For each model parameter w; Initialize the first-order moment estimate m0=0 and the second-order moment estimate v0=0; At iteration t: Compute the first moment estimate: m t First-order moment estimate, representing the exponentially weighted average of the gradient; Compute the second moment estimate: v t The second-order moment estimate represents the exponentially weighted average of the squared gradient, which is used to estimate the variance of the gradient and help adjust the step size of the parameter update; Compute the corrected first moment estimate: Compute the corrected second moment estimate: Update parameters: By repeating the above steps, the Adam optimizer is used to update the model parameters in each iteration, so that the model is gradually optimized during the training process; S4322, regularly calculate the validation set loss: During model training, the loss value is calculated on the validation set in each training cycle. The loss function is the cross entropy loss plus the L2 regularization term, that is, L total =L+λ∑ w ||w|| 2 , where L is the cross entropy loss function, λ is the regularization coefficient, and w is the model parameter: For each sample in the validation set, input it into the model for forward propagation, calculate the prediction result, and then calculate the loss value based on the true label. Finally, the loss values ​​of all samples are averaged to get the average loss of the validation set. Stop training when the change in validation set loss is less than 5% over several consecutive training cycles. If the loss is unstable, continue training and adjust model parameters.

6. A pelletizing blower fault diagnosis method according to claim 1, characterized in that: In step S5, the specific steps are: S5, model evaluation and testing further includes: S51 using the test set to evaluate the trained model; S52 analyzing the diagnosis results of the model to check whether there is any misclassification or missed diagnosis; S53 deploying the model.

7. A pelletizing blower fault diagnosis method as claimed in claim 6, characterized in that: In the step S5, S51, using the test set to evaluate the trained model further includes: calculating the harmonic mean of F1 and recall, where the F1 value is the harmonic mean of precision and recall, which comprehensively considers the precision and recall of the model; Input the samples of the test set into the trained model and calculate the performance indicators of the model on the test set, such as accuracy, recall, F1 value and other parameters; Accuracy calculation formula: Recall calculation formula: The accuracy formula is: F1 formula: According to the evaluation results on the validation set, adjust the model's hyperparameters, such as learning rate, regularization coefficient, and number of neighbors to find the best hyperparameter combination; S52, analyzing the diagnosis results of the model to check whether there is any misclassification or missed diagnosis further includes: traversing each sample in the test set, comparing the predicted category of the model with the actual category; if the predicted category is inconsistent with the actual category, it is a misclassification; recording the features of the misclassified sample and the actual category and the predicted category to analyze the cause; for missed diagnosis, checking whether there is a situation where there is actually a fault but the model has not detected the fault; If the performance is stable, the model is deployed and loaded into the local server.

8. A pelletizing blower fault diagnosis method as claimed in claim 7, characterized in that: S6 fault diagnosis and analysis, the specific steps are: Repeat steps S1 and S2 to process and enhance the real-time data. Use the prototype network model built in S3 to predict the real-time feature vector and output the category with the highest probability as the predicted fault type. If it is found that the prediction results of the model are obviously inconsistent with the actual situation, or new data is collected, etc., return to step S3 to re-divide the data set and re-train the model to improve its accuracy and adaptability.

Images