Multi-source signal tool wear loss prediction method and system based on one-dimensional residual network

Abstract

The invention belongs to the technical field of industrial monitoring, and particularly discloses a multi-source signal tool wear loss prediction method and system based on a one-dimensional residual network, and the method and system achieve the whole-process monitoring from signal collection to tool wear loss prediction. The signal acquisition module is arranged, and the precision of signal analysis is improved through multi-sensor signal conversion and wavelet transformation; a data processing module is arranged, the problems of data imbalance and small samples are solved through data balance operation, and the effectiveness of data processing is improved; a complex relation processing module is arranged, and a one-dimensional residual network is constructed to process a complex relation among multi-source signals, so that comprehensive signal analysis is realized; a feature enhancement module is arranged, and a multi-head attention mechanism is introduced, so that the feature extraction capability is enhanced, and the accuracy of tool wear loss prediction is improved; and a task execution module is arranged, a one-dimensional residual network and a multi-head attention mechanism model are combined, the tool abrasion loss is predicted, and the refined analysis capacity of fault detection is improved.

Description

Technical Field

[0001] The invention relates to the technical field of industrial monitoring, and in particular to a method and system for predicting tool wear using a multi-source signal based on a one-dimensional residual network. Background Art

[0002] High-precision CNC machine tools are one of the important equipments that are indispensable in the development of advanced manufacturing industry. The tool of CNC machine tools will inevitably wear out due to the intense friction between the tool and the parts during the cutting process. The wear state of the tool directly affects the surface processing quality of the parts. Once the tool is severely worn, the processed parts will be unqualified or even scrapped, causing significant losses to the enterprise. Therefore, it is very necessary to monitor the tool wear state of CNC machine tools in real time during the processing. The current monitoring method mainly predicts the tool wear value by using the vibration signal generated by the processing process. This method has a single signal source and is easily affected by noise and environmental disturbances in a complex cutting environment. It has great limitations and inaccuracy. At the same time, most of the collected data are data on the normal wear state of the tool, and data on severe wear are scarce. This unbalanced data under a single signal causes insufficient prediction accuracy, which is not conducive to accurately predicting the tool wear value. With the continuous development of deep learning theory, the multi-signal fusion method has shown obvious advantages in feature extraction and pattern recognition, and has great application potential for tool wear prediction under different working conditions of unbalanced data sets. Summary of the invention

[0003] The purpose of the present invention is to solve the technical problems existing in the background technology. To this end, a multi-source signal tool wear prediction method and system based on a one-dimensional residual network are provided. The accuracy and reliability of tool wear prediction are improved by fusing multi-sensor signals and introducing complex relationship processing and feature enhancement technology.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] The tool wear prediction method based on multi-source signals of one-dimensional residual network includes:

[0006] Step 1: Build a signal acquisition module to convert the collected multi-sensor signals into a spectrum diagram and perform wavelet transform;

[0007] Step 2: Build a data processing module to balance the data and solve problems such as data imbalance and small sample data;

[0008] Step 3: Construct a complex relationship processing module to build a one-dimensional residual network structure to process the complex relationship between multi-source signals;

[0009] Step 4: Build a feature enhancement module and introduce a multi-head attention mechanism to enhance the model’s feature extraction capability;

[0010] Step 5: Build a task execution module and use a model that combines a one-dimensional residual network and a multi-head attention mechanism to perform tool wear prediction tasks.

[0011] The following is a technical solution further defined in the method of the present invention. The construction process of the signal acquisition module in step 1 includes:

[0012] Step 11: Collect signals through multiple sensors, filter the audio information, and perform fast Fourier transform (FFT) to obtain a spectrum diagram;

[0013] Step 12: Use the db4 wavelet function to perform wavelet transform on the spectrum graph, so that the signal can be decomposed and analyzed at different scales and a wavelet scale graph can be obtained;

[0014] Wherein, the wavelet transform in step 12 includes:

[0015] Step 121: performing signal decomposition, firstly performing multi-scale decomposition on the signal to be processed, the signal is decomposed into components at multiple different scales, wherein the low-frequency component reflects the overall trend of the signal, and the high-frequency part reflects the details and mutations of the signal;

[0016] Step 122: using wavelet filters, one is a low-pass filter and the other is a high-pass filter, the two sets of filters are determined by the coefficients of the db4 wavelet function, the low-pass filter is used to extract the low-frequency components of the signal, and the high-pass filter is used to extract the high-frequency components of the signal;

[0017] Step 123: convolution operation, convolving the signal with low-pass and high-pass filters to obtain low-frequency and high-frequency coefficients, where the low-frequency coefficient represents the low-frequency component of the signal, and the high-frequency coefficient represents the high-frequency component of the signal;

[0018] Step 124: downsampling, performing a downsampling operation on the convolution result to reduce the frequency so that the scale of the signal is reduced by half, which can also be called subsampling;

[0019] Step 125: Repeat steps 121-125 to decompose the signal into components of lower scales until the required number of decomposition levels is reached; when necessary, the decomposed signal can be reconstructed to restore the original signal. This operation usually requires the use of wavelet coefficients and inverse wavelet filters for inverse transform operations.

[0020] The following is a technical solution further defined in the method of the present invention. The construction process of the data processing module in step 2 includes:

[0021] Step 21: Data preparation, ensuring that the multi-sensor signal dataset contains various characteristics and distribution of the original data;

[0022] Step 22: Design an autoencoder and construct the encoder and decoder parts of the autoencoder. The encoder maps the original data to a low-dimensional representation, and the decoder attempts to reconstruct the original data from the low-dimensional representation.

[0023] Step 23: Construct a generator and a discriminator to adapt to the characteristics of sensor signal data. The generator is responsible for generating sample data from random noise, and the discriminator tries to distinguish between the generated fake samples and real samples.

[0024] Step 24: Define the loss function, which consists of the reconstruction loss of the autoencoder and the adversarial loss of the generative adversarial network. The reconstruction loss is used to measure the accuracy of data reconstruction, and the adversarial loss is used to encourage the generator to generate real data samples.

[0025] Step 25: Network training, optimize the model by alternately training the autoencoder and the generative adversarial network, first use the reconstruction loss of the autoencoder to train the autoencoder, and then use the adversarial loss to train the generator and discriminator; this process requires multiple iterations until the model converges or reaches the predetermined training rounds.

[0026] Step 26: Evaluate the data samples generated by the generator to check whether they conform to the distribution characteristics of the original data, and then optimize and adjust the model based on the evaluation results, including adjusting the network structure, loss function and training parameters to improve the quality fidelity of the generated results.

[0027] The following is a technical solution further defined by the method of the present invention, wherein the construction process of the autoencoder in step 22 includes:

[0028] Step 221: Define the encoder architecture, select a multi-layer perceptron MLP structure, and use ReLU corrected linear unit as the activation function. The ReLU function can help the model converge better during the training process and avoid the problem of gradient disappearance. During the forward propagation process, the input signal data passes through the various hidden layers of the encoder and is gradually converted into a lower dimensional representation. This process is expressed by the formula:

[0029] h=ReLU(W1x+b1)

[0030] z=W2h+b2

[0031] Where x is the input signal data, h is the hidden layer representation, z is the encoded representation, W1 and W2 are the weight matrices of the hidden layer and the encoding layer, and b1 and b2 are the corresponding bias vectors;

[0032] Step 222: Define the decoder framework. The structure of the decoder, the selection of activation function, and the initialization of the weight matrix are similar to those of the encoder. In the forward propagation process of the decoder, the decoder receives the representation generated by the encoder as input and tries to decode it into the original input data. The forward propagation process of the decoder is expressed by the formula:

[0033] h′=ReLU(W′1x+b′1)

[0034]

[0035] Where h′ is the hidden layer representation of the decoder, is the output of the decoder, representing the reconstructed input data;

[0036] Step 223: Define the loss function and optimizer. The goal of the autoencoder is to minimize the reconstruction error between the input data and the decoder output data. The mean square error (MSE) is used as the loss function. The MSE is defined by the following formula:

[0037]

[0038] Where N is the number of samples, xi is the original input data, is the output data of the decoder;

[0039] Step 224: Back propagation and parameter update, calculate the gradient of the loss with respect to the parameters through back propagation, and update the parameters using the gradient descent optimization algorithm.

[0040] The construction process of the generator and discriminator in step 23 includes:

[0041] Step 231: GAN consists of two parts: a generator and a discriminator. The generator receives a random noise vector z and outputs a generated data sample G(z; θ g ):

[0042]

[0043] Where G is the generator function with parameter θ g , used to map random noise z to generate data samples The discriminator receives real data samples x or generated data samples Through the parameter θ disc Output the probability D(x;θ) that the sample is real data disc )and Where D is the function of the discriminator, and the parameter is θ disc ;

[0044] Step 232: The loss function of the discriminator and the generator is defined. The goal of the discriminator is to maximize the log-likelihood estimate so that the probability of the real sample is close to 1 and the probability of the generated sample is close to 0:

[0045]

[0046] The goal of the generator is to minimize the log-likelihood estimate so that the probability of the generated sample is close to 1 and maximize the probability that the generated sample is considered a real sample by the discriminator:

[0047]

[0048] The following is a technical solution further defined in the method of the present invention. The construction process of network training and definition of loss function in step 24 includes:

[0049] Step 241: Initialization, randomly initialize the parameters θ of the encoder and decoder e ,θ d , the parameters θ of the generator and discriminator g ,θ disc ;

[0050] Step 242: Forward propagation and reconstruction loss calculation of the encoder. The encoder maps the input data x to the latent space representation z:

[0051] z=E(x;θ e )

[0052] The decoder reconstructs the latent space representation z as input data

[0053]

[0054] Calculate the reconstruction loss L recon :

[0055]

[0056] Step 243: GAN forward propagation and adversarial loss calculation, calculate the discriminator loss L D :

[0057]

[0058] Calculate the generator loss L G :

[0059]

[0060] Step 244: Back propagation and parameter update, update the encoder and decoder parameters to minimize the reconstruction loss:

[0061]

[0062]

[0063] Update the discriminator parameters to minimize the discriminator loss:

[0064]

[0065] Update the generator parameters to minimize the generator loss:

[0066]

[0067] Step 245: Iterate training, repeat steps 241-245 until the model converges, each iteration includes an update of the encoder and decoder, and an update of the generator and discriminator.

[0068] The following is a technical solution further defined in the method of the present invention. The construction process of the complex relationship processing module in step 3 includes:

[0069] Step 31: Collect lathe signal data from multiple sensors and use filters to extract and classify signal features;

[0070] Step 32: converting the processed time domain signal into a frequency domain signal through Fourier transform, and further extracting multi-scale features through wavelet transform;

[0071] Step 33: Construct a one-dimensional residual network (1D ResNet) to capture and process the complex relationship between multi-source signals. The residual network introduces skip connections to make the training of the deep network more stable and can effectively integrate and extract the correlation features of multi-source signals.

[0072] In step 31, the process of using a filter to extract and classify signal features includes:

[0073] Step 311: Signal preprocessing, remove noise and highlight the main features of the signal through preprocessing steps. Use a low-pass filter to remove high-frequency noise:

[0074]

[0075] The signals are then normalized to keep their amplitudes within a standard range:

[0076]

[0077] Where μ and σ are the mean and standard deviation of the signal, respectively.

[0078] Step 312: extracting time domain features from the signal, and reflecting the statistical properties of the signal through these features, the time domain features include: maximum value, minimum value, peak-to-peak value, average value, variance, standard deviation, mean square value, kurtosis, skewness, shape factor, peak factor, pulse factor, margin factor, and clearance factor;

[0079] Maximum: The maximum value in the data set, indicating the largest observed value in the data.

[0080] Maximum=max(x1,x2,...,xn)

[0081] Minimum: The smallest value in the data set, representing the smallest observed value in the data.

[0082] Minimum=min(x1,x2,...,xn)

[0083] Peak-to-Peak: The peak-to-peak value is the difference between the maximum and minimum values ​​and represents the range of the data.

[0084] Peak-to-Peak=max(x1,x2,...,xn)-min(x1,x2,...,xn)

[0085] Mean: The mean is the arithmetic average of all observations in a data set and is used to indicate the central tendency of the data.

[0086]

[0087] Variance: Variance is the average of the squares of the differences between the observations and the mean. It indicates the dispersion or spread of the data.

[0088]

[0089] Standard Deviation: Standard deviation is the square root of variance. It measures the dispersion of the data relative to the units of the mean.

[0090]

[0091] Mean Square Value: The mean square value is the average of the squares of all the observations in the data set, then taking the square root. It is often used to represent the root mean square magnitude of a signal or waveform, similar to the RMS value.

[0092]

[0093] Kurtosis: Kurtosis is used to describe the tail and kurtosis of the data distribution. Positive kurtosis means that the data distribution has a heavier tail and more extreme values, while negative kurtosis means that the data distribution has a lighter tail and more data is concentrated around the mean.

[0094] Kurtosis=(1 / n)*∑((xi-Mean) / Standard Deviation)^4-3

[0095] Skewness: Skewness is used to describe the asymmetry of data distribution. Positive skewness means that the data distribution is skewed to the right, negative skewness means that the data distribution is skewed to the left, and zero skewness means that the data distribution is symmetrical.

[0096] Skewness=(1 / n)*∑((xi-Mean) / Standard Deviation)^3

[0097] Form factor: The form factor is used to measure the relationship between the peak value and the RMS value of the signal waveform. It is usually used to analyze the peak nature of the signal. A larger form factor means that the signal has a sharper peak.

[0098] Crest Factor=Peak Value / RMS Value

[0099] Crest Factor: Crest factor is also used to measure the relationship between the peak value and the RMS value of a signal waveform. A larger crest factor means that the signal has a larger peak value.

[0100] Peak Factor=Peak Value / RMS Value

[0101] Pulse factor: The pulse factor is used to describe the pulse nature of the signal, that is, the relationship between the short-term peak value and the root mean square value of the signal. It is usually used to analyze transient signals. A larger pulse factor indicates that the signal contains more transient peaks.

[0102] Pulse Factor=Peak Value / RMS Value

[0103] Margin factor: Margin factor is used to describe the relationship between the peak value and the mean value of a signal. It is often used to analyze the margin of a signal, and a larger margin factor indicates that the signal has a larger peak value relative to the mean value.

[0104] Crest Factor = Peak Value / Mean Value Crest Factor: The crest factor is used to describe the relationship between the signal peak and the signal range. It is often used to analyze the dynamic range of a signal, with a larger crest factor indicating that the signal has a larger peak relative to its range.

[0105] Margin Factor=Peak Value / Range

[0106] In step 32, the process of converting the time domain signal into the frequency domain signal by Fourier transform includes:

[0107] Step 321: Convert the time series signal to the frequency domain through Fourier transform and extract the frequency domain features. The frequency domain feature extraction formula is as follows:

[0108] Fourier transform: The time domain signal is converted into the frequency domain signal through Fourier transform.

[0109]

[0110] Spectral energy: represents the energy distribution of the signal in different frequency bands.

[0111]

[0112] Center Frequency: Indicates the center of gravity of the spectrum.

[0113]

[0114] Spectral entropy: represents the complexity of the spectrum.

[0115]

[0116] in,

[0117] Step 322: extracting time-frequency domain features, combining the time domain and the frequency domain, decomposing the signal into different frequency band energies through wavelet transform, and extracting local features;

[0118]

[0119] Step 323: Combine all extracted features into a feature vector.

[0120] In step 33, the process of constructing the one-dimensional residual network includes:

[0121] Step 331: Input data preparation, using the pre-processed multi-sensor signal data (cutting force signal, vibration signal and acoustic emission signal, etc.) as the input of the network. Assume that the input signal is

[0122] X∈R N×d

[0123] Where N is the time step and d is the dimension of the signal. The purpose of this step is to ensure that the model receives processed high-quality data and is ready for subsequent feature extraction and analysis.

[0124] Step 332: Construct an initial convolution layer to perform initial convolution on the input signal to extract low-level features. The convolution operation captures local timing patterns and frequency features. The specific formula is:

[0125] X conv1 =Conv1D(X, W1, b1)

[0126] Among them, Conv1D represents a one-dimensional convolution operation, W1 and b1 are the convolution kernel weight and bias respectively. The function of the initial convolution layer is to reduce the complexity of the input signal, extract preliminary local features, and lay the foundation for subsequent deep feature extraction.

[0127] Step 333: Construction of residual blocks. Residual blocks are the core part of ResNet. Each residual block includes two convolutional layers and a shortcut connection. This structure allows signals to be directly transmitted through skip connections, thereby alleviating the gradient vanishing problem and facilitating the training of deep networks.

[0128] 1) The first convolutional layer of the residual block:

[0129] X conv2 =ReLU(BatchNorm(Conv1D(X conv1 , W2, b2)))

[0130] Among them, BatchNorm represents the batch normalization operation, and ReLU is the activation function. The role of this convolutional layer is to further extract features from the signal and stabilize the training process through batch normalization.

[0131] 2) The second convolutional layer of the residual block:

[0132] X conv3 =BatchNorm(Conv1D(X conv2 , W3, b3)) The second convolutional layer further processes the extracted features and maintains the stability of the signal through batch normalization.

[0133] 3) Quick connection:

[0134] X res =X conv1 +X conv3

[0135] If the dimensions of the input and output are inconsistent, a linear transformation is required to match the dimensions:

[0136] X res =X conv1 +Conv1D(X conv1 , W skip , b skip )

[0137] The role of the shortcut connection is to directly add the input signal to the output, ensuring the direct transmission of information, reducing gradient disappearance, and improving the training effect of the model.

[0138] 4) Activate output:

[0139] X out=ReLu(X res )

[0140] Through the ReLU activation function, nonlinearity is introduced into the model to improve the expressiveness of the model.

[0141] Repeating multiple residual blocks to build a deep network. The role of these residual blocks is to extract and integrate features layer by layer to capture complex patterns and relationships in the signal.

[0142] Step 334: Global average pooling layer, which reduces feature dimensions and aggregates information. The specific formula is:

[0143]

[0144] The role of the global average pooling layer is to average the features of all time steps to generate a feature vector of fixed size, reducing the dimension while retaining important information.

[0145] Step 335: Construct a fully connected layer and an output layer, input the result of global average pooling into the fully connected layer, and obtain the final output. The specific formula is:

[0146] X fc =ReLU(W fc X gap +b fc )

[0147] Y pred =Softmax(W out X fc +b out )

[0148] Among them, W fc and b fc are the weights and biases of the fully connected layer, W out and b out are the weights and biases of the output layer. The function of the fully connected layer is to perform nonlinear transformation on the pooled features, and the output layer is classified through the Softmax function to generate the final prediction result.

[0149] The following is a technical solution further defined in the method of the present invention. The construction process of the feature enhancement module in step 4 includes:

[0150] Step 41: Preprocess the signal, standardize and normalize the signals collected by multiple sensors to ensure that the dimensions of the signal are on the same scale, which helps the model to better process the input data;

[0151] Step 42: Linear transformation, input the preprocessed signal into the linear layer to generate Query, Key and Value matrices;

[0152] Q=XW Q , K = XW K ,V=XW V

[0153] Where X is the input signal matrix; W Q , W K , W V is a learnable weight matrix; Q, K, V are query, key, and value matrices, respectively.

[0154] Step 43: Calculate the attention score. Calculate the attention score between the query and the key by dot product:

[0155]

[0156] Among them, d k is the dimension of the key, used to scale the dot product result.

[0157] After scaling and normalization, the attention scores are scaled and softmax normalized to ensure the stability and interpretability of the scores. The values ​​are weighted summed using the normalized attention weights to obtain the output of each attention head:

[0158]

[0159] Where h is the number of attention heads. i ,K i ,V i is the query, key, and value matrix of the ith head.

[0160] Step 44: After concatenating the outputs of the attention heads, perform residual connections and layer normalization. Concatenate the outputs of all attention heads and perform linear transformation to generate the final output feature representation:

[0161] MultiHead(Q,K,V)=Concat(head1,...,head h )W O

[0162] Among them: head i =Attention(Q i ,K i ,V i ),W O is the output weight matrix.

[0163] Then the output and input features of the multi-head attention mechanism are residually connected and layer normalized:

[0164] Y=LayerNorm(X+MultiHead(Q, K, V)).

[0165] The following is a technical solution further defined by the method of the present invention. The construction process of the task execution module in step 5 includes:

[0166] Step 51: Combine the one-dimensional residual network and the multi-head attention mechanism;

[0167] Step 52: Define the loss function of the model, use an optimization algorithm to minimize the loss function, and then iteratively train the model until convergence;

[0168] In step 51, the process of combining the one-dimensional residual network and the multi-head attention mechanism includes:

[0169] Step 511: The one-dimensional residual network part is formed by stacking multiple residual blocks in depth to form a deep one-dimensional residual network to enhance the model's modeling ability for complex signals. The residual block is the basic structure of the network. Each residual block contains two or more one-dimensional convolutional layers. Each convolutional layer is followed by a ReLU activation function and batch normalization, and a jump connection is introduced between the input and output.

[0170] y=F(x,{W i})+x

[0171] Among them, F(x,{Wi}) represents the convolution and activation operation in the residual block, x is the input, and y is the output. In the one-dimensional convolution layer, multi-sensor signals are preprocessed and then enter the feature extraction.

[0172] Step 512: Input the output features of the one-dimensional residual network (1DResNet) into the multi-head attention mechanism to enhance the feature representation through the attention mechanism. Perform residual connection and layer normalization on the features output by the attention mechanism to form the final feature representation:

[0173] Y=LayerNorm(X+MulitiHead(Q, K, V)) The following is a technical solution further defined in the method of the present invention, which includes, in step 52:

[0174] Step 521: Define the loss function and use the mean square error as the loss function to measure the difference between the model prediction value and the actual value:

[0175]

[0176] MSE is more sensitive to large errors and can effectively penalize prediction results with large deviations, which helps to improve the accuracy of the model.

[0177] Step 522: Optimize algorithm design and use Adam optimizer, which is an optimization algorithm that combines momentum and RMSProp. It can adaptively adjust the learning rate of each parameter and is suitable for processing sparse gradients and high-dimensional data:

[0178] m t =β1m t-1 +(1-β1)g t

[0179] v t =β2v t-1 +(1-β2)g 2 t

[0180]

[0181] Among them, mt is the first-order moment estimate, vt is the second-order moment estimate, β1, β2 are the exponential decay rates of the momentum term, and g t is the gradient, α is the learning rate, and θt is the model parameter. The Adam optimizer combines the advantages of momentum and RMSProp, can adaptively adjust the learning rate, performs well in processing sparse gradients and high-dimensional data, and can quickly converge to the optimal solution.

[0182] The multi-source signal tool wear prediction system based on one-dimensional residual network is used to implement the multi-source signal tool wear prediction method based on one-dimensional residual network, including:

[0183] A signal acquisition module, used to convert the collected multi-sensor signals into a spectrum diagram and perform wavelet transformation;

[0184] A data processing module, used for performing a balancing operation on the data;

[0185] Complex relationship processing module, used to construct a one-dimensional residual network structure to process complex relationships between multi-source signals;

[0186] Feature enhancement module, which is used to introduce a multi-head attention mechanism to enhance the model's feature extraction capability;

[0187] The task execution module uses a model that combines a one-dimensional residual network and a multi-head attention mechanism to perform tool wear prediction tasks.

[0188] Compared with the prior art, the present invention has the following technical effects:

[0189] The present invention realizes the whole process monitoring from signal acquisition to tool wear prediction; a signal acquisition module is provided to improve the accuracy of signal analysis through multi-sensor signal conversion and wavelet transform; a data processing module is provided to solve the data imbalance and small sample problems through data balancing operation and improve the effectiveness of data processing; a complex relationship processing module is provided to process the complex relationship between multi-source signals through constructing a one-dimensional residual network and realize comprehensive signal analysis; a feature enhancement module is provided to enhance the feature extraction capability and improve the accuracy of tool wear prediction through introducing a multi-head attention mechanism; a task execution module is provided to predict tool wear through a model combining a one-dimensional residual network and a multi-head attention mechanism and improve the detailed analysis capability of fault detection.

[0190] The present invention is further described below in conjunction with the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0191] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art will be briefly introduced below. 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 paying creative work.

[0192] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0193] Figure 2 It is a structural diagram of the AE-GAN network model in the present invention;

[0194] Figure 3 It is a structural diagram of the improved 1DResNet-MHSA model in the present invention;

[0195] Figure 4 This is an example diagram of the data collection process in the present invention. DETAILED DESCRIPTION

[0196] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0197] This embodiment provides a tool wear prediction system based on a multi-source signal of a one-dimensional residual network and a tool wear prediction method based on a multi-source signal of a one-dimensional residual network.

[0198] The system consists of signal acquisition module, data processing module, complex relationship processing module, feature enhancement module and task execution module. The system has developed signal acquisition module, data processing module, complex relationship processing module, feature enhancement module and task execution module. The signal acquisition module converts the collected multi-sensor signals into a spectrum diagram and performs wavelet transform; the data processing module performs a balancing operation on the data to solve problems such as data imbalance and small sample data; the complex relationship processing module handles the complex relationship between multi-source signals by constructing a one-dimensional residual network; the feature enhancement module introduces a multi-head attention mechanism to enhance the feature extraction ability of the model; the task execution module combines a one-dimensional residual network and a multi-head attention mechanism to predict tool wear. It solves the problem that traditional tool wear prediction technology is difficult to extract complex relationships between signals in multi-sensor input and cannot efficiently eliminate highly correlated redundant information, significantly improves the ability to capture complex relationships between multi-sensor signals, and enhances the accuracy and reliability of wear prediction, which has broad engineering application prospects.

[0199] The method includes signal acquisition module construction, data processing module construction, complex relationship processing module construction, feature enhancement module construction and task execution module construction. The specific steps are as follows: Figure 1 As shown:

[0200] S1: Build a signal acquisition module to convert the signals collected by multiple sensors into a spectrum and perform wavelet transform.

[0201] S11, such as Figure 4 As shown in Figure 1, a three-axis dynamometer is installed between the workpiece and the worktable to measure the three-axis cutting force. Three piezoelectric acceleration sensors and an acoustic emission sensor are installed on the workpiece to measure the cutting force signals, vibration signals and acoustic emission signals generated in the X, Y and Z directions during high-speed CNC machine tool cutting.

[0202] S12, the signals collected by the multi-sensors are first filtered to remove noise and interference to ensure the clarity and accuracy of the signals. The filtered signals are subjected to fast Fourier transform (FFT) to convert the time domain signals into frequency domain signals to obtain a spectrum diagram.

[0203] S13, using the db4 wavelet function to perform wavelet transform on the spectrum graph, so that the signal is decomposed and analyzed at different scales to obtain a wavelet scale graph. Specifically, the signal to be processed is decomposed into components at different scales through multi-scale decomposition technology, the low-frequency component reflects the overall trend of the signal, and the high-frequency part reflects the details and mutations of the signal.

[0204] S14, then perform convolution operation on the filtered signal to obtain low-frequency coefficients that represent the overall trend of the signal and high-frequency coefficients that reflect the details and mutations of the signal. Finally, perform downsampling operation on the result to reduce the scale of the signal to reduce the frequency. Repeat the decomposition steps to decompose the signal into the required number of decomposition layers, that is, the target scale.

[0205] S2: Figure 2 As shown in the figure, a data processing module is constructed to solve the problems of small samples and sample imbalance that occur in most offline training tasks.

[0206] S21. Organize the multi-sensor signal data to ensure that the data set contains the corresponding features. Then clean the data, remove outliers and fill in missing values ​​to ensure the integrity and consistency of data points.

[0207] S22. Build an autoencoder module, which includes an encoder part and a decoder part. First, input the high-order input data, which is the feature we have processed, into the input layer of the encoder. After the input layer receives the data, it is passed to the hidden layer for linear transformation and ReLU activation function conversion. The features of the input data are extracted through the hidden layer, and compressed into a low-dimensional representation and passed to the output layer. The output layer of the encoder passes the low-dimensional representation to the decoder. The first layer of the decoder uses linear transformation and ReLU activation function to convert the low-dimensional representation into a hidden representation. Then, the hidden representation is further converted into a high-dimensional representation, and finally the original input is reconstructed through the output layer.

[0208] S23, the generator and discriminator are shown below. A random noise (a vector that conforms to the standard normal distribution) is input into the first hidden layer of the generator. The first hidden layer of the generator converts the random noise into a preliminary hidden representation through linear transformation and ReLU activation function. The hidden representation is further processed by linear transformation and ReLU activation function to generate a more complex hidden representation. Finally, the output layer uses tanh activation function to convert the hidden representation into generated sample data to ensure that the value of the generated data is within a reasonable range. Then input real data to the discriminator for training, and then input the generated data into the discriminator. The first hidden layer of the discriminator converts the input data into a hidden representation through linear transformation and ReLU activation function; the hidden representation is further processed by linear transformation and ReLU activation function to extract deeper features; the output layer uses sigmoid activation function to calculate the probability that the sample is real data. Train the generator and discriminator to improve the quality of generated data through adversarial training. The discriminator tries to distinguish between real data and generated data, while the generator tries to deceive the discriminator so that the generated data is judged as real data.

[0209] S24. The next step is to define the loss functions of the two. The reconstruction loss of the autoencoder is used to measure the difference between the reconstructed data and the original data. The mean square error (MSE) is usually used as the loss function. The loss function of the generative adversarial network includes the discriminator loss and the generator loss. The discriminator loss is used to measure the ability of the discriminator to distinguish between real data and generated data, while the generator loss is used to encourage the generator to generate samples that are closer to real data.

[0210] S25, then continuously optimize the model parameters by alternately training the autoencoder and the generative adversarial network. First, the encoder and decoder are trained using the reconstruction loss of the autoencoder to minimize the reconstruction error. Next, the generator and discriminator are trained using the adversarial loss, and the parameters of the generator and discriminator are alternately optimized by minimizing the discriminator loss and the generator loss until the model converges or reaches the predetermined training rounds. Finally, the data samples generated by the generator are evaluated to check whether they conform to the distribution characteristics of the original data. The model is optimized and adjusted according to the evaluation results to ensure the quality and fidelity of the generated data. If necessary, the structure, training parameters, and loss function of the model can be adjusted to improve the generation results.

[0211] S3: Build a complex relationship processing module that can capture the complex relationships between signals and improve the model's feature extraction capabilities, thereby providing high-quality feature input for subsequent feature enhancement and task execution modules.

[0212] S31. First, preprocess the multi-sensor signal. Use a bandpass filter to remove high-frequency noise and low-frequency drift to ensure the purity of the signal. The filtered signal is normalized to scale the signal amplitude to a uniform range to improve the stability of subsequent feature extraction and model training. Then, extract the maximum, minimum, and variance time domain features; perform Fourier transform to convert the time domain signal into a frequency domain signal to obtain a spectrum diagram; calculate the spectrum energy and analyze the energy distribution of the signal in different frequency bands; extract frequency domain features such as center frequency and spectrum entropy.

[0213] S32, such as Figure 3 As shown in Figure 1, the preprocessed signal is input into a one-dimensional residual network to capture and process the complex relationship between multi-source signals. The residual network introduces skip connections to make the training of deep networks more stable and can effectively integrate and extract the correlation features of multi-source signals.

[0214] S33. Construction of one-dimensional residual network: By stacking multiple residual blocks, each residual block contains two convolutional layers and a shortcut connection. The first convolutional layer extracts the local features of the input signal through a one-dimensional convolution operation, and uses batch normalization and ReLU activation functions for nonlinear transformation. The second convolutional layer further extracts the deep features of the signal, also using batch normalization and ReLU activation functions. The shortcut connection adds the input signal directly to the output of the second convolutional layer to form a residual output. By stacking multiple residual blocks, a deep one-dimensional residual network is constructed. Each residual block can further extract more complex signal features based on the previous residual block.

[0215] S34. Use the global average pooling layer to perform pooling on the output of the residual block. The function of the global average pooling layer is to average the features of all time steps to generate a fixed-size feature vector, reducing the dimension while retaining important information.

[0216] S4: Build a feature enhancement module and introduce a multi-head attention mechanism to enhance the model’s feature extraction capability.

[0217] S41. Based on the preprocessed signal, the signal is first standardized and normalized to ensure that the dimensions of the signal are on the same scale, which helps the model to better process the input data. The standardized and normalized signal is input into the multi-head attention mechanism.

[0218] S42. In the multi-head attention mechanism, the input signal is passed through multiple self-attention layers. Each self-attention layer generates different feature representations that capture the relevance of the signal in different contexts.

[0219] S43: For the output of each self-attention layer, a linear transformation is performed to adjust the dimension of the feature. Then, the outputs of all self-attention layers are concatenated to generate a comprehensive feature vector.

[0220] S44, the concatenated feature vector is passed through a fully connected layer to further process and transform the comprehensive features. The output of this fully connected layer is the enhanced feature representation, which can better reflect the complex relationship and important features of the signal.

[0221] S45, after the fully connected layer processing, the enhanced feature representation is added to the original input signal through the residual connection. Then the layer is normalized, which helps to stabilize and accelerate the network training process while preventing the gradient from disappearing or exploding.

[0222] S46. Finally, the enhanced feature representation is used as the input of the model for the subsequent task execution module. The feature enhancement module can capture the correlation of signals in different contexts by introducing a multi-head attention mechanism, improve the feature extraction capability and prediction accuracy of the model, and thus improve the accuracy and reliability of tool wear prediction.

[0223] S5: Build a task execution module and use a model that combines a one-dimensional residual network and a multi-head attention mechanism to perform tool wear prediction tasks.

[0224] S51. First, after the signal goes through preprocessing, data processing, complex relationship processing and feature enhancement modules, high-quality feature representations are generated. These feature representations are used as inputs to the task execution module.

[0225] S52, input the features into the one-dimensional residual network part. The one-dimensional residual network is formed by stacking multiple residual blocks in depth, forming a deep one-dimensional residual network to enhance the model's modeling ability for complex signals. Each residual block contains two or more one-dimensional convolutional layers, and each convolutional layer is connected to a ReLU activation function and batch normalization. By introducing jump connections, the effective transmission of information in the deep network is ensured, the gradient vanishing problem is reduced, and the stability and effect of model training are improved.

[0226] S53. Input the output features of the one-dimensional residual network into the multi-head attention mechanism. The multi-head attention mechanism improves the model's attention to key features by assigning weights to different time steps. The specific steps are: first, perform a linear transformation to generate query, key, and value matrices; then calculate the attention score between the query and the key through the dot product, and scale and normalize the attention score; finally, use the normalized attention weights to weighted sum the values ​​to obtain the output of each attention head.

[0227] S54, concatenate the features output by the multi-head attention mechanism, perform residual connection and layer normalization to form the final feature representation. This step ensures the comprehensive expression of features at different levels and improves the prediction ability of the model.

[0228] S55, input the final feature representation to the fully connected layer to predict the tool wear. The fully connected layer maps the feature vector to the prediction result and outputs the predicted value of the wear through the activation function.

[0229] S56. Define the loss function of the model and use the mean square error as the loss function to measure the difference between the predicted value and the actual wear amount. Use the Adam optimizer to minimize the loss function, calculate the gradient of the loss relative to the model parameters through the back propagation algorithm, and update the model parameters. Iterate the model training until the loss function converges or reaches the predetermined training rounds. During the training process, monitor the performance indicators of the model to ensure that it performs well on the training set and the validation set.

[0230] The present invention provides a tool wear prediction model based on a one-dimensional residual network and a multi-sensor signal (multi-source signal), which integrates multi-dimensional signals such as cutting force, vibration, acoustic emission and temperature field. Through feature dimensionality reduction, multimodal input is converted into two-dimensional data, thereby accelerating data processing speed and providing efficient input for offline training and online prediction. The model accurately removes complex redundant information in multi-dimensional signals and extracts deeper dependency patterns by combining a residual network with a multi-head attention mechanism. In order to solve the common problems of small samples and sample imbalance in offline training, feature enhancement technology is introduced, and optimization is performed using an automatic encoding adversarial generative network. This model covers the entire process from signal acquisition, data processing to wear prediction, significantly improving the ability to capture complex relationships between multi-sensor signals, enhancing the accuracy and reliability of wear prediction, and has broad engineering application prospects.

[0231] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Any technician familiar with the art can make many possible changes and modifications to the technical solution of the present invention by using the above disclosed methods and technical contents without departing from the scope of the technical solution of the present invention, or modify it into an equivalent embodiment of equivalent changes. Therefore, all equivalent changes made according to the shape, structure and principle of the present invention without departing from the content of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A tool wear prediction method based on a multi-source signal of a one-dimensional residual network, characterized in that: include: Step 1: Build a signal acquisition module to convert the collected multi-sensor signals into a spectrum diagram and perform wavelet transform; Step 2: Build a data processing module to perform balancing operations on the data; Step 3: Construct a complex relationship processing module to build a one-dimensional residual network structure to process the complex relationship between multi-source signals; Step 4: Build a feature enhancement module and introduce a multi-head attention mechanism to enhance the model’s feature extraction capability; Step 5: Build a task execution module and use a model that combines a one-dimensional residual network and a multi-head attention mechanism to perform tool wear prediction tasks.

2. The multi-source signal tool wear prediction method based on one-dimensional residual network as claimed in claim 1 is characterized in that: The construction process of the signal acquisition module includes: Step 11: Collect signals through multiple sensors, filter the audio information, and then perform fast Fourier transform to obtain a spectrum diagram; Step 12: Use the db4 wavelet function to perform wavelet transform on the spectrum graph, so that the signal can be decomposed and analyzed at different scales and a wavelet scale graph can be obtained; Among them, wavelet transform includes: Step 121: performing signal decomposition, firstly performing multi-scale decomposition on the signal to be processed, the signal is decomposed into components at multiple different scales, wherein the low-frequency component reflects the overall trend of the signal, and the high-frequency part reflects the details and mutations of the signal; Step 122: using wavelet filters, one is a low-pass filter and the other is a high-pass filter, the two sets of filters are determined by the coefficients of the db4 wavelet function, the low-pass filter is used to extract the low-frequency components of the signal, and the high-pass filter is used to extract the high-frequency components of the signal; Step 123: Convolve the signal with low-pass and high-pass filters to obtain low-frequency and high-frequency coefficients; Step 124: downsampling the convolution result to reduce the frequency so that the scale of the signal is reduced by half; Step 125: Repeat steps 121-125 to decompose the signal into components of lower scales until the desired number of decomposition levels is reached.

3. The tool wear prediction method based on multi-source signals of one-dimensional residual network as claimed in claim 1 is characterized in that: The construction process of the data processing module includes: Step 21: Data preparation, ensuring that the multi-sensor signal dataset contains various characteristics and distribution of the original data; Step 22: Design an autoencoder and construct the encoder and decoder parts of the autoencoder. The encoder maps the original data to a low-dimensional representation, and the decoder attempts to reconstruct the original data from the low-dimensional representation. Step 23: Construct a generator and a discriminator to adapt to the characteristics of sensor signal data. The generator is responsible for generating sample data from random noise, and the discriminator distinguishes between generated false samples and real samples. Step 24: Define the loss function, which consists of the reconstruction loss of the autoencoder and the adversarial loss of the generative adversarial network. The reconstruction loss is used to measure the accuracy of data reconstruction, and the adversarial loss is used to encourage the generator to generate real data samples. Step 25: Network training, optimize the model by alternately training the autoencoder and the generative adversarial network, first use the reconstruction loss of the autoencoder to train the autoencoder, and then use the adversarial loss to train the generator and discriminator; Step 26: Evaluate the data samples generated by the generator to check whether they conform to the distribution characteristics of the original data, and then optimize and adjust the model based on the evaluation results, including adjusting the network structure, loss function and training parameters to improve the quality fidelity of the generated results.

4. The multi-source signal tool wear prediction method based on one-dimensional residual network as claimed in claim 3 is characterized in that: The construction process of the autoencoder includes: Step 221: Define the encoder architecture, select a multi-layer perceptron MLP structure, use ReLU corrected linear unit as the activation function, and in the forward propagation process, the input signal data passes through each hidden layer of the encoder and is gradually converted into a lower dimensional representation. This process is expressed by the formula: h=ReLU(W1x+b1) z=W2h+b2 Where x is the input signal data, h is the hidden layer representation, z is the encoded representation, W1 and W2 are the weight matrices of the hidden layer and the encoding layer, and b1 and b2 are the corresponding bias vectors; Step 222: Define the decoder framework, the structure of the decoder, and the activation function selection. In the forward propagation process of the decoder, the decoder receives the representation generated by the encoder as input and decodes it into the original input data. The forward propagation process of the decoder is expressed by the formula: h′=ReLU(W′1x+b′1) Where h′ is the hidden layer representation of the decoder, is the output of the decoder, representing the reconstructed input data; Step 223: Define the loss function and optimizer. The goal of the autoencoder is to minimize the reconstruction error between the input data and the decoder output data. The mean square error (MSE) is used as the loss function. The MSE is defined by the following formula: Where N is the number of samples, x i is the original input data, is the output data of the decoder; Step 224: Back propagation and parameter update, calculating the gradient of the loss with respect to the parameters by back propagation, and updating the parameters using a gradient descent optimization algorithm; The construction process of the generator and discriminator includes: Step 231: GAN consists of two parts: a generator and a discriminator. The generator receives a random noise vector z and outputs a generated data sample G(z; θ g ): Where G is the generator function with parameter θ g , used to map random noise z to generate data samples The discriminator receives real data samples x or generated data samples Through the parameter θ disc Output the probability D(x;θ) that the sample is real data disc )and Where D is the function of the discriminator, and the parameter is θ disc ; Step 232: The loss functions of the discriminator and the generator are defined. The goal of the discriminator is to maximize the log-likelihood estimate so that the probability of the real sample is close to 1 and the probability of the generated sample is close to 0. The goal of the generator is to minimize the log-likelihood estimate so that the probability of the generated sample is close to 1 and maximize the probability that the generated sample is considered to be a real sample by the discriminator.

5. The multi-source signal tool wear prediction method based on one-dimensional residual network as claimed in claim 3 is characterized in that: The construction process of network training and defining loss function includes: Step 241: Randomly initialize the parameters θ of the encoder and decoder e ,θ d , the parameters θ of the generator and discriminator g ,θ disc ; Step 242: Forward propagation and reconstruction loss calculation of the encoder. The encoder maps the input data x to the latent space representation z: z=E(x;θ e ) The decoder reconstructs the latent space representation z as input data Calculate the reconstruction loss L recon : Step 243: GAN forward propagation and adversarial loss calculation, calculate the discriminator loss L D : Calculate the generator loss L G : Step 244: Back propagation and parameter update, update the encoder and decoder parameters to minimize the reconstruction loss: Update the discriminator parameters to minimize the discriminator loss: Update the generator parameters to minimize the generator loss: Step 245: Iterate training, repeat steps 241-245 until the model converges, each iteration includes an update of the encoder and decoder, and an update of the generator and discriminator.

6. The multi-source signal tool wear prediction method based on one-dimensional residual network as claimed in claim 1 is characterized in that: The construction process of the complex relationship processing module includes: Step 31: Collect lathe signal data from multiple sensors and use filters to extract and classify signal features; Step 32: converting the processed time domain signal into a frequency domain signal through Fourier transform, and further extracting multi-scale features through wavelet transform; Step 33: Construct a one-dimensional residual network to capture and process the complex relationship between multi-source signals. The residual network stabilizes the training of the deep network by introducing jump connections, and integrates and extracts the correlation features of multi-source signals. In step 31, the process of using a filter to extract and classify signal features includes: Step 311: preprocessing the signals, using a low-pass filter to remove high-frequency noise, and then normalizing the signals to control their amplitudes within a standard range; Step 312: extracting time domain features from the signal, and reflecting the statistical properties of the signal through these features, the time domain features include: maximum value, minimum value, peak-to-peak value, average value, variance, standard deviation, mean square value, kurtosis, skewness, shape factor, peak factor, pulse factor, margin factor, and clearance factor; In step 32, the process of converting the time domain signal into the frequency domain signal by Fourier transform includes: Step 321: convert the time series signal into the frequency domain through Fourier transform and extract the frequency domain features; Step 322: extracting time-frequency domain features, combining the time domain and the frequency domain, decomposing the signal into different frequency band energies through wavelet transform, and extracting local features; Step 323: Combining all extracted features into a feature vector; In step 33, the process of constructing the one-dimensional residual network includes: Step 331: input data preparation, using the pre-processed multi-sensor signal data as the input of the network; Step 332: construct an initial convolution layer to perform initial convolution on the input signal to extract low-level features. The convolution operation captures local timing patterns and frequency features. Step 333: Construction of residual blocks, each residual block includes two convolutional layers and a shortcut connection, allowing the signal to be directly transmitted through the skip connection; Step 334: global average pooling layer, reducing feature dimensions and aggregating information through the global average pooling layer; Step 335: Construct a fully connected layer and an output layer, input the result of global average pooling into the fully connected layer, and obtain the final output.

7. The multi-source signal tool wear prediction method based on one-dimensional residual network as claimed in claim 1 is characterized in that: The construction process of the feature enhancement module includes: Step 41: preprocessing the signal, standardizing and normalizing the signals collected by multiple sensors to ensure that all dimensions of the signal are on the same scale; Step 42: Linear transformation, input the preprocessed signal into the linear layer to generate Query, Key and Value matrices; Step 43: Calculate the attention score. Calculate the attention score between the query and the key by dot product. Scale and softmax normalize the attention score to ensure the stability and interpretability of the score. Use the normalized attention weight to weight the sum of the values ​​to get the output of each attention head. Step 44: Concatenate the outputs of all attention heads and perform linear transformation to generate the final output feature representation. Then, perform residual connection on the output and input features of the multi-head attention mechanism and perform layer normalization.

8. The multi-source signal tool wear prediction method based on one-dimensional residual network as claimed in claim 1 is characterized in that: The construction process of the task execution module includes: Step 51: Combine the one-dimensional residual network and the multi-head attention mechanism; Step 52: Define the loss function of the model, use an optimization algorithm to minimize the loss function, and then iteratively train the model until convergence; In step 51, the process of combining the one-dimensional residual network and the multi-head attention mechanism includes: Step 511: The one-dimensional residual network part is formed by stacking multiple residual blocks in depth to form a deep one-dimensional residual network to enhance the model's modeling ability for complex signals. The residual block is the basic structure of the network. Each residual block contains two or more one-dimensional convolutional layers. Each convolutional layer is followed by a ReLU activation function and batch normalization, and a jump connection is introduced between the input and output. Step 512: Input the output features of the one-dimensional residual network into the multi-head attention mechanism to enhance the feature representation through the attention mechanism.

9. The multi-source signal tool wear prediction method based on one-dimensional residual network as claimed in claim 8, characterized in that: In step 52, it includes: Step 521: define a loss function, and use the mean square error as the loss function to measure the difference between the model prediction value and the actual value; Step 522: Optimize the algorithm design, use the Adam optimizer, and adaptively adjust the learning rate of each parameter, which is suitable for processing sparse gradients and high-dimensional data.

10. A tool wear prediction system based on a multi-source signal of a one-dimensional residual network, used to implement a tool wear prediction method based on a multi-source signal of a one-dimensional residual network as claimed in any one of claims 1 to 9, characterized in that: include: A signal acquisition module, used to convert the collected multi-sensor signals into a spectrum diagram and perform wavelet transformation; A data processing module, used for performing a balancing operation on the data; Complex relationship processing module, used to construct a one-dimensional residual network structure to process complex relationships between multi-source signals; Feature enhancement module, which is used to introduce a multi-head attention mechanism to enhance the model's feature extraction capability; The task execution module uses a model that combines a one-dimensional residual network and a multi-head attention mechanism to perform tool wear prediction tasks.

Images