Transmission tower bolt loosening positioning method, system, equipment, medium and product

Abstract

The invention relates to the technical field of power systems, and discloses a power transmission tower bolt loosening positioning method, system, device, medium and product, the method comprises the steps of decomposing vibration signals through an IVMD algorithm optimized based on a crested porcupine optimization algorithm, extracting feature signals of a plurality of vibration signals, distributing labels to the plurality of feature signals, and determining whether the vibration signals are loose or not. The method comprises the following steps: acquiring a plurality of feature signals to determine the bolt position of each feature signal and whether the bolt is loosened, constructing a training data set through a plurality of feature signals and labels, and performing classified training on the training data set by using a convolutional neural network to obtain a bolt loosening positioning model, thereby predicting whether the bolt is loosened or not for an unidentified bolt. And the loosened bolt is positioned, so that the bolt loosening positioning efficiency and accuracy are improved.

Description

Technical Field

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

[0002] Overhead transmission lines often use transmission towers to suspend transmission wires and overhead ground wires at high altitudes to achieve power transmission goals. Bolt connection is a common connection method for transmission towers, and the stability of the bolt connection of the tower plays a vital role in the normal operation of the power system. During the operation of the transmission tower, due to the loosening of the bolts of the tower, the wires will dance under the action of long-term strong winds, causing the tower to tilt or even collapse, posing a serious threat to the safe operation of the power system.

[0003] Therefore, it is essential to conduct safety inspection on the bolts of transmission towers. However, the traditional manual scheduled maintenance method is not only inefficient, but also difficult to locate the fault position in time. Summary of the invention

[0004] In view of this, the present invention provides a method, system, device, medium and product for locating loose bolts on a transmission tower, which solves the technical problem that the traditional manual scheduled maintenance method is not only inefficient but also difficult to locate the fault position in time.

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

[0006] Collect vibration signals at multiple bolts of the target transmission tower;

[0007] Decomposing the vibration signal based on the IVMD algorithm optimized by the Crested Porcupine Optimization Algorithm to obtain a plurality of modal components, and extracting a plurality of characteristic signals of the vibration signal using the plurality of modal components;

[0008] assigning labels to the plurality of characteristic signals, the labels including bolt positions and whether the bolts are loose;

[0009] Constructing a training data set according to the plurality of characteristic signals and the labels;

[0010] Using a convolutional neural network to perform classification training on the training data set to obtain a bolt loosening positioning model;

[0011] The bolt loosening positioning model is used to predict whether the unidentified bolts are loose, and to locate the loosened bolts.

[0012] Preferably, the method further comprises: optimizing the network parameters of the IVMD algorithm using a crested porcupine optimization algorithm; the optimizing the network parameters of the IVMD algorithm using a crested porcupine optimization algorithm comprises:

[0013] Initializing the parameters of the crested porcupine optimization algorithm, the parameters including porcupine population, population size, maximum number of iterations and fitness function, wherein the porcupine individuals in the porcupine population represent a candidate network parameter solution of the IVMD algorithm;

[0014] Constructing a mathematical model to simulate the defense strategy of the porcupine individual, wherein the defense strategy includes visual defense, sound defense, odor defense and physical attack;

[0015] Updating the current position of the porcupine individual using the defense strategy;

[0016] The fitness function is used to calculate the fitness value of each porcupine individual after the position is updated, and the global optimal solution of the porcupine population is determined;

[0017] Determine whether the current number of iterations reaches the maximum number of iterations, and when it is determined that the current number of iterations does not reach the maximum number of iterations, proceed to the step of updating the current position of the porcupine individual using the defense strategy;

[0018] During the iteration process, when the porcupine population reaches a preset reduction period, a population reduction operation is performed using a cyclic population reduction rule;

[0019] When it is determined that the current number of iterations reaches the maximum number of iterations, the iteration is terminated, and the porcupine individual with the best current fitness value is output as the optimal network parameter of the IVMD algorithm.

[0020] Preferably, the IVMD algorithm optimized based on the crested porcupine optimization algorithm decomposes the vibration signal to obtain multiple modal components, and uses the multiple modal components to extract multiple characteristic signals of the vibration signal, including:

[0021] Decomposing the vibration signal using an IVMD algorithm optimized based on a Crested Porcupine Optimization Algorithm to obtain a plurality of modal components;

[0022] For each of the modal components, determining an envelope of the modal component;

[0023] Determine the envelope kurtosis according to the envelope of the modal component, traverse the envelope kurtosis of all modal components, and determine the local maximum envelope kurtosis and the number of decomposition layers corresponding to the local maximum envelope kurtosis;

[0024] Decomposing each of the vibration signals using the number of decomposition layers to obtain new modal components;

[0025] The new modal component is used to perform frequency domain conversion to obtain a frequency spectrum, and the largest frequency component in the frequency spectrum is identified as a characteristic signal of the vibration signal.

[0026] Preferably, the convolutional neural network is a one-dimensional convolutional neural network.

[0027] Preferably, the convolutional neural network includes an input layer, a convolution layer, a pooling layer, a fully connected layer and a SoftMax classifier;

[0028] The method of using a convolutional neural network to classify and train the training data set to obtain a bolt loosening positioning model includes:

[0029] Inputting the characteristic signal into the convolution layer through the input layer for convolution processing to obtain characteristic data;

[0030] Inputting the feature data into the pooling layer for pooling processing;

[0031] Inputting the pooled feature data into the fully connected layer for feature combination, and mapping the combined features to the sample label space;

[0032] The features in the sample label space are classified by the SoftMax classifier to obtain label classification results of the features in the sample label space, and the bolt loosening positioning model is constructed.

[0033] In a second aspect, the present invention further provides a transmission tower bolt loosening positioning system, comprising:

[0034] A signal acquisition module, used to collect vibration signals at multiple bolts of a target transmission tower;

[0035] A signal decomposition module, used for decomposing the vibration signal based on the IVMD algorithm optimized by the Crested Porcupine Optimization Algorithm to obtain a plurality of modal components, and extracting a plurality of characteristic signals of the vibration signal using the plurality of modal components;

[0036] A label assignment module, used to assign labels to the plurality of characteristic signals, wherein the labels include the position of the bolts and whether the bolts are loose;

[0037] A data set construction module, used to construct a training data set according to the plurality of characteristic signals and the labels;

[0038] A classification training module, used to perform classification training on the training data set using a convolutional neural network to obtain a bolt loosening positioning model;

[0039] The positioning prediction module is used to use the bolt loosening positioning model to predict whether the unidentified bolts are loose, and to locate the loosened bolts.

[0040] 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 method for locating loose bolts of a transmission tower as described in the first aspect.

[0041] 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 locating loose bolts of a transmission tower as described in the first aspect are implemented.

[0042] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program comprises program instructions, wherein when the program instructions are executed by a computer, the computer executes the steps of the method for locating loose bolts of a transmission tower as described in the first aspect.

[0043] It can be seen from the above technical scheme that the present invention decomposes the vibration signal through the IVMD algorithm optimized by the Crested Porcupine Optimization Algorithm, extracts characteristic signals of multiple vibration signals, assigns labels to the multiple characteristic signals to determine the bolt position of each characteristic signal and whether the bolt is loose, constructs a training data set through multiple characteristic signals and labels, and uses a convolutional neural network to perform classification training on the training data set to obtain a bolt loosening positioning model, thereby predicting whether the unidentified bolt is loose, and locating the loose bolt, thereby improving the efficiency and accuracy of bolt loosening positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] 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.

[0045] Figure 1 A flow chart of a method for locating loose bolts on a transmission tower provided by an embodiment of the present invention;

[0046] Figure 2 A schematic structural diagram of a transmission tower bolt loosening positioning system provided by an embodiment of the present invention;

[0047] Figure 3A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] 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.

[0049] An embodiment of the present application provides a method for locating loose bolts of a transmission tower. The embodiment can be applicable to the situation of locating loose bolts of a transmission tower. The method can be performed by a transmission tower bolt loose positioning device. The transmission tower bolt loose positioning device can be implemented in the form of hardware and / or software. The transmission tower bolt loose positioning device can be configured in a computer device.

[0050] like Figure 1 As shown, the embodiment of the present application provides a method for locating loose bolts of a transmission tower, which is characterized by comprising:

[0051] Step S1: collecting vibration signals at multiple bolts of a target transmission tower.

[0052] The vibration signals at the multiple bolts of the target transmission tower can be obtained by using an acceleration sensor.

[0053] Step S2: Decomposing the vibration signal based on the IVMD algorithm optimized by the Crested Porcupine Optimization Algorithm to obtain multiple modal components, and extracting characteristic signals of the multiple vibration signals using the multiple modal components.

[0054] Among them, the Porcupine Quill Optimization (PQO) algorithm is a heuristic optimization algorithm inspired by the behavior of crested porcupines (also known as African porcupines) when they encounter threats. When threatened, crested porcupines will erect their quills to protect themselves, and is designed to solve complex optimization problems.

[0055] The embodiment of the present application also uses the Crested Porcupine Optimization Algorithm to optimize the network parameters of the IVMD algorithm.

[0056] Specifically, the crested porcupine optimization algorithm is used to optimize the network parameters of the IVMD algorithm, including:

[0057] Step S21, initializing the parameters of the crested porcupine optimization algorithm, the parameters including the porcupine population, population size, maximum number of iterations and fitness function, the porcupine individuals in the porcupine population represent a candidate network parameter solution of the IVMD algorithm.

[0058] The fitness function may be a root mean square error function or the like.

[0059] Step S22: construct a mathematical model to simulate the defense strategy of the porcupine individual, the defense strategy includes visual defense, sound defense, odor defense and physical attack.

[0060] Among them, visual defense is when the porcupine senses the threat of predation, it will show the first defensive posture, raise its quills to increase the visual volume, induce the predator to have an illusion, and the predator can only choose to get closer or farther away from the porcupine. Normal distribution random variables are used to quantify and simulate these two decision-making behaviors. The mathematical model under this behavior mode is precisely expressed as follows:

[0061]

[0062] In the formula, It is a random number based on normal distribution; is a random value in the interval [0,1]. is the evaluation function The optimal solution of is a vector generated between the current porcupine and a randomly selected porcupine from the population, representing the position of the predator at iteration t, where r is a random number between [1, N], and the subscript Represents the current porcupine.

[0063] In sound defense, porcupines use acoustic defense to produce a deterrent effect by roaring to deal with predators. As the distance between the predator and the porcupine decreases, the intensity of the porcupine's vocalization increases. The following formula was constructed to simulate this phenomenon:

[0064]

[0065] In the formula, is a random value generated between 0 and 1, and r1 and r2 are two random integers between [1,N].

[0066] In odor defense, porcupines secrete and diffuse pheromones to deter predators. When faced with an already present threat, the intensity of their "bad smell" signal increases as the distance from the predator decreases. To quantify this process, the following mathematical model formula is proposed:

[0067]

[0068] In the formula, Iteration Time The location of an individual, is the odor diffusion factor defined using equation (1.5). is a random number between [1,N], is a random value between the interval [0,1], is a parameter used to control the search direction, and, is the defense factor. As shown below:

[0069]

[0070]

[0071]

[0072] In the formula, Represents iteration Time The objective function value of each individual, is the overall size, is a small value that avoids division by zero, is a vector of randomly generated values ​​between 0 and 1, is a variable that contains a randomly generated number between 0 and 1, is the number of current iterations, is the maximum number of iterations.

[0073] The vector is used to simulate three possible situations in this strategy: when the simulation parameter value is 0, the predator stops approaching and the two sides maintain a constant distance. If the parameter value is 1, it means that the predator approaches the porcupine and the odor release is enhanced. For the parameter combination between 0 and 1, it means that the predator and the porcupine remain within an appropriate safe distance, so the porcupine does not need to emit excessive odor to maintain this relatively stable state.

[0074] Physical attack is when the predator is very close, the porcupine will resort to physical attack, using the short and thick porcupine quills to attack. In physical attack, the two creatures collide violently, representing a one-dimensional inelastic collision. The mathematical formula to express its physical attack behavior is as follows:

[0075]

[0076] In the formula, is the optimal solution obtained, Indicates that the porcupine is iterating Time The location of an individual, is the convergence speed factor, is a random value in the interval [0,1].

[0077] Step S23: Update the current position of the porcupine individual using the defense strategy.

[0078] Among them, each individual updates the current position of the porcupine individual according to the defense strategy from other individuals, and the updated position is:

[0079]

[0080] In the formula, α is the step size coefficient, and rand() is a random number generator, which is used to introduce a certain degree of randomness to avoid falling into the local optimum.

[0081] Step S24: Calculate the fitness value of each porcupine individual after the position is updated using the fitness function to determine the global optimal solution of the porcupine population.

[0082] Step S25, determining whether the current number of iterations has reached the maximum number of iterations. When it is determined that the current number of iterations has not reached the maximum number of iterations, the process proceeds to the step of updating the current position of the porcupine individual using the defense strategy.

[0083] Step S26: During the iteration process, when the porcupine population reaches a preset reduction period, the cyclic population reduction rule is used to perform a population reduction operation.

[0084] Among them, the cyclic population reduction rule uses a selection mechanism (such as fitness-based selection) to determine which individuals are eliminated and which are retained.

[0085] The cyclic population reduction rule (CPR) is a technique that balances convergence efficiency and population diversity maintenance. It selects some CPs that are not threatened, speeds up the convergence of the global search by dynamically shrinking the capacity, and reintegrates these solutions into the population in the subsequent stage, thereby promoting the diversification of the population structure and effectively preventing it from falling into the local optimal solution. This process is regulated by the law of a periodic variable T. The corresponding mathematical model describes this method of cyclically reducing the population size as follows:

[0086]

[0087] In the formula, is the minimum number of individuals in the newly generated population, so the population size cannot be less than ; is the number of individuals before reducing the population size; N is the number of individuals after reducing the population size; t is the current evaluation of the function; % represents the remainder or modulus operator; is the maximum number of function evaluations; T is a variable that determines the number of loops.

[0088] Step S27: When it is determined that the current number of iterations reaches the maximum number of iterations, the iteration is terminated, and the porcupine individual with the best current fitness value is output as the optimal network parameter of the IVMD algorithm.

[0089] Among them, the IVMD algorithm optimized based on the crested porcupine optimization algorithm in step S2 decomposes the vibration signal to obtain multiple modal components, and uses the multiple modal components to extract characteristic signals of multiple vibration signals, including:

[0090] Step S201 : Decomposing the vibration signal based on the IVMD algorithm optimized by the Crested Porcupine Optimization Algorithm to obtain multiple modal components.

[0091] Step S202: for each modal component, determine the envelope of the modal component.

[0092] Among them, the decomposition level K of IVMD is set, and the envelope of each modal component is calculated:

[0093]

[0094] in is the absolute value obtained after Hilbert transform, is the first modal components, is the mode number, Represents a certain vibration signal parameter extracted.

[0095] Step S203: determine the envelope kurtosis according to the envelope of the modal component, traverse the envelope kurtosis of all modal components, and determine the local maximum envelope kurtosis and the number of decomposition layers corresponding to the local maximum envelope kurtosis.

[0096] Among them, calculate the envelope kurtosis of each modal component, find the local maximum envelope kurtosis, The envelope kurtosis calculation formula of a modal component is:

[0097]

[0098] In the formula, yes The average value of for The standard deviation of for The fourth-order central moment of .

[0099] Among them, the local maximum envelope kurtosis is:

[0100]

[0101] In the formula, is the local maximum envelope kurtosis, is the local kurtosis of each envelope.

[0102] The maximum number of IVMD decomposition layers is obtained by calculating the global maximum of the maximum envelope kurtosis and drawing a changing trend graph of the maximum envelope kurtosis.

[0103] The global maximum of the maximum envelope kurtosis is calculated by , the K value corresponding to the global maximum value is the maximum IVMD optimal decomposition layer number.

[0104]

[0105] Step S204: Decompose each vibration signal using the decomposition layer number to obtain new modal components.

[0106] Step S205: Use the new modal component to perform frequency domain conversion to obtain a spectrum diagram, and identify the largest frequency component in the spectrum diagram as a characteristic signal of the vibration signal.

[0107] Among them, the optimized transmission tower vibration signal is decomposed into K layers, and its significant characteristic signals are extracted using spectrum analysis.

[0108] Step S3: assigning labels to the multiple feature signals, where the labels include the bolt positions and whether the bolts are loose.

[0109] Step S4: construct a training data set based on multiple feature signals and labels.

[0110] Among them, each characteristic signal corresponds to a label, and the characteristic signal and the label are in a mapping relationship.

[0111] Step S5: Use a convolutional neural network to perform classification training on the training data set to obtain a bolt loosening positioning model.

[0112] Among them, the convolutional neural network is a one-dimensional convolutional neural network. The convolutional neural network includes an input layer, a convolution layer, a pooling layer, a fully connected layer and a SoftMax classifier.

[0113] Specifically, in step S5, the training data set is classified and trained using a convolutional neural network to obtain a bolt loosening positioning model, including:

[0114] Step S501: Input the feature signal into the convolution layer through the input layer for convolution processing to obtain feature data.

[0115] Among them, the convolution layer convolves the input feature signal, and the local features can be extracted through the activation function in the convolution layer. The convolution process is:

[0116]

[0117] In the formula, Layer The output of the characteristic signal is Indicates; in In the layer, The output of the characteristic signal is Indicates that the offset value is Indicates; at the same time, Layer The characteristic signal and Layer The convolution kernel between neurons is used Indicates. Among them, The value range is , Represents the number of neurons.

[0118] Step S502: input the feature data into the pooling layer for pooling processing.

[0119] The signal features extracted by the convolution layer are introduced into the pooling layer to retain useful features and reduce unnecessary features, so that the extracted features can better reflect the looseness of bolts at different positions. Average pooling and maximum pooling are two common pooling operations. In this article, the maximum pooling operation is used, and the formula is:

[0120]

[0121] In the formula, The first After the feature signal pooling Value Said that Tier The first Activation value Indicates that the pooling area width is express.

[0122] Step S503: input the feature data after pooling processing into the fully connected layer for feature combination, and map the combined features to the sample label space;

[0123] Step S504: Classify the features in the sample label space through the SoftMax classifier to obtain the label classification results of the features in the sample label space, and build a bolt loosening positioning model.

[0124] Among them, the convolutional neural network is used to extract the image depth features and convert the input image into the feature , classify the features through the SoftMax classifier. Process the feature signal with the SoftMax classifier to obtain the probability value of the feature vector belonging to each category:

[0125]

[0126] In the formula, Indicates that the sample data is processed by the SoftMax classifier and the output pixel the probability vector of belonging to each class; Indicates that the sample data belongs to Probability value of the class; Indicates that the l-layer network corresponds to The classification vector of classes.

[0127] The classification error of the SoftMax classifier can be expressed as :

[0128]

[0129] pass The classification error can ensure the maximum classification accuracy in the sample feature learning process, thereby making the signal features more obvious and locating the loose bolts more accurately.

[0130] Step S6: Use the bolt loosening positioning model to predict whether the unidentified bolts are loose, and locate the loose bolts.

[0131] The characteristic signal of the unidentified bolt is input into the bolt loosening positioning model, and the loosening status and positioning information of the unidentified bolt are output.

[0132] It should be noted that the embodiment of the present application decomposes the vibration signal through the IVMD algorithm optimized based on the Crested Porcupine Optimization Algorithm, extracts characteristic signals of multiple vibration signals, assigns labels to the multiple characteristic signals to determine the bolt position of each characteristic signal and whether the bolt is loose, constructs a training data set through multiple characteristic signals and labels, and uses a convolutional neural network to perform classification training on the training data set to obtain a bolt loosening location model, thereby predicting whether the unidentified bolt is loose and locating the loose bolt, thereby improving the efficiency and accuracy of bolt loosening location.

[0133] Based on the same inventive concept, an embodiment of the present application further provides a transmission tower bolt loosening positioning system for implementing the above-mentioned transmission tower bolt loosening positioning method.

[0134] The solution to the problem provided by the system is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the transmission tower bolt loosening positioning system provided below can refer to the limitations of the transmission tower bolt loosening positioning method above, and will not be repeated here.

[0135] like Figure 2 As shown, the embodiment of the present application provides a transmission tower bolt loosening positioning system, comprising:

[0136] The signal acquisition module 100 is used to acquire vibration signals at multiple bolts of a target transmission tower.

[0137] The signal decomposition module 200 is used to decompose the vibration signal based on the IVMD algorithm optimized by the Crested Porcupine Optimization Algorithm to obtain multiple modal components, and extract characteristic signals of the multiple vibration signals using the multiple modal components.

[0138] The label assignment module 300 is used to assign labels to multiple characteristic signals, and the labels include the bolt position and whether the bolt is loose.

[0139] A data set construction module 400 is used to construct a training data set according to a plurality of characteristic signals and labels;

[0140] A classification training module 500 is used to perform classification training on a training data set using a convolutional neural network to obtain a bolt loosening positioning model;

[0141] The positioning prediction module 600 is used to predict whether the unidentified bolts are loose by using the bolt loosening positioning model, and to locate the loosened bolts.

[0142] In some embodiments, the system further comprises: a parameter optimization module for optimizing the network parameters of the IVMD algorithm using the crested porcupine optimization algorithm; optimizing the network parameters of the IVMD algorithm using the crested porcupine optimization algorithm comprises:

[0143] Initialize the parameters of the crested porcupine optimization algorithm, including the porcupine population, population size, maximum number of iterations, and fitness function. The porcupine individuals in the porcupine population represent a candidate network parameter solution for the IVMD algorithm.

[0144] A mathematical model was constructed to simulate the defense strategies of porcupine individuals, including visual defense, sound defense, odor defense, and physical attack.

[0145] Update the current position of the individual porcupine using the defense strategy;

[0146] The fitness function is used to calculate the fitness value of each porcupine individual after the updated position, and the global optimal solution of the porcupine population is determined;

[0147] Determine whether the current number of iterations reaches the maximum number of iterations, and when it is determined that the current number of iterations does not reach the maximum number of iterations, proceed to the step of updating the current position of the porcupine individual using the defense strategy;

[0148] In the iterative process, when the porcupine population reaches the preset reduction period, the cyclic population reduction rule is used to perform the population reduction operation;

[0149] When it is determined that the current number of iterations reaches the maximum number of iterations, the iteration ends, and the porcupine individual with the best current fitness value is output as the optimal network parameter of the IVMD algorithm.

[0150] In some embodiments, the signal decomposition module 200 is used to decompose the vibration signal based on the IVMD algorithm optimized by the crested porcupine optimization algorithm to obtain multiple modal components;

[0151] For each modal component, determine the envelope of the modal component;

[0152] Determine the envelope kurtosis according to the envelope curve of the modal component, traverse the envelope kurtosis of all modal components, and determine the local maximum envelope kurtosis and the number of decomposition layers corresponding to the local maximum envelope kurtosis;

[0153] Decompose each vibration signal using the decomposition layer number to obtain new modal components;

[0154] The new modal components are used to perform frequency domain conversion to obtain a spectrum diagram, and the largest frequency component in the spectrum diagram is identified as the characteristic signal of the vibration signal.

[0155] In some embodiments, the convolutional neural network is a one-dimensional convolutional neural network.

[0156] In some embodiments, the convolutional neural network includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and a SoftMax classifier.

[0157] The classification training module 500 is used to input the feature signal into the convolution layer through the input layer for convolution processing to obtain feature data;

[0158] Input the feature data into the pooling layer for pooling processing;

[0159] The feature data after pooling is input into the fully connected layer for feature combination, and the combined features are mapped to the sample label space;

[0160] The features in the sample label space are classified by the SoftMax classifier to obtain the label classification results of the features in the sample label space and build a bolt loosening positioning model.

[0161] like Figure 3As 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 method for locating loose bolts of a transmission tower as in the above embodiment.

[0162] 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 locating loose bolts of a transmission tower as in the above embodiment are implemented.

[0163] An embodiment of the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the steps of the method for locating loose bolts on a transmission tower as in the above-mentioned embodiment.

[0164] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, electronic devices, computer storage media and computer program products can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0165] 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.

[0166] 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.

[0167] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media, computer program products and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, 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 through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0168] 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.

[0169] 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.

[0170] 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.

[0171] 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 locating loose bolts on a transmission tower, characterized in that: include: Collect vibration signals at multiple bolts of the target transmission tower; Decomposing the vibration signal based on the IVMD algorithm optimized by the crested porcupine optimization algorithm to obtain a plurality of modal components, and extracting a plurality of characteristic signals of the vibration signal using the plurality of modal components; assigning labels to the plurality of characteristic signals, the labels including bolt positions and whether the bolts are loose; Constructing a training data set according to the plurality of characteristic signals and the labels; Using a convolutional neural network to perform classification training on the training data set to obtain a bolt loosening positioning model; The bolt loosening positioning model is used to predict whether the unidentified bolts are loose, and to locate the loosened bolts.

2. The method for locating loose bolts of a transmission tower according to claim 1, characterized in that: The method further includes: optimizing the network parameters of the IVMD algorithm using a crested porcupine optimization algorithm; optimizing the network parameters of the IVMD algorithm using a crested porcupine optimization algorithm includes: Initializing the parameters of the crested porcupine optimization algorithm, the parameters including porcupine population, population size, maximum number of iterations and fitness function, wherein the porcupine individuals in the porcupine population represent a candidate network parameter solution of the IVMD algorithm; Constructing a mathematical model to simulate the defense strategy of the porcupine individual, wherein the defense strategy includes visual defense, sound defense, odor defense and physical attack; Updating the current position of the porcupine individual using the defense strategy; The fitness function is used to calculate the fitness value of each porcupine individual after the position is updated, and the global optimal solution of the porcupine population is determined; Determine whether the current number of iterations reaches the maximum number of iterations, and when it is determined that the current number of iterations does not reach the maximum number of iterations, proceed to the step of updating the current position of the porcupine individual using the defense strategy; During the iteration process, when the porcupine population reaches a preset reduction period, a population reduction operation is performed using a cyclic population reduction rule; When it is determined that the current number of iterations reaches the maximum number of iterations, the iteration is terminated, and the porcupine individual with the best current fitness value is output as the optimal network parameter of the IVMD algorithm.

3. The method for locating loose bolts of a transmission tower according to claim 1, characterized in that: The IVMD algorithm optimized based on the crested porcupine optimization algorithm decomposes the vibration signal to obtain multiple modal components, and uses the multiple modal components to extract multiple characteristic signals of the vibration signal, including: Decomposing the vibration signal using an IVMD algorithm optimized based on a Crested Porcupine Optimization Algorithm to obtain a plurality of modal components; For each of the modal components, determining an envelope of the modal component; Determine the envelope kurtosis according to the envelope of the modal component, traverse the envelope kurtosis of all modal components, and determine the local maximum envelope kurtosis and the number of decomposition layers corresponding to the local maximum envelope kurtosis; Decomposing each of the vibration signals using the number of decomposition layers to obtain new modal components; The new modal component is used to perform frequency domain conversion to obtain a frequency spectrum, and the largest frequency component in the frequency spectrum is identified as a characteristic signal of the vibration signal.

4. The method for locating loose bolts of a transmission tower according to claim 1, characterized in that: The convolutional neural network is a one-dimensional convolutional neural network.

5. The method for locating loose bolts of a transmission tower according to claim 4, characterized in that: The convolutional neural network includes an input layer, a convolution layer, a pooling layer, a fully connected layer and a SoftMax classifier; The method of using a convolutional neural network to classify and train the training data set to obtain a bolt loosening positioning model includes: Inputting the characteristic signal into the convolution layer through the input layer for convolution processing to obtain characteristic data; Inputting the feature data into the pooling layer for pooling processing; Inputting the pooled feature data into the fully connected layer for feature combination, and mapping the combined features to the sample label space; The features in the sample label space are classified by the SoftMax classifier to obtain label classification results of the features in the sample label space, and the bolt loosening positioning model is constructed.

6. A transmission tower bolt loosening positioning system, characterized in that: include: A signal acquisition module, used to collect vibration signals at multiple bolts of a target transmission tower; A signal decomposition module, used for decomposing the vibration signal based on the IVMD algorithm optimized by the Crested Porcupine Optimization Algorithm to obtain a plurality of modal components, and extracting a plurality of characteristic signals of the vibration signal using the plurality of modal components; A label assignment module, used to assign labels to the plurality of characteristic signals, wherein the labels include the position of the bolts and whether the bolts are loose; A data set construction module, used to construct a training data set according to the plurality of characteristic signals and the labels; A classification training module, used to perform classification training on the training data set using a convolutional neural network to obtain a bolt loosening positioning model; The positioning prediction module is used to use the bolt loosening positioning model to predict whether the unidentified bolts are loose, and to locate the loosened bolts.

7. 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 locating loose bolts of a transmission tower as described in any one of claims 1-5.

8. 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 locating loose bolts of a transmission tower as described in any one of claims 1 to 5 are implemented.

9. A computer program product, characterized in that The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer is caused to execute the steps of the method for locating loose bolts of a transmission tower as described in any one of claims 1-5.

Images