Tower bolt loosening diagnosis method, system, equipment and medium

Abstract

The invention relates to the technical field of electric power systems, in particular to a tower bolt looseness diagnosis method, system, equipment and medium, according to the method, a vibration signal is decomposed through a variational mode decomposition algorithm based on spider wasp optimization to obtain a plurality of mode components, so that the global optimization performance is improved, and the reliability of the tower bolt looseness diagnosis is improved. Signal reconstruction is carried out by screening out a plurality of modal components with higher correlation with the vibration signal to obtain a reconstructed signal, a signal feature threshold distance is determined according to the reconstructed signal and a preset normal vibration signal, and the signal feature distance of the bolt position is compared with the signal feature threshold distance to obtain a signal feature threshold distance of the bolt position. The bolt positions with the signal characteristic distances larger than the signal characteristic threshold distance are screened out and judged as the bolt looseness risk positions, the tower bolt looseness diagnosis efficiency is improved, judgment is made in advance for on-site iron tower local bolt looseness detection, and the applicability is good.

Description

Technical Field

[0001] The invention relates to the technical field of power systems, and in particular to a method, system, equipment and medium for diagnosing loose bolts of a tower. Background Art

[0002] As an important part of the power network, transmission towers play an important role in ensuring the safe operation of the power grid. Transmission towers are generally made of galvanized angle steel structures and connected with eccentric bolts. They are very flexible structures. Because the towers stand in the wind field for a long time, the bolts of the transmission towers will become loose or fall off due to long-term wind vibration and conductor dancing. The large number of bolt connection nodes on the towers makes it difficult for maintenance personnel to determine the looseness of the bolts of each rod and to repair each loose bolt in time. When the number of loose bolts is large enough, serious tower collapse accidents may even occur, causing immeasurable losses to the power grid. Therefore, it is very important to accurately detect and diagnose the tightness of the bolts on the transmission towers.

[0003] Traditional tower bolt loosening diagnosis methods have low recognition efficiency for minor structural damage and poor adaptability, making it difficult to accurately detect and diagnose the tightness of transmission tower bolts. Summary of the invention

[0004] In view of this, the present invention provides a pole tower bolt loosening diagnosis method, system, equipment and medium, which solves the technical problems that the traditional pole tower bolt loosening diagnosis method has low recognition efficiency for minor structural damage and poor adaptability, and is difficult to accurately detect and diagnose the tightness of transmission tower bolts.

[0005] A first aspect of the present invention provides a method for diagnosing loose bolts of a tower, comprising:

[0006] Collect vibration signals of target transmission towers;

[0007] Decomposing the vibration signal based on a variational modal decomposition algorithm optimized by spider wasps to obtain a number of modal components;

[0008] Screening the correlations between a plurality of the modal components and the vibration signal, and screening out a plurality of the modal components with a higher correlation;

[0009] Performing signal reconstruction on the plurality of modal components to obtain a reconstructed signal;

[0010] Determining a signal characteristic threshold distance according to the reconstructed signal and a preset normal vibration signal;

[0011] Determine signal characteristic distances of multiple bolt positions according to vibration signals of multiple bolt positions of the target transmission tower;

[0012] The signal feature distance of each bolt position is compared with the signal feature threshold distance, and the bolt positions whose signal feature distance is greater than the signal feature threshold distance are screened out and determined as bolt loosening risk positions.

[0013] Preferably, collecting the vibration signal of the target transmission tower includes:

[0014] Performing multiple pulse excitations on multiple preset excitation positions on the target transmission tower, and acquiring vibration acceleration signals at each bolt of the target transmission tower;

[0015] Establishing a frequency response function according to the vibration acceleration signal and the pulse excitation signal;

[0016] Calculating the mean of the multiple frequency response functions at each of the preset excitation positions to determine an average frequency response function;

[0017] A natural frequency is determined as a vibration signal based on the average frequency response function.

[0018] Preferably, the variational modal decomposition algorithm based on spider wasp optimization decomposes the vibration signal to obtain a plurality of modal components, including:

[0019] Optimizing the network parameters of the variational mode decomposition algorithm using a spider wasp optimization algorithm, and optimizing the network parameters of the variational mode decomposition algorithm using the optimization result;

[0020] The vibration signal is decomposed using an optimized variational modal decomposition algorithm to obtain a number of modal components.

[0021] Preferably, the optimizing the network parameters of the variational mode decomposition algorithm by using the spider wasp optimization algorithm comprises:

[0022] Initializing population parameters of the spider-wasp optimization algorithm, position parameters of the spider swarm, maximum number of iterations, and initial fitness values, wherein the position of each swarm represents a network parameter of the variational mode decomposition algorithm;

[0023] Solve the search space using a population-guiding mechanism;

[0024] updating the position parameters of the swarm using the sine-cosine transformed inertia weights to simulate the swarm's exploration behavior in the search space;

[0025] The fitness value is determined by the fitness function according to the location parameters of the swarm;

[0026] Determine whether the fitness value has reached the optimum or whether the iteration has reached the maximum number of iterations. If it is determined that the fitness value has not reached the optimum or the iteration has not reached the maximum number of iterations, proceed to the step of updating the position parameters of the bee colony using the sine-cosine transform inertia weight to simulate the exploration behavior of the bee colony in the search space;

[0027] If it is determined that the fitness value reaches the optimum or the iteration reaches the maximum number of iterations, the optimization process is stopped and the current position parameters of the swarm are output as the optimal network parameters of the variational mode decomposition algorithm.

[0028] Preferably, the step of screening the correlations between the plurality of modal components and the vibration signal to screen out a plurality of modal components with higher correlations comprises:

[0029] Determine the correlation coefficients between the modal components and the vibration signal respectively using the Pearson correlation coefficient analysis method;

[0030] A plurality of the modal components whose correlation coefficients are greater than a preset correlation coefficient threshold are screened out.

[0031] Preferably, the performing signal reconstruction on the plurality of modal components to obtain a reconstructed signal comprises:

[0032] Performing three-layer wavelet packet decomposition on each of the modal components using wavelet basis functions;

[0033] Taking energy logarithmic entropy as a cost function, using the cost function to determine the cost function size of each node coefficient obtained by wavelet packet decomposition, and determining the minimum wavelet packet basis;

[0034] A soft threshold function is selected based on an adaptive threshold criterion, and wavelet packet reconstruction is performed using the soft threshold function to obtain a reconstructed signal.

[0035] Preferably, determining the signal characteristic threshold distance according to the reconstructed signal and a preset normal vibration signal includes:

[0036] Extracting characteristic values ​​from the reconstructed signal to form a characteristic vector;

[0037] Determine a characteristic vector standard value and a characteristic vector variance value according to the characteristic vector;

[0038] The Mahalanobis distance between the reconstructed signal and the preset normal vibration signal is determined as the signal feature threshold distance according to the feature vector standard value, the feature vector variance value and the feature vector standard value of the preset normal vibration signal.

[0039] In a second aspect, the present invention further provides a tower bolt loosening diagnosis system, comprising:

[0040] A signal acquisition module, used to collect vibration signals of target transmission towers;

[0041] A signal decomposition module, used for decomposing the vibration signal based on a variational modal decomposition algorithm optimized by spider wasps to obtain a plurality of modal components;

[0042] A component screening module, used for screening the correlation between a plurality of the modal components and the vibration signal, and screening out a plurality of the modal components with a higher correlation;

[0043] A signal reconstruction module, used for performing signal reconstruction on the plurality of modal components to obtain a reconstructed signal;

[0044] A threshold distance determination module, used to determine a signal feature threshold distance according to the reconstructed signal and a preset normal vibration signal;

[0045] A characteristic distance determination module, used to determine the signal characteristic distances of multiple bolt positions according to the vibration signals of multiple bolt positions of the target transmission tower;

[0046] The bolt loosening identification module is used to compare the signal feature distance of each bolt position with the signal feature threshold distance, and screen out the bolt positions whose signal feature distance is greater than the signal feature threshold distance to determine them as bolt loosening risk positions.

[0047] In a third aspect, the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the tower bolt loosening diagnosis method as described in the first aspect.

[0048] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the steps of the method for diagnosing loosening of tower bolts as described in the first aspect are implemented.

[0049] It can be seen from the above technical scheme that the present invention decomposes the vibration signal through the variational modal decomposition algorithm based on spider wasp optimization to obtain several modal components, thereby improving the global optimization performance, and reconstructs the signal by screening out multiple modal components with high correlation with the vibration signal to obtain a reconstructed signal, determines the signal characteristic threshold distance according to the reconstructed signal and the preset normal vibration signal, and compares the signal characteristic distance of each bolt position with the signal characteristic threshold distance, and screens out the bolt position with a signal characteristic distance greater than the signal characteristic threshold distance, and determines it as the risk position of bolt loosening, thereby improving the efficiency of tower bolt loosening diagnosis, making early judgment for on-site local bolt loosening detection of the tower, and having good applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0051] Figure 1 A flowchart of a method for diagnosing loose bolts of a tower provided by an embodiment of the present invention;

[0052] Figure 2 A schematic diagram of the structure of a tower bolt loosening diagnosis system provided by an embodiment of the present invention;

[0053] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0055] The pole tower bolt loosening diagnosis method provided in the embodiment of the present application can be applied to the situation of pole tower bolt loosening diagnosis. The method can be executed by a pole tower bolt loosening diagnosis device. The pole tower bolt loosening diagnosis device can be implemented in the form of hardware and / or software. The pole tower bolt loosening diagnosis device can be configured in a computer device.

[0056] like Figure 1 As shown, the embodiment of the present application provides a method for diagnosing loose bolts of a tower, comprising the following steps S1 to S7. Among them:

[0057] Step S1: collecting vibration signals of target transmission towers.

[0058] Specifically, the step S1 of collecting the vibration signal of the target transmission tower includes:

[0059] Step S101: performing multiple pulse excitations on multiple preset excitation positions on a target transmission tower, and obtaining vibration acceleration signals at each bolt of the target transmission tower.

[0060] Among them, external force can be used to excite multiple preset excitation positions on the target transmission tower.

[0061] Exemplarily, the vibration response of the transmission tower is generated by the impact provided by the hammer, and the piezoelectric acceleration sensor is adsorbed on the main material near the bolt through the magnetic base to collect the vibration acceleration signal at each bolt on the target transmission tower.

[0062] Step S102: establishing a frequency response function according to the vibration acceleration signal and the pulse excitation signal.

[0063] The frequency response function is:

[0064]

[0065] In the formula, m is the number of modes, k is the mode order, For the extreme point, is the pole conjugate value, For the remainder, is the residue conjugate value, is the natural frequency, Indicates the output response position, represents the input excitation position, and J is the imaginary unit.

[0066] It can be understood that the frequency response function describes the response of the system to different excitations. In this process, the external force hits the tower and the vibration acceleration signal is collected. The frequency response function can also be simplified as h(w)=A(w) / F(w), which means that at different frequencies w, the acceleration response A(w) changes with the excitation force F(w). The pulse excitation signal is the process of applying pulse excitation using external force.

[0067] and As an extension of h(w), the frequency response function can be calculated based on the ratio of the acceleration spectrum to the excitation force spectrum, which includes the information of the acceleration response signal and the pulse excitation signal.

[0068] Step S103: Calculate the mean of multiple frequency response functions at each preset excitation position to determine an average frequency response function.

[0069] Among them, when pulse excitation is provided by tapping, the size and position of each tap cannot be guaranteed to be exactly the same. It is necessary to tap each measurement point multiple times, take the average of the frequency response function (FRF) after the tapping, and take the average value of multiple times to reduce the error to a certain extent.

[0070] Step S104: determine the natural frequency as a vibration signal according to the average frequency response function.

[0071] Step S2: Decompose the vibration signal based on the variational modal decomposition algorithm optimized by the spider wasp to obtain a plurality of modal components.

[0072] Among them, the network parameters of the variational mode decomposition algorithm (modal number K value and quadratic penalty factor value), taking sample entropy as fitness function, given K and The range is: , ,The Spider Wasp Optimization Algorithm (SWO) is used to optimize the network parameters.

[0073] Specifically, the variational modal decomposition algorithm based on spider wasp optimization in step S2 decomposes the vibration signal to obtain several modal components, including:

[0074] Step S201: Optimize the network parameters of the variational mode decomposition algorithm using the spider wasp optimization algorithm, and optimize the network parameters of the variational mode decomposition algorithm using the optimization result.

[0075] Among them, the spider wasp optimization algorithm is used to optimize the network parameters of the variational mode decomposition algorithm, including:

[0076] Step S2011, initializing the population parameters of the spider wasp optimization algorithm, the position parameters of the spider swarm, the maximum number of iterations and the initial fitness value, wherein the position of each swarm represents a network parameter of the variational mode decomposition algorithm.

[0077] Among them, the population can be initialized using the cubic mapping method, and the process is expressed as:

[0078]

[0079] In the formula, represents the population at the previous moment, Represents the population at the next moment.

[0080] Step S2012: Solve the search space using the population guidance mechanism.

[0081] Among them, the population guidance mechanism is used to solve the search space and enhance the exploration ability of the population individuals to the surrounding space. The process is expressed as:

[0082]

[0083] In the formula, represents the solution of the population guidance mechanism, represents an individual in n-dimensional space at time t, , represents the dynamic upper and lower bounds of the entire population at f iterations, , , random number , represents the expansion and contraction control amount of the dynamic boundary.

[0084] Step S2013: Use the sine-cosine transform inertia weights to update the position parameters of the swarm to simulate the swarm's exploration behavior in the search space.

[0085] In order to further improve the convergence speed and accuracy of the algorithm, the inertia weight of the sine-cosine transform is introduced in the SWO position update, as shown in the following formula:

[0086]

[0087] In the formula, the random number , represents the inertia weight coefficient, is the current iteration number, Indicates the maximum number of iterations.

[0088] The introduced sine-cosine transform inertia weight is dynamically adjusted with the number of iterations, which helps SWO to jump out of the local optimum, thereby improving the convergence efficiency of the algorithm. The subsequent position update process is expressed as:

[0089]

[0090] In the formula, express The position parameters at the moment, express The position parameters at the moment, , are two random position parameters, is the search coefficient.

[0091] Among them, SWO search is a nonlinear optimization strategy, which cannot adjust the global search ability and local development ability well. In order to prevent the search from falling into the local optimal situation, a nonlinear convergence factor is introduced in the search coefficient. , whose expression is:

[0092]

[0093] Introduction The search coefficient after It is expressed as:

[0094]

[0095] In the formula, the random number , random number It follows a normal distribution.

[0096] Among them, spider swarm behavior: In this algorithm, "swarming", "pursuit", "hunting", and "reproduction" refer to different stages or strategies of the optimization process simulated in the algorithm:

[0097] Swarming: refers to the algorithm's multiple search agents (simulating spider bees) gathering in more promising areas of the search space, similar to how real animal groups gather in places with abundant food.

[0098] Pursuit: refers to the process of tracking the current optimal solution or a more promising solution during the search. It will adjust its position according to the rules of the algorithm, trying to approach or surpass the current optimal solution.

[0099] Hunting: It includes two phases: exploration (extensive search in the search space) and exploitation (concentrated search in promising areas).

[0100] Reproduction: refers to the process of updating the position of the search agent or generating new search agents in the algorithm. This is similar to biological reproduction, which explores the search space by generating new individuals and improves the quality of the solution.

[0101] Step S2014: determine the fitness value through a fitness function according to the location parameters of the bee colony.

[0102] Among them, the fitness function can adopt the error mean function and the like.

[0103] Step S2015, determine whether the fitness value has reached the optimum or whether the iteration has reached the maximum number of iterations. If it is determined that the fitness value has not reached the optimum or the iteration has not reached the maximum number of iterations, go to step S2013.

[0104] Step S2016: If it is determined that the fitness value has reached the optimum or the iteration has reached the maximum number of iterations, the optimization process is stopped and the current position parameters of the swarm are output as the optimal network parameters of the variational mode decomposition algorithm.

[0105] Step S202: Decompose the vibration signal using the optimized variational modal decomposition algorithm to obtain a plurality of modal components.

[0106] Step S3: screening the correlations between the multiple modal components and the vibration signal, and screening out multiple modal components with higher correlations.

[0107] Specifically, in step S3, the correlation between several modal components and the vibration signal is screened to screen out multiple modal components with higher correlation, including:

[0108] Step S301: using the Pearson correlation coefficient analysis method to determine the correlation coefficients between a plurality of modal components and the vibration signal.

[0109] Step S302: Filter out multiple modal components whose correlation coefficients are greater than a preset correlation coefficient threshold.

[0110] For example, the original vibration signal is assumed to be , the vibration signal is decomposed into K modal components by using the variational modal decomposition algorithm, namely and a residual component ,Right now:

[0111]

[0112] For the IMF components decomposed by the variational mode decomposition algorithm, select any modal component The original vibration signal is Pearson correlation coefficient analysis was performed.

[0113] Among them, The data sequence is , the data sequence of the original signal is , then their Pearson correlation coefficient is calculated as follows:

[0114]

[0115] The threshold is selected The IMF components of are determined as effective components that meet the high correlation requirements, where is the maximum correlation value between each component and the original vibration signal.

[0116] Step S4: reconstruct the signals of the multiple modal components to obtain reconstructed signals.

[0117] The embodiment of the present application performs wavelet packet denoising on the screened modal components to remove noise interference and improve the purity of the signal. By setting a threshold, the main modal components are retained and the signal is reconstructed to avoid excessive denoising.

[0118] Specifically, the signal reconstruction of multiple modal components in step S4 to obtain the reconstructed signal includes:

[0119] Step S401: perform three-layer wavelet packet decomposition on each modal component using wavelet basis functions.

[0120] Among them, the wavelet basis function and the number of decomposition levels are selected, Daubechies (dbN) is selected as the wavelet basis function, and the number of decomposition levels is three.

[0121] Step S402: Use the energy logarithmic entropy as a cost function, use the cost function to determine the cost function size of each node coefficient obtained by wavelet packet decomposition, and determine the minimum wavelet packet basis.

[0122] Among them, the optimal wavelet packet basis is determined according to the entropy value of each node. First, a cost function needs to be set. Here, the energy logarithmic entropy is selected, and then the basis that minimizes the cost function is found. The cost function size of each node coefficient is calculated according to the cost function, and the cost function size is the entropy value.

[0123] Step S403: select a soft threshold function based on the adaptive threshold criterion, and use the soft threshold function to perform wavelet packet reconstruction to obtain a reconstructed signal.

[0124] Among them, threshold processing is performed on the nodes of the wavelet packet base layer, the threshold is obtained based on the adaptive threshold criterion, and the threshold function selects the soft threshold function, which can effectively prevent excessive noise reduction.

[0125] Soft threshold function:

[0126]

[0127] In the formula, is the original decomposition coefficient, is the threshold value, is the decomposition coefficient after reconstruction.

[0128] Step S5: determining a signal feature threshold distance according to the reconstructed signal and a preset normal vibration signal.

[0129] Specifically, determining the signal feature threshold distance according to the reconstructed signal and the preset normal vibration signal in step S5 includes:

[0130] Step S501: extract eigenvalues ​​from the reconstructed signal to form a eigenvector.

[0131] Step S502: Determine the characteristic vector standard value and the characteristic vector variance value according to the characteristic vector.

[0132] Step S503: Determine the Mahalanobis distance between the reconstructed signal and the preset normal vibration signal as the signal feature threshold distance according to the feature vector standard value, the feature vector variance value and the feature vector standard value of the preset normal vibration signal.

[0133] Among them, two working conditions can be set as normal working conditions and abnormal working conditions, and the eigenvalues ​​of the reconstructed signal are extracted to form a eigenvector , and obtain the standard eigenvector of the normal working condition data set and variance var( ), by comparing the test signal and The Mahalanobis distance The working conditions are judged by the differences.

[0134] Specifically, two working conditions are set, working condition 1 is the normal state and working condition 2 is the abnormal state. A total of 50 data sets are extracted for working condition 1 and working condition 2. For the modal component of the reconstructed signal in each sensor, the feature information of each modal component is extracted to obtain a series of feature vector sets. , where each vector contains the frequency, amplitude, energy, kurtosis and other correlation information of IMF, consisting of 50 sets of data feature vectors Constructing the test feature matrix .

[0135] set up For working conditions Next The extracted feature vector representing the state is , select the working condition The mean of the response signal eigenvector is taken as the standard eigenvector of this condition. , and calculate its variance var( ), then a signal The eigenvector of and The MD between can be expressed as the state identification of bolt loosening using the Mahalanobis distance method, as shown in the following formula. Here, the mean of the characteristic vector of working condition 1 is obtained: , as the standard feature vector for state recognition. Obtain the feature vector of each reconstructed signal With the standard eigenvector Mahalanobis distance As the signal feature threshold distance.

[0136] .

[0137] Step S6: determining signal characteristic distances of multiple bolt positions according to vibration signals of multiple bolt positions of the target transmission tower.

[0138] Among them, for the vibration signals of multiple bolt positions of the target transmission tower under actual measurement, the Mahalanobis distance between the feature vectors of the multiple bolt positions and the standard feature vector can be determined as the signal feature distance according to the above steps S1 to S5.

[0139] Step S7: compare the signal feature distance of each bolt position with the signal feature threshold distance, and screen out the bolt positions whose signal feature distance is greater than the signal feature threshold distance and determine them as bolt loosening risk positions.

[0140] Among them, by screening out the bolt positions whose signal feature distance is greater than the signal feature threshold distance and determining them as risk positions for bolt loosening, potential accidents can be predicted in advance, more serious accidents can be avoided, and the safe and stable operation of the power grid can be promoted.

[0141] It should be noted that the embodiment of the present application decomposes the vibration signal by using a variational modal decomposition algorithm based on spider wasp optimization to obtain several modal components, thereby improving the global optimization performance, and reconstructs the signal by screening out multiple modal components with a high correlation with the vibration signal to obtain a reconstructed signal, determines the signal characteristic threshold distance according to the reconstructed signal and a preset normal vibration signal, and compares the signal characteristic distance of each bolt position with the signal characteristic threshold distance, and screens out the bolt positions with signal characteristic distances greater than the signal characteristic threshold distance, and determines them as risk positions for bolt loosening, thereby improving the efficiency of diagnosing loose bolts on pole towers, making early judgments for on-site detection of local bolt loosening on towers, and having good applicability.

[0142] Based on the same inventive concept, an embodiment of the present application further provides a tower bolt loosening diagnosis system for implementing the tower bolt loosening diagnosis method mentioned above.

[0143] The solution to the problem provided by the system is similar to the solution described in the above method, so the specific limitations in one or more embodiments of the tower bolt loosening diagnosis system provided below can refer to the limitations on the tower bolt loosening diagnosis method above, and will not be repeated here.

[0144] like Figure 2 As shown, the embodiment of the present application also provides a tower bolt loosening diagnosis system, comprising:

[0145] The signal acquisition module 100 is used to collect the vibration signal of the target transmission tower;

[0146] A signal decomposition module 200, for decomposing the vibration signal based on a variational modal decomposition algorithm optimized by spider wasps to obtain a plurality of modal components;

[0147] The component screening module 300 is used to screen the correlation between a plurality of modal components and the vibration signal, and screen out a plurality of modal components with a higher correlation;

[0148] The signal reconstruction module 400 is used to reconstruct the signals of the multiple modal components to obtain the reconstructed signals;

[0149] A threshold distance determination module 500, configured to determine a signal feature threshold distance according to the reconstructed signal and a preset normal vibration signal;

[0150] A characteristic distance determination module 600 is used to determine the signal characteristic distances of multiple bolt positions according to the vibration signals of multiple bolt positions of the target transmission tower;

[0151] The bolt loosening identification module 700 is used to compare the signal feature distance of each bolt position with the signal feature threshold distance, and screen out the bolt positions whose signal feature distance is greater than the signal feature threshold distance to determine them as bolt loosening risk positions.

[0152] In some embodiments, the signal acquisition module 100 is used to perform multiple pulse excitations on multiple preset excitation positions on the target transmission tower, and obtain vibration acceleration signals at each bolt of the target transmission tower; establish a frequency response function based on the vibration acceleration signal and the pulse excitation signal; perform mean calculation on multiple frequency response functions of each preset excitation position to determine an average frequency response function; and determine the natural frequency as the vibration signal based on the average frequency response function.

[0153] In some embodiments, the vibration signal is decomposed based on the variational modal decomposition algorithm optimized by spider wasps to obtain several modal components, including:

[0154] The spider wasp optimization algorithm is used to optimize the network parameters of the variational mode decomposition algorithm, and the optimization results are used to optimize the network parameters of the variational mode decomposition algorithm;

[0155] The vibration signal is decomposed using the optimized variational mode decomposition algorithm to obtain several modal components.

[0156] In some embodiments, the network parameters of the variational mode decomposition algorithm are optimized using the spider wasp optimization algorithm, including:

[0157] Initialize the population parameters of the spider-wasp optimization algorithm, the position parameters of the spider swarm, the maximum number of iterations, and the initial fitness value, where the position of each swarm represents a network parameter of the variational mode decomposition algorithm;

[0158] Solve the search space using a population-guiding mechanism;

[0159] The position parameters of the swarm are updated using the sine-cosine transform inertia weights to simulate the swarm's exploration behavior in the search space;

[0160] The fitness value is determined by the fitness function according to the location parameters of the swarm;

[0161] Determine whether the fitness value has reached the optimum or whether the iteration has reached the maximum number of iterations. If it is determined that the fitness value has not reached the optimum or the iteration has not reached the maximum number of iterations, proceed to the step of updating the position parameters of the swarm using the sine-cosine transform inertia weights to simulate the swarm's exploration behavior in the search space;

[0162] If the fitness value is judged to be optimal or the iteration reaches the maximum number of iterations, the optimization process is stopped and the current position parameters of the swarm are output as the optimal network parameters of the variational mode decomposition algorithm.

[0163] In some embodiments, the component screening module 300 is used to determine the correlation coefficients between a plurality of modal components and the vibration signal respectively using the Pearson correlation coefficient analysis method; and screen out a plurality of modal components whose correlation coefficients are greater than a preset correlation coefficient threshold.

[0164] In some embodiments, the signal reconstruction module 400 is used to perform three-layer wavelet packet decomposition on each modal component using wavelet basis functions; using energy logarithmic entropy as a cost function, using the cost function to determine the cost function size of each node coefficient obtained by wavelet packet decomposition, and determining the minimum wavelet packet basis; performing threshold selection based on an adaptive threshold criterion, using the soft threshold function to perform wavelet packet reconstruction, and obtaining a reconstructed signal.

[0165] In some embodiments, determining a signal characteristic threshold distance according to the reconstructed signal and a preset normal vibration signal includes:

[0166] Extract eigenvalues ​​from the reconstructed signal to form a eigenvector;

[0167] Determine the eigenvector standard value and the eigenvector variance value according to the eigenvector;

[0168] The Mahalanobis distance between the reconstructed signal and the preset normal vibration signal is determined as the signal feature threshold distance according to the feature vector standard value, the feature vector variance value and the feature vector standard value of the preset normal vibration signal.

[0169] like Figure 3 As shown, an embodiment of the present application further provides an electronic device, the electronic device 10 includes a memory 20 and a processor 30, the memory 20 stores a computer program, and when the computer program is executed by the processor 30, the processor 30 executes the steps of the tower bolt loosening diagnosis method as described in the above embodiment.

[0170] The embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed, the steps of the method for diagnosing loose bolts of a tower as described in the above embodiment are implemented.

[0171] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, electronic device and computer storage medium can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0172] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or apparatus.

[0173] In several embodiments provided by the present invention, it is understood that each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and a part of a module, a program segment or a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.

[0174] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0175] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0176] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or 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 software functional units.

[0177] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for executing all or part of the steps of the method described in each embodiment of the present invention through a computer device (which can be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store program codes.

[0178] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for diagnosing loose bolts of a tower, characterized in that: include: Collect vibration signals of target transmission towers; Decomposing the vibration signal based on a variational modal decomposition algorithm optimized by spider wasps to obtain a number of modal components; Screening the correlations between a plurality of the modal components and the vibration signal, and screening out a plurality of the modal components with a higher correlation; Performing signal reconstruction on the plurality of modal components to obtain a reconstructed signal; Determining a signal characteristic threshold distance according to the reconstructed signal and a preset normal vibration signal; Determine signal characteristic distances of multiple bolt positions according to vibration signals of multiple bolt positions of the target transmission tower; The signal feature distance of each bolt position is compared with the signal feature threshold distance, and the bolt positions whose signal feature distance is greater than the signal feature threshold distance are screened out and determined as bolt loosening risk positions.

2. The method for diagnosing loose tower bolts according to claim 1, characterized in that: The collecting of the vibration signal of the target transmission tower includes: Performing multiple pulse excitations on multiple preset excitation positions on the target transmission tower, and acquiring vibration acceleration signals at each bolt of the target transmission tower; Establishing a frequency response function according to the vibration acceleration signal and the pulse excitation signal; Calculating the mean of the multiple frequency response functions at each of the preset excitation positions to determine an average frequency response function; A natural frequency is determined as a vibration signal based on the average frequency response function.

3. The method for diagnosing loose tower bolts according to claim 1, characterized in that: The variational modal decomposition algorithm based on spider wasp optimization decomposes the vibration signal to obtain several modal components, including: Optimizing the network parameters of the variational mode decomposition algorithm using a spider wasp optimization algorithm, and optimizing the network parameters of the variational mode decomposition algorithm using the optimization result; The vibration signal is decomposed using an optimized variational modal decomposition algorithm to obtain a number of modal components.

4. The method for diagnosing loose tower bolts according to claim 3, characterized in that: The method of optimizing the network parameters of the variational mode decomposition algorithm by using the spider wasp optimization algorithm includes: Initializing population parameters of the spider-wasp optimization algorithm, position parameters of the spider swarm, maximum number of iterations, and initial fitness values, wherein the position of each swarm represents a network parameter of the variational mode decomposition algorithm; Solve the search space using a population-guiding mechanism; updating the position parameters of the swarm using the sine-cosine transformed inertia weights to simulate the swarm's exploration behavior in the search space; The fitness value is determined by the fitness function according to the location parameters of the swarm; Determine whether the fitness value has reached the optimum or whether the iteration has reached the maximum number of iterations. If it is determined that the fitness value has not reached the optimum or the iteration has not reached the maximum number of iterations, proceed to the step of updating the position parameters of the bee colony using the sine-cosine transform inertia weight to simulate the exploration behavior of the bee colony in the search space; If it is determined that the fitness value reaches the optimum or the iteration reaches the maximum number of iterations, the optimization process is stopped and the current position parameters of the swarm are output as the optimal network parameters of the variational mode decomposition algorithm.

5. The method for diagnosing loose tower bolts according to claim 1, characterized in that: The step of screening the correlations between the plurality of modal components and the vibration signal to screen out a plurality of modal components with higher correlations comprises: Determine the correlation coefficients between the modal components and the vibration signal respectively using the Pearson correlation coefficient analysis method; A plurality of the modal components whose correlation coefficients are greater than a preset correlation coefficient threshold are screened out.

6. The method for diagnosing loose bolts of a tower according to claim 1, characterized in that: The step of reconstructing the signals of the plurality of modal components to obtain the reconstructed signals comprises: Performing three-layer wavelet packet decomposition on each of the modal components using wavelet basis functions; Taking energy logarithmic entropy as a cost function, using the cost function to determine the cost function size of each node coefficient obtained by wavelet packet decomposition, and determining the minimum wavelet packet basis; A soft threshold function is selected based on an adaptive threshold criterion, and wavelet packet reconstruction is performed using the soft threshold function to obtain a reconstructed signal.

7. The method for diagnosing loose tower bolts according to claim 1, characterized in that: The determining of the signal characteristic threshold distance according to the reconstructed signal and the preset normal vibration signal includes: Extracting characteristic values ​​from the reconstructed signal to form a characteristic vector; Determine a characteristic vector standard value and a characteristic vector variance value according to the characteristic vector; The Mahalanobis distance between the reconstructed signal and the preset normal vibration signal is determined as the signal feature threshold distance according to the feature vector standard value, the feature vector variance value and the feature vector standard value of the preset normal vibration signal.

8. A tower bolt loosening diagnosis system, characterized in that: include: A signal acquisition module, used to collect vibration signals of target transmission towers; A signal decomposition module, used for decomposing the vibration signal based on a variational modal decomposition algorithm optimized by spider wasps to obtain a plurality of modal components; A component screening module, used for screening the correlation between a plurality of the modal components and the vibration signal, and screening out a plurality of the modal components with a higher correlation; A signal reconstruction module, used for performing signal reconstruction on the plurality of modal components to obtain a reconstructed signal; A threshold distance determination module, used to determine a signal feature threshold distance according to the reconstructed signal and a preset normal vibration signal; A characteristic distance determination module, used to determine the signal characteristic distances of multiple bolt positions according to the vibration signals of multiple bolt positions of the target transmission tower; The bolt loosening identification module is used to compare the signal feature distance of each bolt position with the signal feature threshold distance, and screen out the bolt positions whose signal feature distance is greater than the signal feature threshold distance to determine them as bolt loosening risk positions.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the method for diagnosing loosening of a tower bolt according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the steps of the method for diagnosing loosening of tower bolts according to any one of claims 1 to 7 are implemented.

Images