Track disease detection method based on LSTM-BiGRU network

Abstract

The invention provides a track disease detection method based on an LSTM-BiGRU network, and the method comprises the steps: obtaining a multi-mode vibration signal of a track, and carrying out the electric signal conversion of the multi-mode vibration signal; tearing and separating the electric signal; separating single and double components from the modulated sample signal, performing feature extraction on the single and double components by using a long and short term memory network, and performing weighting processing on a feature signal; performing information entropy calculation on the weighted feature signals according to an improved ECER entropy change extremum feature recognition method, and calculating a feature weight of each weighted feature signal according to the information entropy; and fusing the weighted feature signals according to the feature weights, inputting the fused features into a BiGRU network for fault detection, and outputting a track disease detection result according to a final representation vector. According to the embodiment of the invention, efficient and accurate track disease diagnosis can be realized.

Description

Technical Field

[0001] The present invention relates to the field of track detection technology, and in particular to a track defect detection method based on an LSTM-BiGRU network. Background Art

[0002] With the rapid expansion of high-speed railway networks, track defect monitoring has become a key link in ensuring the safety of train operations.

[0003] However, existing methods for detecting track defects face many challenges. Traditional manual inspections and periodic maintenance are not only time-consuming and labor-intensive, but also difficult to maintain the consistency of detection conditions and the accuracy of detection results in actual environments. In addition, these methods are highly dependent on the experience and technical level of operators, resulting in a high subjective error rate. In recent years, although some intelligent monitoring systems based on sensors and preliminary data processing technologies have been applied to track defect detection, they still have significant technical defects. For example, early automatic detection systems usually rely on fixed thresholds to judge the state of the track. This method cannot adapt to changes in different environmental conditions and is prone to false alarms or missed alarms. In addition, these systems have limited capabilities in extracting track defect features, especially in complex and changing environments, making it difficult to achieve efficient and accurate defect identification. Summary of the invention

[0004] The purpose of the present invention is to provide a track defect detection method based on LSTM-BiGRU network, aiming to solve the problem that traditional technologies are prone to false positives or false negatives when performing track defect detection, making it difficult to achieve efficient and accurate defect identification.

[0005] In a first aspect, the present invention provides a track defect detection method based on an LSTM-BiGRU network, the method comprising: Acquiring a multimodal vibration signal of the track, and converting the multimodal vibration signal into an electrical signal to obtain an electrical signal; Performing tearing and separation on the electrical signal to combine the electrical signal with the angular frequency and modulation phase of the radio frequency sampling clock to obtain a modulated sample signal; Separating single and double components from the modulated sample signal, and applying a long short-term memory network to extract features of the single and double components respectively, to obtain feature signals at each moment corresponding to the single and double components respectively, and performing weighted processing on the feature signals to obtain weighted feature signals; The information entropy of the weighted feature signal is calculated according to the improved ECER entropy change extreme value feature recognition method, and the feature weight of each weighted feature signal is calculated according to the information entropy; The weighted feature signals are fused according to the feature weights to obtain fused features, and the fused features are input into the BiGRU network for fault detection to obtain a final characterization vector, and the track disease detection result is output according to the final characterization vector.

[0006] In a second aspect, the present invention provides a track defect detection system based on an LSTM-BiGRU network, the system comprising: A signal conversion module, used for acquiring a multi-modal vibration signal of the track, and converting the multi-modal vibration signal into an electrical signal to obtain an electrical signal; A signal modulation module, used for tearing and separating the electrical signal, so as to combine the electrical signal with the angular frequency and modulation phase of the radio frequency sampling clock to obtain a modulated sample signal; A signal separation module, used to separate the single and double components from the modulated sample signal, and use a long short-term memory network to extract features of the single and double components respectively, to obtain feature signals at each moment corresponding to the single and double components respectively, and to perform weighted processing on the feature signals to obtain weighted feature signals; An information entropy acquisition module is used to calculate the information entropy of the weighted feature signal according to the improved ECER entropy change extreme value feature recognition method, and calculate the feature weight of each weighted feature signal according to the information entropy; The feature fusion module is used to fuse the weighted feature signals according to the feature weights to obtain fused features, and input the fused features into the BiGRU network for fault detection to obtain a final characterization vector, and output the track disease detection result according to the final characterization vector.

[0007] In a third aspect, the present invention provides a storage medium, which stores one or more programs, which, when executed by a processor, implement the above-mentioned track defect detection method based on the LSTM-BiGRU network.

[0008] In a fourth aspect, the present invention provides an electronic device, the electronic device comprising a memory and a processor, wherein: The memory is used to store computer programs; When the processor is used to execute the computer program stored in the memory, the above-mentioned track defect detection method based on the LSTM-BiGRU network is implemented.

[0009] Compared with the prior art, the embodiments of the present invention have the following advantages: 1. According to the above-mentioned rail disease detection method based on LSTM-BiGRU network, accurate monitoring of rail diseases is achieved by adopting LSTM-BiGRU network fused with attention mechanism. This method removes noise and highlights key fault features through refined preprocessing technology. The membership function is introduced to weight the extracted features, and the weight of important features is enhanced. In this way, the LSTM network is used to extract deep features of the signal and realize two-way information mining, thereby capturing the long-term dependency of rail diseases. In addition, the ECER entropy change extreme value feature recognition method is introduced to deal with the uncertainty of features, and the feature weights are further optimized through the attention mechanism to ensure that the model can focus on key information. Finally, fault classification is performed through the fully connected layer and the Softmax layer, achieving efficient and accurate fault diagnosis. This method is suitable for complex and changing environments, effectively overcomes the shortcomings of early systems in adapting to environmental changes and feature extraction, and significantly improves the accuracy and robustness of disease identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of a track defect detection method based on an LSTM-BiGRU network proposed in one embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a track defect detection system based on an LSTM-BiGRU network proposed in one embodiment of the present invention.

[0011] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein should be understood by people with general skills in the field to which the present invention belongs. "Including" and similar words used in this article mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0013] like Figure 1 As shown, an embodiment of the present invention proposes a track defect detection method based on an LSTM-BiGRU network, the method comprising steps S101 to S105, wherein: Step S101: acquiring a multimodal vibration signal of a track, and converting the multimodal vibration signal into an electrical signal to obtain an electrical signal; It should be noted that the present embodiment uses an IEPE vibration sensor for signal acquisition, covering multi-modal vibration signals in different frequency ranges and multiple modes. The multi-modal vibration signals may be acceleration, velocity, displacement and other signals.

[0014] In addition, in some embodiments, in order to facilitate the transmission of the collected signals to the computer for subsequent processing and analysis, a new super-dimensional multivariate signal electrical domain conversion technology is used, and the electrical signal conversion is performed specifically according to the following formula: ; in, is the converted electrical signal, is the frequency response function, is the gain coefficient, is the sensitivity coefficient of the signal acquisition sensor, and are the amplitudes of the i-th and j-th multimodal vibration signals, respectively, and are the angular frequencies of the i-th and j-th multimodal vibration signals, respectively, and are the jth and ith phase angles respectively, is the number of modes, i.e. the intrinsic characteristics of vibration, such as vertical vibration, lateral vibration and other different modes, is the number of types of multimodal vibration signals, that is, the external types of vibration signals, such as vibration caused by external factors such as train running, track defects and environmental factors (such as wind or earthquake), For the The gain coefficient of the multi-modal vibration signal, is the sensitivity coefficient of the i-th multimodal vibration signal, is the wave function of the i-th multimodal vibration signal, is the attenuation coefficient of the i-th multimodal vibration signal, t is the current time, is the integration variable.

[0015] Step S102: tearing and separating the electrical signal to combine the electrical signal with the angular frequency and modulation phase of the radio frequency sampling clock to obtain a modulated sample signal; It should be pointed out that in this step, a refined preprocessing method based on single and double tearing separation is used for the input electrical signal, which can remove noise and interference and highlight the fault characteristics.

[0016] In some embodiments, tearing separation is first performed according to the following formula: ; In order to analyze the characteristics of the signal and obtain the spectrum information of the modulated signal, the modulated sample signal is expanded. The formula is as follows: ; in, is the RF sampling clock angular frequency, is the modulation phase, is the modulated sample signal, is an imaginary unit, is the harmonic order, .

[0017] Step S103: separating the single and double components from the modulated sample signal, and applying the long short-term memory network to extract features of the single and double components respectively, to obtain feature signals at each moment corresponding to the single and double components respectively, and performing weighted processing on the feature signals to obtain weighted feature signals; It should be pointed out that in order to further highlight the fault characteristics, a digital filter is used to analyze the single harmonic and double harmonic components in the spectrum, and the modulated sample signal is filtered to separate the modulated sample signal into single and double components.

[0018] In some embodiments, a single component is separated specifically according to the following formula: ; The two components are separated according to the following formula: ; In addition, the long short-term memory network (LSTM network) is applied to extract features of the modulated single and double component signals respectively, capture long-term dependencies and weight the features through membership functions, amplify the weights of important features, and propose the ECER entropy change extreme feature recognition method to deal with feature uncertainty. The feature fusion method is used to integrate the processed single and double component features to form a comprehensive feature signal, which provides a basis for model diagnosis.

[0019] In some embodiments, the single component and the dual component are each input as independent data samples into the LSTM network, and each input is an input matrix of 1×1024 dimensions. Then the input data is sequentially fed into the LSTM layer according to the time step number of 32 and the dimension of each time step of 1×32. There are 3 LSTM layers in the model, and each LSTM layer has 128 hidden nodes. Through these three LSTM layers, the deep features of the multimodal vibration signal of the track can be deeply mined.

[0020] Specifically, feature extraction is performed according to the following formula: ; in, is the input signal of the LSTM network, including single component and double component. is the characteristic signal output by the LSTM network at the t-1th moment, , , , is the weight matrix for each gate, , , , is the bias matrix for each gate, , , is the output value of each gate at the tth moment of the i-th layer network, is the new information candidate value, is the memory unit at the tth moment, is the memory unit at the t-1th moment, For the The characteristic signal output by the LSTM network at the tth moment of the layer network, , Both are activation functions.

[0021] Step S104: performing information entropy calculation on the weighted feature signal according to the improved ECER entropy change extreme value feature recognition method, and calculating the feature weight of each weighted feature signal according to the information entropy; In this step, the feature signal extracted by the LSTM network may contain redundant features or noise features. The dynamic Gaussian membership function is introduced to dynamically calculate the membership value of the feature by adjusting the number of Gaussian kernels, contribution, scaling factor and other parameters. The features are weighted using the membership value to amplify the weights of important features. Important features are highlighted, the impact of irrelevant features is reduced, and the accuracy of fault diagnosis is improved.

[0022] In some embodiments, the membership value of each characteristic signal is calculated according to: ; in, is the membership value of each characteristic signal, m is the total number of dynamic Gaussian kernels, For the The contribution of the Gaussian components, For the The scaling factor of the Gaussian kernel is For the The mean of the Gaussian kernels, For the The standard deviation of the Gaussian kernel, is the dynamic adjustment coefficient, For the The weight parameters of the Gaussian components, For the The bias term of the Gaussian components adjusts the activation threshold.

[0023] Then the weighted feature signal Its membership value Multiply them together to get the weighted feature signal.

[0024] In addition, due to the uncertainty of the weighted feature signal, an improved ECER entropy change extreme feature recognition method is proposed. This recognition method identifies the extreme points of entropy change by calculating the information entropy change of the weighted feature, and further adjusts the weights or selects features for the feature signals corresponding to these extreme points. Information entropy is used to quantify the uncertainty of the feature set (the fault information contained has different richness) to evaluate the contribution of the feature to fault diagnosis. By calculating the information entropy of the weighted feature signal, the change trend of the entire feature set can be analyzed, thereby determining the extreme points of entropy change that can reflect the significant change in the uncertainty of the feature. Usually, these extreme points of entropy change correspond to feature signals with higher information value, that is, these features contribute more to the classification or recognition task in the feature set.

[0025] In some embodiments, first, all weighted characteristic signals obtained from a single component are aggregated into a first characteristic signal set, and all weighted characteristic signals obtained from a dual component are aggregated into a second characteristic signal set; Then the information entropy is calculated according to the following formula: ; in, is the information entropy of the i-th feature signal in the first feature signal set or the second feature signal set, is the i-th characteristic signal in the first characteristic signal set or the second characteristic signal set, is the probability of the i-th characteristic signal appearing at the j-th moment, is the smoothing factor, h is the total number of characteristic signal moments; By calculating the information entropy of the weighted feature signal, we can analyze the changing trend of the entire feature set, thereby determining the extreme points of entropy change that can reflect significant changes in the uncertainty of the feature. Usually, these extreme points of entropy change correspond to feature signals with higher information value, that is, these features contribute more to the classification or recognition task in the feature set.

[0026] Based on the information entropy and entropy change extreme points of the feature signals, the richness of the fault information contained in each feature signal can be evaluated. In view of the differences in the information content of the feature signals, it is necessary to adjust the weights of these feature signals to optimize the quality of the feature set.

[0027] The feature weight is calculated according to the following formula: ; in, is the feature weight of the i-th weighted feature signal, is the information entropy of the i-th weighted feature signal, is the extreme value of entropy change, is the adjustment factor, is the adjustment coefficient for the interaction between characteristic signals, is the relationship adjustment coefficient between characteristic signals, is the total number of characteristic signals in the first characteristic signal set or the second characteristic signal set, is the covariance between the ith feature signal and the jth feature signal, where the two feature signals belong to the same feature signal set, is the correlation between the i-th feature signal and the j-th feature signal, where the two feature signals belong to the same feature signal set, is the jth feature signal in the first feature signal set or the second feature signal set.

[0028] Step S105: The weighted feature signals are fused according to the feature weights to obtain fused features, and the fused features are input into the BiGRU network for fault detection to obtain a final characterization vector, and the track disease detection result is output according to the final characterization vector.

[0029] In this step, more weights are preferentially assigned to features with higher information value while reducing the weights of features with less information, thereby eliminating redundant features and highlighting important features while reducing the interference of noise features.

[0030] In some embodiments, first, all first target feature weights greater than a first preset threshold are screened out from all feature weights corresponding to the first feature signal set; and all second target feature weights greater than the first preset threshold are screened out from all feature weights corresponding to the second feature signal set; Then, the corresponding characteristic signal is adjusted according to the target characteristic weight: ; ; in, is the adjusted feature signal set corresponding to a single component, is the adjusted characteristic signal set corresponding to the dual component, , , The first, second, and The first target feature weight, , , The first, second, and The feature signal corresponding to the first target feature weight, For the The target feature weights, , , The first, second, and The second target feature weight, , , The first, second, and The feature signal corresponding to the second target feature weight, For the The target feature weights, , are the total number of the first target feature weights and the second target weights respectively.

[0031] In addition, after the weights of the weighted feature signals are adjusted, the single-component and dual-component signals obtained at this time may contain complementary information. In order to make full use of this information, they need to be fused into a comprehensive feature representation to provide a basis for the next step of model diagnosis. Specifically, the adjusted feature signal set is fused according to the following formula: ; in, To fusion features, is the global weight parameter, is the correlation coefficient between the kth adjusted feature signal and the jth adjusted feature signal, is the sum of the first target feature weight and the second target weight, is a regularization term, It is a global weight parameter used to balance the contribution of single-component features and dual-component features in the final fusion features. It can be obtained by sampling and updating the posterior distribution using a conventional Bayesian optimization algorithm (such as Gaussian process) to gradually approach the optimal value. It can also be obtained by randomly selecting several values ​​to train the model and verifying which parameter is optimal using a conventional random search method.

[0032] Then, the BiGRU network is used to conduct deeper information mining on the fused features from both positive and negative directions, and the attention mechanism is introduced to strengthen the model's focus on key information, ignore some information irrelevant to the results, and reduce the risk of information loss to improve accuracy.

[0033] The fused features are input into the BiGRU network. The Bidirectional Gated Recurrent Unit Network (BiGRU Network) is a combination of two GRU models, one forward and one reverse. Since the input is the entire time series, when mining information, the features at a certain moment rely on both previous information and future information at that moment as a basis.

[0034] In some embodiments, the calculation formula of the BiGRU network is as follows: ; in, and are the forward and backward hidden states, respectively. , , are weight matrices for linear transformations of input gates, reset gates, and candidate hidden states, is the long sequence after processing, , , are normalized weight matrices, is a candidate hidden state, , are the outputs of the update gate and the reset gate, respectively. To update the part that the gate did not pass, is the hidden state at the t-1th moment, is the hidden state at the t+1th moment, is the input signal at the tth moment, that is , is the output signal at the tth moment, and the arrow is the propagation direction.

[0035] Since not all feature vectors output by the BiGRU network play a decisive role in track disease diagnosis, different BiGRU hidden states are normalized and assigned different weights to achieve the effect of focusing on important information. The expression is as follows: ; ; in, and are bias and weight respectively, The tth weight assigned to the attention mechanism, , is the hidden state at the tth moment, T is the total time step, S is the final representation vector, is the function used to process the output signal of the BiGRU network.

[0036] In addition, in the output of the BiGRU network, since the features at different time steps may have different contributions to the final disease diagnosis, an attention mechanism is introduced to strengthen the model's focus on key information, ignore some information that is irrelevant to the results, and reduce the risk of information loss to improve accuracy.

[0037] Finally, the final representation vector is sent to the fault classification layer for integration. The classification layer contains a fully connected Softmax Each neuron in the fully connected layer is connected to the neurons in the upper layer, and the extracted features are mapped. The Softmax layer performs fault classification and compares the results with the fault values ​​set in advance. The ones with a larger proportion can be identified as faults, thus completing the fault diagnosis. Specifically, the final representation vector is input into the fully connected layer and Layer, get the probability of each disease category: select the maximum disease probability from the probabilities of all disease categories, and determine whether the maximum disease probability is greater than the second preset threshold; if the maximum disease probability is greater than the second preset threshold, it is determined that the track has a disease. Exemplarily, the track has diseases such as: long-term train operation causes track surface wear; environmental factors (such as humidity, salt, etc.) cause track material corrosion; cracks due to fatigue, impact or material defects, which may lead to fractures in severe cases.

[0038] In summary, according to the above-mentioned rail disease detection method based on LSTM-BiGRU network, accurate monitoring of rail diseases is achieved by adopting LSTM-BiGRU network fused with attention mechanism. This method removes noise and highlights key fault features through refined preprocessing technology. The membership function is introduced to weight the extracted features, and the weight of important features is enhanced. In this way, the LSTM network is used to extract deep features of the signal and realize bidirectional information mining, thereby capturing the long-term dependency of rail diseases. In addition, the ECER entropy change extreme value feature recognition method is introduced to deal with the uncertainty of features, and the feature weights are further optimized through the attention mechanism to ensure that the model can focus on key information. Finally, fault classification is performed through the fully connected layer and the Softmax layer, achieving efficient and accurate fault diagnosis. This method is suitable for complex and changing environments, effectively overcomes the shortcomings of early systems in adapting to environmental changes and feature extraction, and significantly improves the accuracy and robustness of disease identification.

[0039] like Figure 2 As shown, an embodiment of the present invention further proposes a track defect detection system based on an LSTM-BiGRU network, the system comprising: The signal conversion module 10 is used to obtain a multi-modal vibration signal of the track and convert the multi-modal vibration signal into an electrical signal to obtain an electrical signal; The signal modulation module 20 is used to tear and separate the electrical signal to combine the electrical signal with the angular frequency and modulation phase of the radio frequency sampling clock to obtain a modulated sample signal; A signal separation module 30 is used to separate the single and double components from the modulated sample signal, and use a long short-term memory network to extract features of the single and double components respectively to obtain feature signals at each moment corresponding to the single and double components respectively, and perform weighted processing on the feature signals to obtain weighted feature signals; An information entropy acquisition module 40 is used to calculate the information entropy of the weighted feature signal according to the improved ECER entropy change extreme value feature recognition method, and calculate the feature weight of each weighted feature signal according to the information entropy; The feature fusion module 50 is used to fuse the weighted feature signals according to the feature weights to obtain fused features, and input the fused features into the BiGRU network for fault detection to obtain a final characterization vector, and output the track disease detection result according to the final characterization vector.

[0040] On the other hand, the present invention further proposes a storage medium on which one or more programs are stored, and when the program is executed by a processor, the above-mentioned track defect detection method based on the LSTM-BiGRU network is implemented.

[0041] On the other hand, the present invention further proposes an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the above-mentioned track defect detection method based on the LSTM-BiGRU network.

[0042] Those skilled in the art will appreciate that the logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For purposes of this specification, "computer-readable medium" may be any device that can contain storage, communication, propagation or transmission of a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0043] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0044] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0045] Although the embodiments of the present invention are described in detail above, it is obvious to those skilled in the art that various modifications and variations can be made to these embodiments. However, it should be understood that such modifications and variations are within the scope and spirit of the present invention as described in the claims. Moreover, the present invention described herein may have other embodiments and may be implemented or realized in a variety of ways.

Claims

1. A track defect detection method based on LSTM-BiGRU network, characterized in that: The method comprises: Acquiring a multimodal vibration signal of the track, and converting the multimodal vibration signal into an electrical signal to obtain an electrical signal; Performing tearing and separation on the electrical signal to combine the electrical signal with the angular frequency and modulation phase of the radio frequency sampling clock to obtain a modulated sample signal; Separating single and double components from the modulated sample signal, and applying a long short-term memory network to extract features of the single and double components respectively, to obtain feature signals at each moment corresponding to the single and double components respectively, and performing weighted processing on the feature signals to obtain weighted feature signals; The information entropy of the weighted feature signal is calculated according to the improved ECER entropy change extreme value feature recognition method, and the feature weight of each weighted feature signal is calculated according to the information entropy; The weighted feature signals are fused according to the feature weights to obtain fused features, and the fused features are input into the BiGRU network for fault detection to obtain a final characterization vector, and the track disease detection result is output according to the final characterization vector.

2. The track defect detection method based on LSTM-BiGRU network according to claim 1 is characterized in that: The step of acquiring a multimodal vibration signal of the track and converting the multimodal vibration signal into an electrical signal to obtain an electrical signal comprises: The electrical signal conversion is performed according to the following formula: ； in, is the converted electrical signal, is the frequency response function, is the gain coefficient, is the sensitivity coefficient of the signal acquisition sensor, and are the amplitudes of the i-th and j-th multimodal vibration signals, respectively, and are the angular frequencies of the i-th and j-th multimodal vibration signals, respectively, and are the jth and ith phase angles respectively, is the number of modes, is the number of types of multimodal vibration signals, For the The gain coefficient of the multi-modal vibration signal, is the sensitivity coefficient of the i-th multimodal vibration signal, is the wave function of the i-th multimodal vibration signal, is the attenuation coefficient of the i-th multimodal vibration signal, t is the current time, is the integration variable.

3. The track defect detection method based on LSTM-BiGRU network according to claim 2 is characterized in that: The step of tearing and separating the electrical signal to combine the electrical signal with the angular frequency and modulation phase of the radio frequency sampling clock to obtain a modulated sample signal comprises: The tear separation is performed according to the following formula: ； Expand the modulated sample signal: ； in, is the RF sampling clock angular frequency, is the modulation phase, is the modulated sample signal, is an imaginary unit, is the harmonic order, .

4. The track defect detection method based on LSTM-BiGRU network according to claim 3 is characterized in that: The steps of separating single and double components from the modulated sample signal, and extracting features of the single and double components using a long short-term memory network to obtain feature signals at each moment corresponding to the single and double components include: The single components are separated according to the following formula: ； The two components are separated according to the following formula: ； The single component and the double component are each input into the LSTM network as independent data samples, and feature extraction is performed according to the following formula: ； in, is the input signal of the LSTM network, including single component and double component. is the characteristic signal output by the LSTM network at the t-1th moment, , , , is the weight matrix for each gate, , , , is the bias matrix for each gate, , , is the output value of each gate at the tth moment of the i-th layer network, is the new candidate value of information, is the memory unit at the tth moment, is the memory unit at the t-1th moment, For the The characteristic signal output by the LSTM network at the tth moment of the layer network, , Both are activation functions.

5. The track defect detection method based on LSTM-BiGRU network according to claim 4 is characterized in that: The step of performing weighted processing on the characteristic signal to obtain a weighted characteristic signal comprises: The dynamic Gaussian membership function is introduced to calculate the membership value of each characteristic signal: ； in, is the membership value of each characteristic signal, m is the total number of dynamic Gaussian kernels, For the The contribution of the Gaussian components, For the The scaling factor of the Gaussian kernel is For the The mean of the Gaussian kernels, For the The standard deviation of the Gaussian kernel, is the dynamic adjustment coefficient, For the The weight parameters of the Gaussian components, For the The bias term of the Gaussian components; The weighted characteristic signal Its membership value Multiply them together to get the weighted feature signal.

6. The track defect detection method based on LSTM-BiGRU network according to claim 5 is characterized in that: The step of calculating the information entropy of the weighted feature signal according to the improved ECER entropy change extreme value feature recognition method, and calculating the feature weight of each weighted feature signal according to the information entropy comprises: Aggregating all weighted characteristic signals obtained from a single component into a first characteristic signal set, and aggregating all weighted characteristic signals obtained from a dual component into a second characteristic signal set; Information entropy is calculated according to the following formula: ； in, is the information entropy of the i-th feature signal in the first feature signal set or the second feature signal set, is the i-th characteristic signal in the first characteristic signal set or the second characteristic signal set, is the probability of the i-th characteristic signal appearing at the j-th moment, is the smoothing factor, h is the total number of characteristic signal moments; The feature weight is calculated according to the following formula: ； in, is the feature weight of the i-th weighted feature signal, is the information entropy of the i-th weighted feature signal, is the extreme value of entropy change, is the adjustment factor, is the adjustment coefficient for the interaction between characteristic signals, is the relationship adjustment coefficient between characteristic signals, is the total number of characteristic signals in the first characteristic signal set or the second characteristic signal set, is the covariance between the ith feature signal and the jth feature signal, where the two feature signals belong to the same feature signal set, is the correlation between the i-th feature signal and the j-th feature signal, where the two feature signals belong to the same feature signal set, is the jth feature signal in the first feature signal set or the second feature signal set.

7. The track defect detection method based on LSTM-BiGRU network according to claim 6 is characterized in that: The step of fusing the weighted feature signals according to the feature weights to obtain fused features comprises: Filter out all first target feature weights greater than a first preset threshold from all feature weights corresponding to the first feature signal set; Filter out all second target feature weights greater than a first preset threshold from all feature weights corresponding to the second feature signal set; According to the target feature weight, the corresponding feature signal is adjusted: ； ； in, is the adjusted feature signal set corresponding to a single component, is the adjusted characteristic signal set corresponding to the dual component, , , The first, second, and The first target feature weight, , , The first, second, and The feature signal corresponding to the first target feature weight, For the The target feature weights, , , The first, second, and The second target feature weight, , , The first, second, and The feature signal corresponding to the second target feature weight, For the The target feature weights, , are the total number of the first target feature weights and the second target weights respectively; The adjusted feature signal set is fused according to the following formula: ； in, To fusion features, is the global weight parameter, is the correlation coefficient between the kth adjusted feature signal and the jth adjusted feature signal, is the sum of the first target feature weight and the second target weight, is a small regularization term.

8. The track defect detection method based on LSTM-BiGRU network according to claim 7 is characterized in that: The step of inputting the fused features into the BiGRU network for fault detection to obtain the final representation vector includes: The calculation formula of the BiGRU network is as follows: ； in, and are the forward and backward hidden states, respectively. , , are weight matrices for linear transformations of input gates, reset gates, and candidate hidden states, is the long sequence after processing, , , are normalized weight matrices, is a candidate hidden state, , are the outputs of the update gate and the reset gate, respectively. To update the part that the gate did not pass, is the hidden state at the t-1th moment, is the hidden state at the t+1th moment, is the input signal at the tth moment, that is , is the output signal at the tth moment, and the arrow is the propagation direction; Normalize the output signal of the BiGRU network and assign different weights: ； ； in, and are bias and weight respectively, The tth weight assigned to the attention mechanism, , is the hidden state at the tth moment, T is the total time step, S is the final representation vector, is the function used to process the output signal of the BiGRU network.

9. The track defect detection method based on LSTM-BiGRU network according to claim 8 is characterized in that: The step of outputting the track defect detection result according to the final characterization vector comprises: The final representation vector is input into the fully connected layer and Layer, get the probability of each disease category: Screening out a maximum disease probability from the probabilities of all disease categories, and determining whether the maximum disease probability is greater than a second preset threshold; If the maximum defect probability is greater than a second preset threshold, it is determined that there is a defect on the track.

Images