Remote monitoring method and system for hydraulic power plant equipment
By remotely monitoring the vibration and sound signals of hydropower plant equipment, time-frequency characteristic analysis and intelligent diagnosis are carried out, the problem of inefficiency of traditional monitoring methods is solved, efficient remote monitoring and intelligent fault diagnosis of hydropower plant equipment is realized, and safety and Operational benefits.
Patent Information
- Application Number
- CN202411356065.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-27
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional hydropower plant equipment monitoring methods rely on manual inspection and regular maintenance, are inefficient and cannot achieve continuous monitoring, and it is difficult to detect abnormal equipment operation in a timely manner, resulting in faults and safety accidents. The existing automated monitoring systems lack in-depth analysis and intelligent diagnosis capabilities.
Using remote monitoring method, by obtaining the operating vibration signals and sound signals of hydropower plant equipment, using remote monitoring servers to extract time-frequency features and select feature optimization, combined with attention combined feature analysis, determine whether there are abnormalities in the working state of the equipment, and realize intelligent failures. diagnosis.
It improves the monitoring efficiency and safety of hydropower plant equipment, reduces maintenance costs, and improves overall operating efficiency.
Smart Images

Figure CN119045386A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to a remote monitoring method and system for hydropower plant equipment. Background Art
[0002] As global energy demand continues to grow, hydropower, as a clean, renewable energy form, occupies an increasingly important position in the energy structure. However, the complexity of hydropower plant equipment and the diversity of operating environments make their maintenance and management face many challenges.
[0003] Traditional hydropower plant equipment monitoring methods mainly rely on manual inspections and regular maintenance. This method is not only inefficient, but also unable to continuously monitor the equipment status, making it difficult to detect potential equipment operating anomalies in a timely manner, which may lead to equipment failures and even safety accidents, seriously affecting the normal operation and economic benefits of the hydropower plant.
[0004] At present, with the development of Internet of Things technology, although some hydropower plants have introduced partial automated monitoring systems, most of them focus on simple monitoring of equipment operating parameters and lack the ability to conduct in-depth analysis and intelligent diagnosis of operating data, making it difficult to effectively detect potential problems with equipment. Summary of the invention
[0005] The present disclosure aims to solve at least one of the problems existing in the prior art and provide a method and system for remote monitoring of hydropower plant equipment.
[0006] In one aspect of the present disclosure, a method for remote monitoring of hydropower plant equipment is provided, the method comprising:
[0007] Obtain the operating vibration signal and operating sound signal of the monitored hydropower plant equipment;
[0008] Sending the operation vibration signal and the operation sound signal to a device remote monitoring server;
[0009] Utilizing the equipment remote monitoring server, respectively extracting the time-frequency characteristics of the operation vibration signal and the operation sound signal to obtain a vibration time-frequency characteristic graph and a sound time-frequency characteristic graph;
[0010] Using the device remote monitoring server, feature selection optimization is performed on the sound time-frequency feature graph and the vibration time-frequency feature graph respectively to obtain a sound time-frequency enhancement feature graph and a vibration time-frequency enhancement feature graph;
[0011] By utilizing the equipment remote monitoring server, it is determined whether there is any abnormality in the working state of the monitored hydropower plant equipment based on the attention joint features of the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map.
[0012] Optionally, respectively extracting the time-frequency features of the operation vibration signal and the operation sound signal to obtain a vibration time-frequency feature graph and a sound time-frequency feature graph includes:
[0013] Performing wavelet analysis on the operation vibration signal and the operation sound signal to obtain a vibration two-dimensional time-frequency graph and a sound two-dimensional time-frequency graph;
[0014] Inputting the vibration two-dimensional time-frequency graph into a vibration time-frequency feature extractor based on a first hole convolutional neural network model to obtain the vibration time-frequency feature graph;
[0015] The sound two-dimensional time-frequency graph is input into a sound time-frequency feature extractor based on a second hole convolutional neural network model to obtain the sound time-frequency feature graph.
[0016] Optionally, feature selection optimization is performed on the sound time-frequency feature map and the vibration time-frequency feature map respectively to obtain a sound time-frequency enhancement feature map and a vibration time-frequency enhancement feature map, including:
[0017] The sound time-frequency feature map and the vibration time-frequency feature map are respectively input into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map.
[0018] Optionally, the sound time-frequency feature map and the vibration time-frequency feature map are respectively input into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map, including:
[0019] Splitting the sound time-frequency feature map into a set of sound time-frequency local feature vectors;
[0020] Calculating the characteristic energy distribution spectrum vector of each sound time-frequency local feature vector in the set of sound time-frequency local feature vectors to obtain a set of sound time-frequency characteristic energy distribution spectrum vectors;
[0021] Calculating the position mean vector of the set of energy distribution spectrum vectors of the sound time-frequency characteristics as the center vector of the energy spectrum distribution;
[0022] Calculating the Mahalanobis distance between the energy spectrum distribution center vector and each sound time-frequency feature energy distribution spectrum vector in the set of the sound time-frequency feature energy distribution spectrum vectors as the energy distribution spectrum span factor, and obtaining a set of energy distribution spectrum span factors;
[0023] Comparing each energy distribution spectrum span factor in the set of energy distribution spectrum span factors with a preset threshold, and selecting sound time-frequency local feature vectors corresponding to energy distribution spectrum span factors less than the preset threshold to form a set of selected sound time-frequency local feature vectors;
[0024] The set of the selected sound time-frequency local feature vectors is trans-aggregated in a feature splitting manner to obtain the sound time-frequency enhanced feature map.
[0025] Optionally, calculating the characteristic energy distribution spectrum vector of each sound time-frequency local feature vector in the set of sound time-frequency local feature vectors to obtain the set of sound time-frequency characteristic energy distribution spectrum vectors includes:
[0026] Calculate the feature distribution energy synergistic representation vector between the i-th sound time-frequency local feature vector and each other sound time-frequency local feature vector in the set of the sound time-frequency local feature vectors in a manner of taking the maximum eigenvalue element by element, and obtain a set of feature distribution energy synergistic representation vectors;
[0027] The characteristic distribution energy synergy factor of each characteristic distribution energy synergy representation vector in the set of characteristic distribution energy synergy representation vectors is calculated to obtain the characteristic energy distribution spectrum vector of the i-th sound time-frequency local characteristic vector composed of multiple characteristic distribution energy synergy factors.
[0028] Optionally, calculating the characteristic distribution energy synergy factor of each characteristic distribution energy synergy representation vector in the set of characteristic distribution energy synergy representation vectors to obtain the characteristic energy distribution spectrum vector of the i-th sound time-frequency local characteristic vector composed of multiple characteristic distribution energy synergy factors includes:
[0029] The square of the difference between the maximum eigenvalue of the feature distribution energy synergistic representation vector and its eigenmean, the feature variance of the feature distribution energy synergistic representation vector, and the weighted sum of the bias terms are calculated, and then the variance of the feature distribution energy synergistic representation vector and the bias term are added and divided by the weighted sum to obtain the feature distribution energy synergy factor.
[0030] Optionally, determining whether there is an abnormality in the working state of the monitored hydropower plant equipment based on the attention joint feature of the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map includes:
[0031] Inputting the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map into an attention-driven feature joint saliency perception network to obtain a sound-vibration time-frequency joint saliency perception feature map;
[0032] The sound-vibration time-frequency joint significant perception feature map is input into a classifier-based monitoring result generation module to obtain a monitoring result, which is used to indicate whether there is an abnormality in the working state of the monitored hydropower plant equipment.
[0033] Optionally, inputting the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map into an attention-driven feature joint saliency perception network to obtain a sound-vibration time-frequency joint saliency perception feature map comprises:
[0034] Performing feature shape reshaping on the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map to obtain a sound time-frequency enhancement feature shape reshaping matrix and a vibration time-frequency enhancement feature shape reshaping matrix;
[0035] Input the sound time-frequency enhancement feature shape reshaping matrix and the vibration time-frequency enhancement feature shape reshaping matrix into a feature channel-by-channel interactive perception module to obtain a sound-vibration dependency matrix and a vibration-sound dependency matrix;
[0036] Inputting the sound-vibration dependency matrix and the vibration-sound dependency matrix into a random dropout module to obtain a pruned sound-vibration dependency matrix and a pruned vibration-sound dependency matrix;
[0037] Performing matrix multiplication of the pruned sound-vibration dependency matrix and the vibration time-frequency enhancement feature shape reshaping matrix to obtain a dependency-optimized vibration time-frequency enhancement feature matrix;
[0038] Performing matrix multiplication of the pruned vibration-sound dependency matrix and the sound time-frequency enhancement feature shape reshaping matrix to obtain a dependency-optimized sound time-frequency enhancement feature matrix;
[0039] Reshaping the dependency-optimized sound time-frequency enhancement feature matrix and the dependency-optimized vibration time-frequency enhancement feature matrix to obtain an optimized sound time-frequency enhancement feature map and an optimized vibration time-frequency enhancement feature map;
[0040] The weighted sum of the optimized sound time-frequency enhancement feature map and the optimized vibration time-frequency enhancement feature map is calculated to obtain the multi-scale joint perception significant feature map of the plant state.
[0041] Optionally, the sound time-frequency enhancement feature shape reshaping matrix and the vibration time-frequency enhancement feature shape reshaping matrix are input into a feature channel-by-channel interactive perception module to obtain a sound-vibration dependency matrix and a vibration-sound dependency matrix, including:
[0042] Calculate the transposed matrix of the sound time-frequency enhancement characteristic shape reshaping matrix multiplied by the vibration time-frequency enhancement characteristic shape reshaping matrix to obtain a sound-vibration association representation matrix;
[0043] The sound-vibration association representation matrix is divided by the scale of the vibration time-frequency enhancement feature shape reshaping matrix and then input into a softmax function to obtain the sound-vibration dependency matrix;
[0044] Calculate the transposed matrix of the sound time-frequency enhancement characteristic shape reshaping matrix and multiply it by the vibration time-frequency enhancement characteristic shape reshaping matrix to obtain a plant state vibration-sound correlation representation matrix;
[0045] The vibration-sound association representation matrix is divided by the scale of the sound time-frequency enhancement feature shape reshaping matrix and then input into the softmax function to obtain the vibration-sound dependency matrix.
[0046] Another aspect of the present disclosure provides a remote monitoring system for hydropower plant equipment, the remote monitoring system for hydropower plant equipment comprising:
[0047] A signal acquisition module, used to acquire operating vibration signals and operating sound signals of monitored hydropower plant equipment;
[0048] A signal transmission module, used for transmitting the operation vibration signal and the operation sound signal to a device remote monitoring server;
[0049] A time-frequency feature extraction module, used to use the device remote monitoring server to extract the time-frequency features of the operation vibration signal and the operation sound signal respectively, and obtain a vibration time-frequency feature graph and a sound time-frequency feature graph;
[0050] A feature selection optimization module, used to use the device remote monitoring server to perform feature selection optimization on the sound time-frequency feature graph and the vibration time-frequency feature graph respectively, to obtain a sound time-frequency enhancement feature graph and a vibration time-frequency enhancement feature graph;
[0051] The abnormality judgment module is used to use the equipment remote monitoring server to determine whether there is an abnormality in the working state of the monitored hydropower plant equipment based on the attention joint features of the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map.
[0052] Compared with the prior art, the present invention utilizes the vibration signals and sound signals during the operation of hydropower plant equipment, and combines the Internet of Things technology and deep learning technology to extract time-frequency features and perform correlation interaction analysis on the vibration signals and sound signals of the hydropower plant equipment, so as to mine potential abnormal information of the equipment operation status, thereby realizing remote monitoring of the operation status of the hydropower plant equipment and intelligent fault diagnosis, effectively improving the monitoring efficiency and safety of the hydropower plant equipment, reducing maintenance costs, and improving the overall operation efficiency of the hydropower plant. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.
[0054] Figure 1 A flow chart of a method for remote monitoring of hydropower plant equipment provided in one embodiment of the present disclosure;
[0055] Figure 2 A data flow diagram of a method for remote monitoring of hydropower plant equipment provided by another embodiment of the present disclosure;
[0056] Figure 3 A flowchart of step S5 in a method for remote monitoring of hydropower plant equipment provided by another embodiment of the present disclosure;
[0057] Figure 4 A schematic structural diagram of a remote monitoring system for hydropower plant equipment provided in another embodiment of the present disclosure. DETAILED DESCRIPTION
[0058] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the embodiments of the present disclosure will be described in detail below in conjunction with the accompanying drawings. However, it can be understood by those skilled in the art that in each embodiment of the present disclosure, many technical details are proposed in order to enable readers to better understand the present disclosure. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed for protection in the present disclosure can also be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present disclosure. The various embodiments can be combined and referenced with each other without contradiction.
[0059] Unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.
[0060] Although the present disclosure makes various references to certain modules in the system according to embodiments of the present disclosure, any number of different modules can be used and run on a user terminal and / or server. The modules are illustrative only, and different aspects of the systems and methods can use different modules.
[0061] The present disclosure uses a flow chart to illustrate the operations performed by the method according to the embodiments of the present disclosure. It should be understood that the preceding or following operations are not necessarily performed precisely in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.
[0062] The traditional monitoring method of hydropower plant equipment mainly relies on manual inspections and regular maintenance. This method is not only inefficient, but also unable to continuously monitor the status of the equipment. It is difficult to detect potential operating anomalies of the equipment in a timely manner, which may lead to equipment failures and even safety accidents, seriously affecting the normal operation and economic benefits of the hydropower plant. At present, with the development of Internet of Things technology, although some hydropower plants have introduced partial automation monitoring systems, most of them focus on simple monitoring of equipment operating parameters, lack of in-depth analysis of operating data and intelligent diagnosis capabilities, and it is difficult to effectively detect potential problems with the equipment. Therefore, in response to the above problems, the present disclosure provides a remote monitoring method and system for hydropower plant equipment to improve the monitoring efficiency and safety of hydropower plant equipment, reduce maintenance costs, and improve the overall operating efficiency of hydropower plants.
[0063] One embodiment of the present disclosure relates to a method for remote monitoring of hydropower plant equipment, the process of which is as follows: Figure 1 As shown, the data flow is as follows Figure 2 shown.
[0064] Combined Figure 1 and Figure 2 The method for remote monitoring of hydropower plant equipment provided in this embodiment includes the following steps:
[0065] Step S1, obtaining the operating vibration signal and the operating sound signal of the monitored hydropower plant equipment.
[0066] Specifically, when the internal parts of the hydropower plant equipment are worn, loose, unbalanced or have other faults, they are first manifested in the form of vibration and sound. Therefore, this embodiment monitors the operating status of the equipment by collecting the operating vibration signal and operating sound signal of the hydropower plant equipment, so as to timely discover potential abnormal conditions. Among them, the operating vibration signal can be collected by a vibration sensor set in the hydropower plant equipment, and the operating sound signal can be collected by a sound sensor set in the hydropower plant equipment.
[0067] Step S2, sending the operation vibration signal and the operation sound signal to a device remote monitoring server.
[0068] Specifically, step S2 can transmit the operation vibration signal and the operation sound signal to the equipment remote monitoring server through a wireless communication module that is communicatively connected to the equipment remote monitoring server, so as to realize remote monitoring and centralized data analysis of the equipment operation status.
[0069] Step S3, using the equipment remote monitoring server to respectively extract the time-frequency characteristics of the operation vibration signal and the operation sound signal to obtain a vibration time-frequency characteristic graph and a sound time-frequency characteristic graph.
[0070] Exemplarily, in step S3, respectively extracting the time-frequency features of the operation vibration signal and the operation sound signal to obtain a vibration time-frequency feature graph and a sound time-frequency feature graph includes:
[0071] Wavelet analysis is performed on the operation vibration signal and the operation sound signal to obtain a vibration two-dimensional time-frequency diagram and a sound two-dimensional time-frequency diagram. That is, wavelet analysis is performed on the operation vibration signal and the operation sound signal, and the signal is decomposed into components of different scales and frequencies by projecting the signal onto a series of wavelet basis functions to obtain a two-dimensional time-frequency representation of the signal, namely, a vibration two-dimensional time-frequency diagram and a sound two-dimensional time-frequency diagram, thereby intuitively showing the changes of the vibration signal and the sound signal in the two dimensions of time and frequency, and providing a more intuitive information basis for the diagnosis of abnormal operation of the equipment.
[0072] The vibration two-dimensional time-frequency graph is input into a vibration time-frequency feature extractor based on the first atrous convolutional neural network model to obtain the vibration time-frequency feature graph. Similarly, the sound two-dimensional time-frequency graph is input into a sound time-frequency feature extractor based on the second atrous convolutional neural network model to obtain the sound time-frequency feature graph.
[0073] That is to say, step S3 can use a deep learning algorithm to extract more useful time-frequency feature representations from the vibration two-dimensional time-frequency diagram and the sound two-dimensional time-frequency diagram. At the same time, considering that the vibration signal is more related to the mechanical structure and motion state of the equipment, while the sound signal may contain more acoustic characteristics and environmental noise, therefore, in order to better adapt to different types of signal characteristics, this embodiment uses a first hole convolutional neural network model and a second hole convolutional neural network model of different scales as a vibration time-frequency feature extractor and a sound time-frequency feature extractor to process the vibration two-dimensional time-frequency diagram and the sound two-dimensional time-frequency diagram, and uses a hole convolution operation to capture the long-distance dependency in the two-dimensional time-frequency diagram, extract possible abnormal time-frequency feature patterns in the vibration signal and the sound signal, such as frequency offset, energy mutation, etc., so as to obtain a vibration time-frequency feature diagram and a sound time-frequency feature diagram.
[0074] Step S4, using the device remote monitoring server to perform feature selection optimization on the sound time-frequency feature graph and the vibration time-frequency feature graph respectively, to obtain a sound time-frequency enhancement feature graph and a vibration time-frequency enhancement feature graph.
[0075] Considering that the operating vibration signal and the operating sound signal may be interfered by noise during the acquisition process, resulting in redundant information in the feature map, in order to further improve the feature quality, step S4 can introduce a feature selection optimization module based on the feature energy distribution spectrum, and input the sound time-frequency feature map and the vibration time-frequency feature map into the feature selection optimization module based on the feature energy distribution spectrum to obtain the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map. The feature selection optimization module based on the feature energy distribution spectrum is used to select a useful feature subset from the original features, while removing the interference of redundant and irrelevant features, thereby improving the accuracy and efficiency of fault diagnosis.
[0076] Exemplarily, in step S4, feature selection optimization is performed on the sound time-frequency feature graph and the vibration time-frequency feature graph respectively to obtain the sound time-frequency enhancement feature graph and the vibration time-frequency enhancement feature graph, which may include: inputting the sound time-frequency feature graph and the vibration time-frequency feature graph respectively into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain the sound time-frequency enhancement feature graph and the vibration time-frequency enhancement feature graph.
[0077] Among them, the feature selection optimization module based on the characteristic energy distribution spectrum first splits the input feature map into a set of local feature vectors, and reveals the distribution of each local feature vector at different energy levels by calculating the characteristic energy distribution spectrum vector of each local feature vector. Then, the characteristic energy distribution spectrum vector of each local feature vector is further calculated according to the position mean to obtain the energy spectrum distribution center vector, and this is used as reference information. By calculating the difference between the energy spectrum distribution center vector and each characteristic energy distribution spectrum vector, the degree of closeness of each local feature vector to the overall distribution is evaluated, and the corresponding energy distribution spectrum span factor is generated. Then, based on the energy distribution spectrum span factor, the original set of local feature vectors is screened to remove redundant information that deviates from the overall distribution, retain useful features close to the overall distribution, and re-aggregate the selected local features to generate the final sound time-frequency enhancement feature map and vibration time-frequency enhancement feature map.
[0078] That is to say, the sound time-frequency feature map and the vibration time-frequency feature map are respectively input into the feature selection optimization module based on the characteristic energy distribution spectrum to obtain the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map, which specifically includes: splitting the sound time-frequency feature map into a set of sound time-frequency local feature vectors; calculating the characteristic energy distribution spectrum vector of each sound time-frequency local feature vector in the set of sound time-frequency local feature vectors to obtain a set of sound time-frequency characteristic energy distribution spectrum vectors; calculating the positional mean vector of the set of sound time-frequency characteristic energy distribution spectrum vectors as the energy spectrum distribution center vector; calculating the energy spectrum distribution vector. The Mahalanobis distance between the center vector and each sound time-frequency feature energy distribution spectrum vector in the set of the sound time-frequency feature energy distribution spectrum vector is used as the energy distribution spectrum span factor to obtain a set of energy distribution spectrum span factors; each energy distribution spectrum span factor in the set of energy distribution spectrum span factors is compared with a preset threshold, and the sound time-frequency local feature vectors corresponding to the energy distribution spectrum span factors less than the preset threshold are selected to form a set of selected sound time-frequency local feature vectors; the set of selected sound time-frequency local feature vectors is trans-aggregated in a feature splitting manner to obtain the sound time-frequency enhanced feature map. The sound time-frequency feature map in the above process is replaced by a vibration time-frequency feature map to obtain a vibration time-frequency enhanced feature map.
[0079] Among them, calculating the characteristic energy distribution spectrum vector of each sound time-frequency local feature vector in the set of sound time-frequency local feature vectors to obtain the set of sound time-frequency characteristic energy distribution spectrum vectors includes: calculating the characteristic distribution energy synergy representation vector between the i-th sound time-frequency local feature vector and each other sound time-frequency local feature vector in the set of sound time-frequency local feature vectors in a manner of taking the maximum eigenvalue element by element to obtain a set of characteristic distribution energy synergy representation vectors; calculating the characteristic distribution energy synergy factor of each characteristic distribution energy synergy representation vector in the set of characteristic distribution energy synergy representation vectors to obtain the characteristic energy distribution spectrum vector of the i-th sound time-frequency local feature vector composed of multiple characteristic distribution energy synergy factors.
[0080] Furthermore, the characteristic distribution energy synergy factor of each characteristic distribution energy synergy representation vector in the set of the characteristic distribution energy synergy representation vector is calculated to obtain the characteristic energy distribution spectrum vector of the i-th sound time-frequency local characteristic vector composed of multiple characteristic distribution energy synergy factors, specifically including: calculating the square of the difference between the maximum eigenvalue of the characteristic distribution energy synergy representation vector and its characteristic mean, the characteristic variance of the characteristic distribution energy synergy representation vector and the weighted sum of the bias terms, and then adding the variance of the characteristic distribution energy synergy representation vector and the bias term and dividing the result by the weighted sum to obtain the characteristic distribution energy synergy factor.
[0081] In other words, step S4 inputs the sound time-frequency feature map and the vibration time-frequency feature map into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map. The step S4 may be specifically as follows: the set of the sound time-frequency local feature vectors is processed according to the following feature selection optimization formula to obtain the selected set of the sound time-frequency local feature vectors, wherein the feature selection optimization formula is:
[0082] X i,j =max(X i , X j )
[0083]
[0084] P i ={e i,1 , e i,2 , ..., e i,n}
[0085]
[0086]
[0087] X′={V i |Dist(c,P i )≤θ}
[0088] Among them, X i and X j denote the i-th and j-th sound time-frequency local feature vectors in the set of the sound time-frequency local feature vectors, respectively; max(·,·) denotes taking the maximum eigenvalue of the two feature vectors element by element; X i,j represents the characteristic distribution energy synergy representation vector between the i-th sound time-frequency local feature vector and the j-th sound time-frequency local feature vector, μ, σ 2 and max(X i,j ) represent the characteristic mean, characteristic variance and maximum eigenvalue of the characteristic distribution energy cooperative representation vector, λ is the bias term, e i,1 、e i,2 、e i,j and e i,n Respectively represent the characteristic distribution energy synergy factors between the ith sound time-frequency local feature vector and the first, second, th and nth sound time-frequency local feature vectors in the set of the sound time-frequency local feature vectors, where the value of n is the number of the sound time-frequency local feature vectors, P irepresents the energy distribution spectrum vector of the i-th sound time-frequency feature, P represents the set of energy distribution spectrum vectors of the sound time-frequency feature, c represents the center vector of the energy spectrum distribution, S i,c represents the covariance matrix between the energy distribution spectrum vector of the i-th sound time-frequency feature and the center vector of the energy spectrum distribution, Dist(·, ·) represents the distance measurement function, D i represents the span factor of the i-th energy distribution spectrum, θ represents the preset threshold, X′ represents the set of selected sound time-frequency local feature vectors, V i Represents the selected local time-frequency feature vector of the i-th sound.
[0089] Step S5, using the equipment remote monitoring server, based on the attention joint features of the sound time-frequency enhancement feature graph and the vibration time-frequency enhancement feature graph, determine whether there is any abnormality in the working status of the monitored hydropower plant equipment.
[0090] Considering that the vibration signal mainly reflects the dynamic characteristics of the equipment structure and the interaction between mechanical components, while the sound signal contains the acoustic characteristics of the equipment during operation and potential fault source information, the two reflect different aspects of the equipment's operating status and have certain correlations. Therefore, in order to achieve more comprehensive equipment status monitoring and fault diagnosis, step S5 can introduce an attention-driven feature joint saliency perception network to jointly analyze the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map, and mine the correlation dependency between the vibration signal and the sound signal based on the dual attention interaction mechanism, so as to achieve deep fusion of features.
[0091] For example, Figure 3 As shown, in step S5, based on the attention joint features of the sound time-frequency enhancement feature graph and the vibration time-frequency enhancement feature graph, it is determined whether the working state of the monitored hydropower plant equipment is abnormal, including:
[0092] Step S51, inputting the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map into an attention-driven feature joint saliency perception network to obtain a sound-vibration time-frequency joint saliency perception feature map.
[0093] Specifically, the attention-driven feature joint saliency perception network can first reshape the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map into feature matrices so that their dimensions and shapes are adapted to the input requirements of the attention mechanism. Then, by calculating the bidirectional dependency between the two sets of feature matrices, the interaction between the features is quantified, and the sound-vibration and vibration-sound dependency matrices are generated, which serve as a guide for feature attention optimization fusion. At the same time, in order to reduce the model's sensitivity to noise and prevent overfitting, the two sets of dependency matrices are further pruned to enhance the generalization ability of the model by selectively ignoring or suppressing a part of the feature interaction. Furthermore, based on the pruned two sets of dependency matrices, the two sets of feature matrices after reshaping are weighted optimized, and then restored to the original feature map form by reshaping and weighted fusion to generate a sound-vibration time-frequency joint saliency perception feature map, thereby achieving a comprehensive description of the operating status of the hydropower plant equipment.
[0094] Exemplarily, step S51 specifically includes: reshaping the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map to obtain a sound time-frequency enhancement feature shape reshaping matrix and a vibration time-frequency enhancement feature shape reshaping matrix; inputting the sound time-frequency enhancement feature shape reshaping matrix and the vibration time-frequency enhancement feature shape reshaping matrix into a feature channel-by-channel interactive perception module to obtain a sound-vibration dependency matrix and a vibration-sound dependency matrix; inputting the sound-vibration dependency matrix and the vibration-sound dependency matrix into a random inactivation module to obtain a pruned sound-vibration dependency matrix and a pruned vibration-sound dependency matrix; and inputting the pruned sound-vibration dependency matrix into a random inactivation module to obtain a pruned sound-vibration dependency matrix and a pruned vibration-sound dependency matrix. The dependency matrix is matrix-multiplied with the vibration time-frequency enhancement feature shape reshaping matrix to obtain the dependency optimized vibration time-frequency enhancement feature matrix; the pruned vibration-sound dependency matrix is matrix-multiplied with the sound time-frequency enhancement feature shape reshaping matrix to obtain the dependency optimized sound time-frequency enhancement feature matrix; the dependency optimized sound time-frequency enhancement feature matrix and the dependency optimized vibration time-frequency enhancement feature matrix are feature-reshaped to obtain an optimized sound time-frequency enhancement feature map and an optimized vibration time-frequency enhancement feature map; the weighted sum of the optimized sound time-frequency enhancement feature map and the optimized vibration time-frequency enhancement feature map is calculated to obtain the multi-scale joint perception significant feature map of the plant state.
[0095] Among them, the sound time-frequency enhancement feature shape reshaping matrix and the vibration time-frequency enhancement feature shape reshaping matrix are input into the feature channel-by-channel interactive perception module to obtain the sound-vibration dependency matrix and the vibration-sound dependency matrix, which further includes: calculating the sound time-frequency enhancement feature shape reshaping matrix multiplied by the transposed matrix of the vibration time-frequency enhancement feature shape reshaping matrix to obtain the sound-vibration association representation matrix; dividing the sound-vibration association representation matrix by the scale of the vibration time-frequency enhancement feature shape reshaping matrix and then inputting it into the softmax function to obtain the sound-vibration dependency matrix; calculating the transposed matrix of the sound time-frequency enhancement feature shape reshaping matrix multiplied by the vibration time-frequency enhancement feature shape reshaping matrix to obtain the plant state vibration-sound association representation matrix; dividing the vibration-sound association representation matrix by the scale of the sound time-frequency enhancement feature shape reshaping matrix and then inputting it into the softmax function to obtain the vibration-sound dependency matrix.
[0096] In other words, step S51 inputs the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map into the attention-driven feature joint saliency perception network to obtain the sound-vibration time-frequency joint saliency perception feature map, which can be specifically as follows: according to the following joint perception formula, the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map are processed to obtain the sound-vibration time-frequency joint saliency perception feature map, wherein the joint perception formula is:
[0097] M 1 =Reshape(F 1 )
[0098] M 2 =Reshape(F 2 )
[0099]
[0100]
[0101] W 21 = Dropout(M 21 )
[0102] W 12 = Dropout(M 12 )
[0103]
[0104]
[0105] F 1 '=Reshape(M 1 ′)
[0106] F 2 '=Reshape(M 2 ′)
[0107] F c =αF 1 +βF 2
[0108] Among them, F 1 represents the sound time-frequency enhancement feature map, F 2 represents the vibration time-frequency enhancement feature map, reshape(·) represents the feature shape reshaping, M 1 Represents the sound time-frequency enhancement feature reshape matrix, M 2 represents the vibration time-frequency enhanced characteristic shape reshaping matrix, represents matrix multiplication, (·) T represents the transpose of the matrix, S represents the scale of the sound time-frequency enhancement feature shape reshaping matrix, that is, the width of the sound time-frequency enhancement feature shape reshaping matrix multiplied by the height, and the scale of the sound time-frequency enhancement feature shape reshaping matrix and the vibration time-frequency enhancement feature shape reshaping matrix are the same, Softmax is a normalized exponential function, M 21 represents the sound-vibration dependency matrix, M 12 represents the vibration-sound dependency matrix, Dropout(·) represents random inactivation processing, W 21 represents the pruned sound-vibration dependency matrix, W 12 represents the pruned vibration-sound dependency matrix, M 1 ′ represents the dependency relationship optimization sound time-frequency enhancement feature matrix, M 2 ′ represents the dependency relationship optimization vibration time-frequency enhancement feature matrix, F 1 ' represents the optimized sound time-frequency enhancement feature map, F 2 ' represents the optimized vibration time-frequency enhancement feature map, α and β represent different weight coefficients, F c Represents the sound-vibration time-frequency joint significant perceptual feature map.
[0109] Step S52, inputting the sound-vibration time-frequency joint significant perception feature map into a monitoring result generation module based on a classifier to obtain a monitoring result, wherein the monitoring result is used to indicate whether there is an abnormality in the working state of the monitored hydropower plant equipment.
[0110] Specifically, the labels of the classifier may include the normal working state of the hydropower plant equipment (first label), and the abnormal working state of the hydropower plant equipment (second label), wherein the classifier determines which classification label it belongs to through the features in the sound-vibration time-frequency joint significant perception feature map. Step S52 may expand the sound-vibration time-frequency joint significant perception feature map into a classification feature vector based on a row vector or a column vector; use multiple fully connected layers of the classifier to fully connect encode the classification feature vector to obtain an encoded classification feature vector; pass the encoded classification feature vector through the Softmax classification function of the classifier to obtain the monitoring result.
[0111] Compared with the prior art, the remote monitoring method for hydropower plant equipment provided in the embodiment of the present disclosure utilizes the vibration signals and sound signals during the operation of the hydropower plant equipment, and combines the Internet of Things technology and deep learning technology to extract time-frequency features and perform correlation interaction analysis on the vibration signals and sound signals of the hydropower plant equipment, so as to mine potential abnormal information of the equipment operation status, thereby realizing remote monitoring of the operation status of the hydropower plant equipment and intelligent fault diagnosis, effectively improving the monitoring efficiency and safety of the hydropower plant equipment, reducing maintenance costs, and improving the overall operation efficiency of the hydropower plant.
[0112] Another embodiment of the present disclosure relates to a remote monitoring system 300 for hydropower plant equipment, such as Figure 4 As shown, including:
[0113] The signal acquisition module 310 is used to acquire the operating vibration signal and the operating sound signal of the monitored hydropower plant equipment.
[0114] The signal transmission module 320 is used to transmit the operation vibration signal and the operation sound signal to the equipment remote monitoring server.
[0115] The time-frequency feature extraction module 330 is used to use the device remote monitoring server to extract the time-frequency features of the operation vibration signal and the operation sound signal respectively, and obtain a vibration time-frequency feature graph and a sound time-frequency feature graph.
[0116] The feature selection optimization module 340 is used to utilize the device remote monitoring server to perform feature selection optimization on the sound time-frequency feature graph and the vibration time-frequency feature graph respectively to obtain a sound time-frequency enhancement feature graph and a vibration time-frequency enhancement feature graph.
[0117] The abnormality judgment module 350 is used to use the equipment remote monitoring server to determine whether there is an abnormality in the working state of the monitored hydropower plant equipment based on the attention joint features of the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map.
[0118] The specific implementation method of the remote monitoring system for hydropower plant equipment provided in the embodiment of the present disclosure can be found in the remote monitoring method for hydropower plant equipment provided in the embodiment of the present disclosure, and will not be repeated here.
[0119] Exemplarily, the remote monitoring system 300 for hydropower plant equipment can be implemented in various wireless terminals, such as a server with a remote monitoring algorithm for hydropower plant equipment based on the Internet of Things technology. In a possible implementation, the remote monitoring system 300 for hydropower plant equipment can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the remote monitoring system 300 for hydropower plant equipment can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the remote monitoring system 300 for hydropower plant equipment can also be one of the many hardware modules of the wireless terminal.
[0120] In another example, the hydropower plant equipment remote monitoring system 300 and the wireless terminal may also be separate devices, and the hydropower plant equipment remote monitoring system 300 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in an agreed data format.
[0121] Compared with the prior art, the remote monitoring system for hydropower plant equipment provided by the embodiments of the present invention utilizes the vibration signals and sound signals during the operation of the hydropower plant equipment, and combines the Internet of Things technology and deep learning technology to extract time-frequency features and perform correlation interaction analysis on the vibration signals and sound signals of the hydropower plant equipment, so as to mine potential abnormal information of the equipment operation status, thereby realizing remote monitoring of the operation status of the hydropower plant equipment and intelligent fault diagnosis, effectively improving the monitoring efficiency and safety of the hydropower plant equipment, reducing maintenance costs, and improving the overall operation efficiency of the hydropower plant.
[0122] The basic principles of the technical solutions provided by the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present disclosure. In addition, the specific details of the above embodiments are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the present disclosure to being implemented by adopting the above specific details.
[0123] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.
[0124] In addition, each functional module in each embodiment of the present disclosure may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0125] Those skilled in the art will appreciate that the above-mentioned embodiments are specific embodiments for implementing the present disclosure, and in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present disclosure.
[0126] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units included in the system can also be implemented by one unit through software or hardware.
[0127] Finally, it should be noted that the above description has been given for the purpose of illustration and description. In addition, the above embodiments are only used to illustrate the technical solution of the present disclosure and are not intended to limit it. Although the present disclosure is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present disclosure can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present disclosure.
Claims
1. A method for remote monitoring of hydropower plant equipment, characterized in that: The remote monitoring method for hydropower plant equipment comprises: Obtain the operating vibration signal and operating sound signal of the monitored hydropower plant equipment; Sending the operation vibration signal and the operation sound signal to a device remote monitoring server; Utilizing the equipment remote monitoring server, respectively extracting the time-frequency characteristics of the operation vibration signal and the operation sound signal to obtain a vibration time-frequency characteristic graph and a sound time-frequency characteristic graph; Using the device remote monitoring server, feature selection optimization is performed on the sound time-frequency feature graph and the vibration time-frequency feature graph respectively to obtain a sound time-frequency enhancement feature graph and a vibration time-frequency enhancement feature graph; By utilizing the equipment remote monitoring server, it is determined whether there is any abnormality in the working state of the monitored hydropower plant equipment based on the attention joint features of the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map.
2. The method for remote monitoring of hydropower plant equipment according to claim 1, characterized in that: Extracting the time-frequency features of the operation vibration signal and the operation sound signal respectively to obtain a vibration time-frequency feature graph and a sound time-frequency feature graph, including: Performing wavelet analysis on the operation vibration signal and the operation sound signal to obtain a vibration two-dimensional time-frequency graph and a sound two-dimensional time-frequency graph; Inputting the vibration two-dimensional time-frequency graph into a vibration time-frequency feature extractor based on a first hole convolutional neural network model to obtain the vibration time-frequency feature graph; The sound two-dimensional time-frequency graph is input into a sound time-frequency feature extractor based on a second hole convolutional neural network model to obtain the sound time-frequency feature graph.
3. The method for remote monitoring of hydropower plant equipment according to claim 1, characterized in that: The sound time-frequency feature map and the vibration time-frequency feature map are respectively subjected to feature selection optimization to obtain a sound time-frequency enhancement feature map and a vibration time-frequency enhancement feature map, including: The sound time-frequency feature map and the vibration time-frequency feature map are respectively input into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map.
4. The method for remote monitoring of hydropower plant equipment according to claim 3, characterized in that: The sound time-frequency feature map and the vibration time-frequency feature map are respectively input into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map, including: Splitting the sound time-frequency feature map into a set of sound time-frequency local feature vectors; Calculating the characteristic energy distribution spectrum vector of each sound time-frequency local feature vector in the set of sound time-frequency local feature vectors to obtain a set of sound time-frequency characteristic energy distribution spectrum vectors; Calculating the position mean vector of the set of energy distribution spectrum vectors of the sound time-frequency characteristics as the center vector of the energy spectrum distribution; Calculating the Mahalanobis distance between the energy spectrum distribution center vector and each sound time-frequency feature energy distribution spectrum vector in the set of the sound time-frequency feature energy distribution spectrum vectors as the energy distribution spectrum span factor, and obtaining a set of energy distribution spectrum span factors; Comparing each energy distribution spectrum span factor in the set of energy distribution spectrum span factors with a preset threshold, and selecting sound time-frequency local feature vectors corresponding to energy distribution spectrum span factors less than the preset threshold to form a set of selected sound time-frequency local feature vectors; The set of the selected sound time-frequency local feature vectors is trans-aggregated in a feature splitting manner to obtain the sound time-frequency enhanced feature map.
5. The method for remote monitoring of hydropower plant equipment according to claim 4, characterized in that: Calculating the characteristic energy distribution spectrum vector of each sound time-frequency local feature vector in the set of sound time-frequency local feature vectors to obtain a set of sound time-frequency characteristic energy distribution spectrum vectors, including: Calculate the feature distribution energy synergistic representation vector between the i-th sound time-frequency local feature vector and each other sound time-frequency local feature vector in the set of the sound time-frequency local feature vectors in a manner of taking the maximum eigenvalue element by element, and obtain a set of feature distribution energy synergistic representation vectors; The characteristic distribution energy synergy factor of each characteristic distribution energy synergy representation vector in the set of characteristic distribution energy synergy representation vectors is calculated to obtain the characteristic energy distribution spectrum vector of the i-th sound time-frequency local characteristic vector composed of multiple characteristic distribution energy synergy factors.
6. The method for remote monitoring of hydropower plant equipment according to claim 5, characterized in that: Calculating the characteristic distribution energy synergy factor of each characteristic distribution energy synergy representation vector in the set of characteristic distribution energy synergy representation vectors to obtain the characteristic energy distribution spectrum vector of the i-th sound time-frequency local characteristic vector composed of multiple characteristic distribution energy synergy factors, including: The square of the difference between the maximum eigenvalue of the feature distribution energy synergistic representation vector and its eigenmean, the feature variance of the feature distribution energy synergistic representation vector, and the weighted sum of the bias terms are calculated, and then the variance of the feature distribution energy synergistic representation vector and the bias term are added and divided by the weighted sum to obtain the feature distribution energy synergy factor.
7. The method for remote monitoring of hydropower plant equipment according to claim 1, characterized in that: Based on the attention joint feature of the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map, determining whether the working state of the monitored hydropower plant equipment is abnormal includes: Inputting the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map into an attention-driven feature joint saliency perception network to obtain a sound-vibration time-frequency joint saliency perception feature map; The sound-vibration time-frequency joint significant perception feature map is input into a classifier-based monitoring result generation module to obtain a monitoring result, which is used to indicate whether there is an abnormality in the working state of the monitored hydropower plant equipment.
8. The method for remote monitoring of hydropower plant equipment according to claim 7, characterized in that: Inputting the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map into an attention-driven feature joint saliency perception network to obtain a sound-vibration time-frequency joint saliency perception feature map, including: Performing feature shape reshaping on the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map to obtain a sound time-frequency enhancement feature shape reshaping matrix and a vibration time-frequency enhancement feature shape reshaping matrix; Input the sound time-frequency enhancement feature shape reshaping matrix and the vibration time-frequency enhancement feature shape reshaping matrix into a feature channel-by-channel interactive perception module to obtain a sound-vibration dependency matrix and a vibration-sound dependency matrix; Inputting the sound-vibration dependency matrix and the vibration-sound dependency matrix into a random dropout module to obtain a pruned sound-vibration dependency matrix and a pruned vibration-sound dependency matrix; Performing matrix multiplication of the pruned sound-vibration dependency matrix and the vibration time-frequency enhancement feature shape reshaping matrix to obtain a dependency-optimized vibration time-frequency enhancement feature matrix; Performing matrix multiplication of the pruned vibration-sound dependency matrix and the sound time-frequency enhancement feature shape reshaping matrix to obtain a dependency-optimized sound time-frequency enhancement feature matrix; Reshaping the dependency-optimized sound time-frequency enhancement feature matrix and the dependency-optimized vibration time-frequency enhancement feature matrix to obtain an optimized sound time-frequency enhancement feature map and an optimized vibration time-frequency enhancement feature map; The weighted sum of the optimized sound time-frequency enhancement feature map and the optimized vibration time-frequency enhancement feature map is calculated to obtain the multi-scale joint perception significant feature map of the plant state.
9. The method for remote monitoring of hydropower plant equipment according to claim 8, characterized in that: The sound time-frequency enhancement feature shape reshaping matrix and the vibration time-frequency enhancement feature shape reshaping matrix are input into the feature channel-by-channel interactive perception module to obtain a sound-vibration dependency matrix and a vibration-sound dependency matrix, including: Calculate the transposed matrix of the sound time-frequency enhancement characteristic shape reshaping matrix multiplied by the vibration time-frequency enhancement characteristic shape reshaping matrix to obtain a sound-vibration association representation matrix; The sound-vibration association representation matrix is divided by the scale of the vibration time-frequency enhancement feature shape reshaping matrix and then input into a softmax function to obtain the sound-vibration dependency matrix; Calculate the transposed matrix of the sound time-frequency enhancement characteristic shape reshaping matrix and multiply it by the vibration time-frequency enhancement characteristic shape reshaping matrix to obtain a plant state vibration-sound correlation representation matrix; The vibration-sound association representation matrix is divided by the scale of the sound time-frequency enhancement feature shape reshaping matrix and then input into the softmax function to obtain the vibration-sound dependency matrix.
10. A remote monitoring system for hydropower plant equipment, characterized in that: The hydropower plant equipment remote monitoring system comprises: A signal acquisition module, used to acquire operating vibration signals and operating sound signals of monitored hydropower plant equipment; A signal transmission module, used for transmitting the operation vibration signal and the operation sound signal to a device remote monitoring server; A time-frequency feature extraction module, used to use the device remote monitoring server to extract the time-frequency features of the operation vibration signal and the operation sound signal respectively, and obtain a vibration time-frequency feature graph and a sound time-frequency feature graph; A feature selection optimization module, used to use the device remote monitoring server to perform feature selection optimization on the sound time-frequency feature graph and the vibration time-frequency feature graph respectively, to obtain a sound time-frequency enhancement feature graph and a vibration time-frequency enhancement feature graph; The abnormality judgment module is used to use the equipment remote monitoring server to determine whether there is an abnormality in the working state of the monitored hydropower plant equipment based on the attention joint features of the sound time-frequency enhancement feature map and the vibration time-frequency enhancement feature map.
Citation Information
Patent Citations
Visual inspection system and method for new energy automobile manufacturing
CN118941558A
Hydraulic power plant safety protection method and system
CN119229590A
Multi-level network identity verification and access control system
CN119337353A
Risk identification method and system in construction process of cement fiberboard
CN119672638A
Cited By
Earthquake early warning information issuing method based on 4G telephone terminal
CN120913347A
