Power transmission line monitoring method, device, equipment, medium and program product

Abstract

The invention relates to a power transmission line monitoring method, device and equipment, a medium and a program product, and relates to the technical field of line monitoring. The power transmission line monitoring method comprises the following steps: acquiring a sound signal of a to-be-monitored range where a power transmission line is located; under the condition that the signal intensity of the sound signal is greater than a signal intensity threshold value, determining feature data of the sound signal; according to the feature data, determining the vehicle weight of the target vehicle running in the to-be-monitored range; and when the vehicle weight is greater than a preset weight threshold value, obtaining a monitoring video of the to-be-monitored range to monitor the target vehicle. Through the steps, the power transmission line can be monitored without continuous operation of the monitoring system, and the monitoring energy consumption is reduced. And meanwhile, monitoring videos or images captured in real time do not need to be analyzed, so that the monitoring efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of line monitoring, and particularly to a monitoring method, device, equipment, medium and program product for a transmission line. Background Art

[0002] With the continuous growth of power demand, the power system has become increasingly complex, and more transmission lines are required to transmit electrical energy. These transmission lines need to be properly protected and managed to ensure the reliability and safety of the power system.

[0003] In the traditional method, the transmission line is usually continuously monitored by a monitoring system to avoid external damage to the transmission line caused by construction excavators, cranes, etc. However, in this way, the monitoring system needs to run continuously, resulting in high energy consumption. At the same time, it is necessary to analyze the real-time captured monitoring videos or images to identify vehicles such as excavators and cranes, with a large amount of data processing and low monitoring efficiency. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a monitoring method, device, equipment, medium and program product for a transmission line that can reduce monitoring energy consumption and improve monitoring efficiency.

[0005] In a first aspect, the present application provides a monitoring method for a transmission line, including:

[0006] Obtaining a sound signal in a to-be-monitored range where the transmission line is located;

[0007] When the signal intensity of the sound signal is greater than a signal intensity threshold, determining characteristic data of the sound signal;

[0008] Determining the vehicle weight of a target vehicle operating in the to-be-monitored range according to the characteristic data;

[0009] When the vehicle weight is greater than a preset weight threshold, obtaining a monitoring video of the to-be-monitored range to monitor the target vehicle.

[0010] In one embodiment, determining the vehicle weight of a target vehicle operating in the to-be-monitored range according to the characteristic data includes: inputting the characteristic data into a vehicle weight recognition model to obtain the vehicle weight of the target vehicle; wherein, the vehicle weight recognition model is obtained through the following steps: obtaining different training sample data and the corresponding sample vehicle weights; the training sample data includes sample sound signals of the sample vehicle in an operating state; using different training sample data as input data of the vehicle weight recognition model, and using the corresponding sample vehicle weights as training labels to train the vehicle weight recognition model.

[0011] In one embodiment, training the vehicle weight recognition model includes: for each iteration process, obtaining the target population in the current iteration process, where the target population includes at least one population individual, and the population individual is used to represent the model parameters of the vehicle weight recognition model; for each population individual, determining the predicted vehicle weight obtained by the vehicle recognition model for different training sample data under the model parameters corresponding to the population individual; determining the mean error between each predicted vehicle weight and the corresponding sample vehicle weight; when the iteration termination condition is not satisfied, screening at least one target population individual from the target population according to each mean error, and performing crossover and mutation on the target population individual to obtain the target population in the next iteration process; when the iteration termination condition is satisfied, using the population individual corresponding to the minimum mean error among each mean error as the model parameters of the vehicle weight recognition model.

[0012] In one embodiment, it further includes: determining the target difference between the actual weight of the target vehicle and the vehicle weight; when the target difference is greater than the preset difference, adjusting the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight, and the loss function of the vehicle weight recognition model.

[0013] In one embodiment, determining the characteristic data of the sound signal includes: reconstructing the sound signal according to the low-frequency component and the high-frequency component of the sound signal to obtain at least one band signal; for each band signal, determining the characteristic vector of the band signal; using the characteristic vectors of different band signals as the characteristic data of the sound signal.

[0014] In one embodiment, obtaining the sound signal of the range to be monitored where the transmission line is located includes: obtaining the initial sound signal of the range to be monitored where the transmission line is located; converting the initial sound signal into an initial frequency domain signal; segmenting the initial frequency domain signal to obtain at least one sound signal.

[0015] In a second aspect, the present application further provides a monitoring device for a transmission line, including:

[0016] A first acquisition module, configured to acquire the sound signal of the range to be monitored where the transmission line is located;

[0017] A first determination module, configured to determine the characteristic data of the sound signal when the signal intensity of the sound signal is greater than the signal intensity threshold;

[0018] A second determination module, configured to determine the vehicle weight of the target vehicle running in the range to be monitored according to the characteristic data;

[0019] A second acquisition module, configured to acquire the monitoring video of the range to be monitored to monitor the target vehicle when the vehicle weight is greater than the preset weight threshold.

[0020] In a third aspect, the present application further provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0021] Obtain a sound signal in the range to be monitored where the transmission line is located;

[0022] When the signal strength of the sound signal is greater than the signal strength threshold, determine the characteristic data of the sound signal;

[0023] Determine the vehicle weight of the target vehicle running in the range to be monitored according to the characteristic data;

[0024] When the vehicle weight is greater than the preset weight threshold, obtain a monitoring video of the range to be monitored to monitor the target vehicle.

[0025] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0026] Obtain a sound signal in the range to be monitored where the transmission line is located;

[0027] When the signal strength of the sound signal is greater than the signal strength threshold, determine the characteristic data of the sound signal;

[0028] Determine the vehicle weight of the target vehicle running in the range to be monitored according to the characteristic data;

[0029] When the vehicle weight is greater than the preset weight threshold, obtain a monitoring video of the range to be monitored to monitor the target vehicle.

[0030] In a fifth aspect, the present application further provides a computer program product, including a computer program. When the computer program is executed by a processor, the following steps are implemented:

[0031] Obtain a sound signal in the range to be monitored where the transmission line is located;

[0032] When the signal strength of the sound signal is greater than the signal strength threshold, determine the characteristic data of the sound signal;

[0033] Determine the vehicle weight of the target vehicle running in the range to be monitored according to the characteristic data;

[0034] When the vehicle weight is greater than the preset weight threshold, obtain a monitoring video of the range to be monitored to monitor the target vehicle.

[0035] The above-mentioned monitoring method, device, equipment, medium and program product for transmission lines provide a raw data basis for the monitoring of transmission lines by acquiring the sound signals in the monitored range where the transmission lines are located. By comparing the signal intensity of the sound signals with a signal intensity threshold, filtering of invalid data is achieved, avoiding invalid data from participating in subsequent processing and calculations. When the signal intensity of the sound signals is greater than the signal intensity threshold, the characteristic data of the sound signals are determined, and based on the characteristic data, the vehicle weight of the target vehicle operating within the monitored range is determined. When the vehicle weight is greater than the preset weight threshold, it indicates that there is a large vehicle near the transmission line. At this time, by acquiring the monitoring video of the monitored range, the target vehicle is monitored to avoid external force damage to the transmission line caused by the target vehicle. Through the above method, the monitoring of the transmission line can be achieved without the continuous operation of the monitoring system, reducing the monitoring energy consumption. At the same time, there is no need to continuously capture monitoring videos or images and perform continuous recognition and analysis on them, improving the monitoring efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 It is a schematic flowchart of the monitoring method for transmission lines in an embodiment;

[0038] Figure 2 It is a schematic flowchart of the training steps of the vehicle weight recognition model in an embodiment;

[0039] Figure 3 It is a schematic flowchart of the determination steps of the characteristic data in an embodiment;

[0040] Figure 4 It is a schematic flowchart of the monitoring method for transmission lines in another embodiment;

[0041] Figure 5 It is a structural block diagram of the monitoring device for transmission lines in an embodiment;

[0042] Figure 6 It is an internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application.

[0044] In one embodiment, as Figure 1 shown, a method for monitoring a transmission line is provided. In this embodiment, it is exemplified that this method is applied to a terminal. It can be understood that this method can also be applied to a server or a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0045] S110. Obtain the sound signal of the area to be monitored where the transmission line is located.

[0046] Among them, the sound signal can be understood as the ambient sound signal of the area to be monitored where the transmission line is located. It is not difficult to understand that when different vehicles or objects pass through the area to be monitored, the ambient sound signal will also change accordingly. By obtaining the sound signal of the area to be monitored where the transmission line is located, an original data basis is provided for the monitoring of the transmission line.

[0047] In an optional embodiment, after obtaining the sound signal of the area to be monitored where the transmission line is located, the sound signal can also be preprocessed. Among them, the preprocessing includes at least one of filtering, denoising, and normalizing the sound signal.

[0048] In an optional embodiment, the initial sound signal of the area to be monitored where the transmission line is located can be obtained; the initial sound signal is converted into an initial frequency-domain signal; the initial frequency-domain signal is segmented to obtain at least one sound signal. Exemplarily, the initial sound signal can be converted into an initial frequency-domain signal through a fast Fourier transform.

[0049] S120. When the signal intensity of the sound signal is greater than the signal intensity threshold, determine the characteristic data of the sound signal.

[0050] In an optional embodiment, the signal intensity of the sound signal can be determined by determining the amplitude spectrum of the sound signal.

[0051] In an optional embodiment, the signal intensity threshold can be set to a preset multiple of the ambient noise. In another optional embodiment, the signal intensity threshold can also be determined by the mean, standard deviation, and sensitivity coefficient of the ambient noise. It should be noted that the signal intensity threshold can also be set by a technician according to needs or experience, or determined through a large number of experiments. This application does not make any limitations in this regard.

[0052] Exemplarily, the signal intensity threshold T kIt can be determined by the following formula:

[0053] ;

[0054] where T k represents the signal strength threshold; represents the mean value of the ambient noise; represents the standard deviation of the ambient noise; a represents the sensitivity coefficient. Among them, the sensitivity coefficient a can be set by those skilled in the art according to needs or experience, or determined through a large number of experiments. This application does not make any limitations on this. Exemplarily, the sensitivity coefficient a can be set to 2 or 3.

[0055] Exemplarily, the mean value of the ambient noise can be determined by the following formula:

[0056]

[0057] where represents the mean value of the ambient noise; represents the reference sound signal for estimating the ambient noise; S i represents the signal strength of the reference sound signal; N ref represents the number of reference sound signals.

[0058] Among them, the reference sound signal can be understood as the sound signal used to estimate the ambient noise in the sliding window. The reference sound signal can include the left reference sound signal and the right reference sound signal. The left reference sound signal can be understood as the reference sound signal before the target object passes through the range to be monitored, and the right reference sound signal can be understood as the reference sound signal after the target object passes through the range to be monitored.

[0059] Exemplarily, the standard deviation of the ambient noise can be determined by the following formula:

[0060]

[0061] where represents the standard deviation of the ambient noise; represents the reference sound signal for estimating the ambient noise; S i represents the signal strength of the reference sound signal; N ref represents the number of reference sound signals; represents the mean value of the ambient noise.

[0062] S130. Determine the vehicle weight of the target vehicle running within the range to be monitored according to the characteristic data.

[0063] In an optional embodiment, feature data can be input into a vehicle weight recognition model to obtain the vehicle weight of a target vehicle. The vehicle weight recognition model is obtained through the following steps: obtaining different training sample data and the corresponding sample vehicle weights of the training sample data; the training sample data includes sample sound signals of a sample vehicle in an operating state; using the different training sample data as input data of the vehicle weight recognition model and the corresponding sample vehicle weights as training labels, training the vehicle weight recognition model.

[0064] Among them, the vehicle weight recognition model can be a neural network model. Exemplarily, the vehicle weight recognition model can be trained through a genetic algorithm, and a trained vehicle weight recognition model is obtained through selection, crossover, and mutation. Exemplarily, the model training termination condition of the vehicle weight recognition model can be that the number of training sample data reaches a preset number, the model tends to converge, or the model accuracy reaches a preset accuracy threshold, etc., and the present application does not make any limitation in this regard.

[0065] S140. When the vehicle weight is greater than a preset weight threshold, obtain a monitoring video of a range to be monitored to monitor the target vehicle.

[0066] Among them, the preset weight threshold can be set by a technician according to needs or experience, or determined through a large number of experiments, and the present application does not make any limitation in this regard. It is not difficult to understand that when the vehicle weight is greater than the preset weight threshold, it means that there is a large target vehicle in the range to be monitored where the transmission line is located, and the construction operation of this large target vehicle is likely to damage the transmission line. In this case, obtain a monitoring video of the range to be monitored to monitor the target vehicle.

[0067] In an optional embodiment, when the vehicle weight is greater than the preset weight threshold, a wake-up request can be sent to a control terminal, so that in response to the wake-up request, the control terminal wakes up the corresponding monitoring device according to the location information in the wake-up request and obtains a monitoring video of the range to be monitored. In addition, the target vehicle can also be photographed and tracked in real time.

[0068] In an optional embodiment, the target distance between the target vehicle and a monitoring position within the monitoring range can be determined. When the target distance is greater than a preset distance, a stop message is sent to the monitoring device to terminate photographing and tracking of the target vehicle. Among them, the preset distance can be set by a technician according to needs or experience, and the present application does not make any limitation in this regard. Exemplarily, the preset distance can be 100m.

[0069] Exemplarily, the control terminal can determine the target distance between the target vehicle and the monitoring position within the monitoring range through the monitoring video or monitoring image; correspondingly, the control terminal can send a stop message to the monitoring device.

[0070] Exemplarily, the real-time position of the target vehicle can be determined by the intelligent road stud, and thus, according to the real-time position of the target vehicle, the target distance between the target vehicle and the monitoring position within the monitoring range can be determined. Correspondingly, the intelligent road stud can send a stop message to the monitoring device.

[0071] In an optional embodiment, when the vehicle weight is not greater than the preset weight threshold, a sleep request can be sent to the control terminal, so that the control terminal, in response to the sleep request, switches the corresponding monitoring device to the sleep state according to the position information in the wake-up request.

[0072] In an optional embodiment, YOLOv5 (You Only Look Once version 5, a deep learning-based object detection algorithm) can be used to predict the current position of the target vehicle by using the detection results in each frame of the monitoring image and the position information of the target vehicle in the previous frame of the image, and a Kalman filter is used for real-time tracking of the target vehicle. Thus, the advantages of object detection and tracking algorithms are combined, and the accuracy and robustness of target vehicle tracking are improved.

[0073] In the embodiments of the present application, by acquiring the sound signal in the monitoring range where the power transmission line is located, a raw data basis is provided for the monitoring of the power transmission line. By comparing the signal intensity of the sound signal with the signal intensity threshold, filtering of invalid data is achieved, and invalid data is avoided from participating in subsequent processing and calculation. When the signal intensity of the sound signal is greater than the signal intensity threshold, the characteristic data of the sound signal is determined, and according to the characteristic data, the vehicle weight of the target vehicle running within the range to be monitored is determined. When the vehicle weight is greater than the preset weight threshold, it indicates that there is a large vehicle near the power transmission line. At this time, by acquiring the monitoring video of the range to be monitored, the target vehicle is monitored to avoid external force damage to the power transmission line caused by the target vehicle. Through the above method, the monitoring of the power transmission line can be realized without the continuous operation of the monitoring system, reducing the monitoring energy consumption. At the same time, it is also unnecessary to continuously capture monitoring videos or images and perform continuous identification and analysis on them, improving the monitoring efficiency.

[0074] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment, in which the training steps of the vehicle weight recognition model are refined.

[0075] See Figure 2 The training steps of the vehicle weight recognition model shown in

[0076] S210. For each iteration process, obtain the target population in the current iteration process. The target population includes at least one population individual, and the population individual is used to represent the model parameters of the vehicle weight recognition model.

[0077] In an alternative embodiment, the initial model parameters of the vehicle weight recognition model can be used as the initial population individuals, and the initial population individuals are mutated and crossed to obtain the target population in the initial iteration process. In another alternative embodiment, multiple initial model parameters can be pre-selected and used as the target population in the initial iteration process.

[0078] S220. For each population individual, determine the predicted vehicle weight obtained by identifying different training sample data through the vehicle recognition model under the model parameters corresponding to the population individual.

[0079] S230. Determine the mean error between each predicted vehicle weight and the corresponding sample vehicle weight.

[0080] For ease of understanding, taking the population individual x as an example, F i represents the i-th training sample data, represents the sample vehicle weight corresponding to the i-th training sample data, represents the predicted vehicle weight corresponding to the i-th training sample data under the model parameters corresponding to the population individual x.

[0081] Exemplarily, the mean square error E between each predicted vehicle weight and the corresponding sample vehicle weight can be determined by the following formula:

[0082]

[0083] where E represents the sum of squared errors; N represents the number of training sample data.

[0084] Exemplarily, the sum of squared errors E can be divided by the number of training sample data N to obtain the mean squared error MSE corresponding to the population individual x.

[0085] S240. In the case where the iteration termination condition is not satisfied, at least one target population individual is screened out from the target population according to each mean squared error, and the target population individual is crossed and mutated to obtain the target population in the next iteration process.

[0086] In an alternative embodiment, screening out at least one target population individual from the target population may include: determining the fitness of different population individuals; determining the selection probability of different population individuals according to the fitness of the population individuals; and screening out at least one target population individual from the target population according to the selection probability of different population individuals.

[0087] Optionally, the fitness f of the population individual can be determined by the following formula:

[0088] f = -MSE;

[0089] Among them, MSE represents the mean error corresponding to the individuals in the population.

[0090] Optionally, the fitness f of each individual in the population can be summed to obtain the total fitness; for each individual in the population, the proportion of the fitness of the individual in the total fitness is determined to obtain the selection probability.

[0091] It should be further noted that performing crossover on the target population individuals can be understood as randomly selecting two or more parent individuals from the target population individuals for crossover operations, that is, generating new offspring individuals by exchanging some genes or parameters of the parent individuals.

[0092] It should be further noted that performing mutation on the target population individuals can be understood as performing mutation operations on the newly generated offspring individuals, that is, randomly changing some gene or parameter values of the offspring individuals according to a preset probability to increase the diversity of the population and prevent the algorithm from falling into a local optimal solution.

[0093] S250. When the iteration termination condition is satisfied, the individual in the population corresponding to the minimum mean error among the mean errors is used as the model parameter of the vehicle weight recognition model.

[0094] In an optional embodiment, the iteration termination condition can be that the number of iterations reaches a preset number, the model tends to converge, or the model accuracy reaches a preset accuracy threshold, etc. The present application does not make any limitation on this.

[0095] It can be understood that after the training is completed, the vehicle weight recognition model can be used to recognize the vehicle weight of the corresponding target vehicle through the sound signal. Among them, the relationship between the vehicle weight label output by the vehicle weight recognition model and the output layer can be expressed as:

[0096]

[0097] Among them, represents the activation value of the p-th node in the output layer; represents the vehicle weight label finally output by the vehicle weight recognition model.

[0098] In an optional embodiment, the monitoring method of the transmission line further includes: determining the target difference between the actual weight of the target vehicle and the vehicle weight; when the target difference is greater than the preset difference, adjusting the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight, and the loss function of the vehicle weight recognition model.

[0099] Optionally, the parameters of each layer of neural network in the vehicle weight recognition model can be adjusted sequentially by the backpropagation method.

[0100] Optionally, the model parameters of the vehicle weight recognition model can be adjusted by the following gradient update formula:

[0101]

[0102] Among them, represents the adjusted model parameters; represents the model parameters before adjustment; represents the learning rate; represents the loss function; y final represents the actual weight of the target vehicle; y represents the vehicle weight identified by the vehicle weight recognition model.

[0103] In the embodiments of the present application, through the iteration of the target population, the target population can be effectively optimized in each iteration process. By determining the mean error between the predicted vehicle weight and the sample vehicle weight of different training sample data under the corresponding model parameters, the accuracy of the model prediction can be quantified. When the iteration termination condition is not satisfied, high-quality target population individuals are selected based on the mean error, and through crossover and mutation operations, the population evolution is promoted, and the optimization efficiency of the model parameters is improved. When the iteration termination condition is satisfied, the population individual with the minimum mean error can be directly obtained as the optimal model parameters, thereby improving the recognition accuracy and generalization ability of the vehicle weight recognition model.

[0104] Based on the technical solutions of the above embodiments, the present application also provides an alternative embodiment, in which the step of determining the feature data is refined.

[0105] See Figure 3 The steps for determining the feature data shown include:

[0106] S310. Reconstruct the sound signal based on the low-frequency component and the high-frequency component of the sound signal to obtain at least one band signal.

[0107] Exemplarily, the sound signal can be decomposed into a low-frequency component and a high-frequency component by means of wavelet packet transform, which can be specifically expressed as:

[0108]

[0109] Among them, is the wavelet packet component; j represents the decomposition level; n represents the number of nodes; t represents the time or the signal sampling point index; h m represents the coefficient of the low-pass filter; g m represents the coefficient of the high-pass filter; m is the index of the filter, that is, the window width.

[0110] Exemplarily, through the weighted sum of the low-frequency component and the high-frequency component, the original signal can be reconstructed as:

[0111]

[0112] The frequency range corresponding to each frequency band is as follows:

[0113]

[0114] Among them, f s represents the signal sampling frequency; j represents the decomposition level; n represents the number of nodes; represents the width of each frequency band.

[0115] S320. For each frequency band signal, determine the feature vector of the frequency band signal.

[0116] Among them, the feature vector of the frequency band signal can include at least one of the frequency band signal energy, the frequency band signal mean, the frequency band signal standard deviation, and the frequency band signal skewness, etc.

[0117] For the sake of easy understanding, the following further explains the frequency band signal where j represents the decomposition level; n represents the number of nodes; t represents the time or the signal sampling point index, and T represents the time length of the frequency band signal.

[0118] Exemplarily, the frequency band signal energy can be extracted through the following formula :

[0119]

[0120] Exemplarily, the frequency band signal mean can be extracted through the following formula :

[0121]

[0122] Exemplarily, the frequency band signal standard deviation can be extracted through the following formula :

[0123]

[0124] Exemplarily, the frequency band signal skewness can be extracted through the following formula :

[0125]

[0126] Exemplarily, the feature vector F of the frequency band signal can be determined through the following formula :

[0127] .

[0128] S330. Use the feature vectors of different frequency band signals as the feature data of the sound signal.

[0129] In the embodiments of the present application, the sound signal is reconstructed according to the low-frequency component and the high-frequency component of the sound signal, so as to facilitate the analysis of the characteristics of the sound signal in different frequency ranges, and at the same time, it is also beneficial to improve the calculation efficiency. For each band signal, its feature vector is determined, so that the key attributes of the sound signal in this band can be captured. The feature vectors of these different band signals are integrated and used as the feature data of the sound signal, which can improve the richness of feature expression and weaken the influence of interference noise on the overall features.

[0130] Based on the technical solutions of the above embodiments, the present application also provides an alternative embodiment, in which the monitoring method of the transmission line is described in detail.

[0131] See Figure 4 The monitoring method of the shown transmission line includes:

[0132] S401. Obtain the initial sound signal of the area to be monitored where the transmission line is located.

[0133] S402. Convert the initial sound signal into an initial frequency-domain signal.

[0134] S403. Segment the initial frequency-domain signal to obtain at least one sound signal.

[0135] S404. When the signal intensity of the sound signal is greater than the signal intensity threshold, reconstruct the sound signal according to the low-frequency component and the high-frequency component of the sound signal to obtain at least one band signal.

[0136] S405. For each band signal, determine the feature vector of the band signal.

[0137] S406. Use the feature vectors of different band signals as the feature data of the sound signal.

[0138] S407. Input the feature data into the vehicle weight recognition model to obtain the vehicle weight of the target vehicle.

[0139] S408. When the vehicle weight is greater than the preset weight threshold, obtain the monitoring video of the area to be monitored to monitor the target vehicle.

[0140] S409. Determine the target difference between the actual weight of the target vehicle and the vehicle weight.

[0141] S410. When the target difference is greater than the preset difference, adjust the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight, and the loss function of the vehicle weight recognition model.

[0142] The detailed technical content of the above steps S401 - S410 has been recorded in the above examples and will not be repeated here.

[0143] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the indications of the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless there is a clear description in this article, there is no strict order limit for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0144] Based on the same inventive concept, an embodiment of the present application also provides a monitoring device for a transmission line for implementing the monitoring method of the transmission line involved above. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the monitoring device for the transmission line provided below can refer to the limitations on the monitoring method of the transmission line in the above text, and will not be repeated here.

[0145] In an exemplary embodiment, as Figure 5 shown, a monitoring device for a transmission line is provided, including: a first acquisition module 510, a first determination module 520, a second determination module 530, and a second acquisition module 540, where:

[0146] The first acquisition module 510 is configured to acquire a sound signal in a range to be monitored where the transmission line is located.

[0147] The first determination module 520 is configured to determine characteristic data of the sound signal when the signal intensity of the sound signal is greater than a signal intensity threshold.

[0148] The second determination module 530 is configured to determine the vehicle weight of the target vehicle running in the range to be monitored according to the characteristic data.

[0149] The second acquisition module 540 is configured to acquire a monitoring video of the range to be monitored to monitor the target vehicle when the vehicle weight is greater than a preset weight threshold.

[0150] In one embodiment, the second determination module 530 includes: an input unit configured to input the feature data into the vehicle weight recognition model to obtain the vehicle weight of the target vehicle. The vehicle weight recognition model is obtained through the following steps: obtaining different training sample data and the corresponding sample vehicle weights of the training sample data; the training sample data includes sample sound signals of the sample vehicle in an operating state; using the different training sample data as input data of the vehicle weight recognition model, and using the corresponding sample vehicle weights as training labels, training the vehicle weight recognition model.

[0151] In one embodiment, the monitoring device of the transmission line further includes a training module, including: a first acquisition unit configured to, for each iteration process, acquire the target population in the current iteration process, the target population including at least one population individual, and the population individual being used to represent the model parameters of the vehicle weight recognition model; a first determination unit configured to, for each population individual, determine the predicted vehicle weight obtained by recognizing different training sample data through the vehicle recognition model under the model parameters corresponding to the population individual; a second determination unit configured to determine the mean error between each predicted vehicle weight and the corresponding sample vehicle weight; a first processing unit configured to, when the iteration termination condition is not satisfied, screen out at least one target population individual from the target population according to each mean error, and perform crossover and mutation on the target population individual to obtain the target population in the next iteration process; a second processing unit configured to, when the iteration termination condition is satisfied, use the population individual corresponding to the minimum mean error among each mean error as the model parameters of the vehicle weight recognition model.

[0152] In one embodiment, the monitoring device of the transmission line further includes: a third determination module configured to determine the target difference between the actual weight of the target vehicle and the vehicle weight; an adjustment module configured to, when the target difference is greater than the preset difference, adjust the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight, and the loss function of the vehicle weight recognition model.

[0153] In one embodiment, the first determination module 520 includes: a reconstruction unit configured to reconstruct the sound signal according to the low-frequency component and the high-frequency component of the sound signal to obtain at least one band signal; a third determination unit configured to, for each band signal, determine the feature vector of the band signal; an integration unit configured to use the feature vectors of different band signals as the feature data of the sound signal.

[0154] In one embodiment, the first acquisition module 510 includes: a second acquisition unit configured to acquire the initial sound signal in the to-be-monitored range where the transmission line is located; a conversion unit configured to convert the initial sound signal into an initial frequency domain signal; a segmentation unit configured to segment the initial frequency domain signal to obtain at least one sound signal.

[0155] Each module in the above monitoring device for a transmission line can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above modules can be embedded in the processor in the computer device in hardware form or independent of it, or stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0156] In an exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 6 shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC), or other technologies. The computer program, when executed by the processor, implements a method for monitoring a transmission line. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0157] Those skilled in the art can understand that Figure 6 the structure shown in

[0158] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0158] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0159] Obtain the sound signal in the range to be monitored where the transmission line is located;

[0160] When the signal strength of the sound signal is greater than the signal strength threshold, determine the characteristic data of the sound signal;

[0161] According to the characteristic data, determine the vehicle weight of the target vehicle running within the range to be monitored;

[0162] When the vehicle weight is greater than the preset weight threshold, obtain the monitoring video of the range to be monitored to monitor the target vehicle.

[0163] In one embodiment, when the processor executes the computer program, the following steps are further implemented: input the characteristic data into the vehicle weight recognition model to obtain the vehicle weight of the target vehicle; wherein, the vehicle weight recognition model is obtained through the following steps: obtain different training sample data and the corresponding sample vehicle weights of the training sample data; the training sample data includes the sample sound signals of the sample vehicle in the running state; use different training sample data as the input data of the vehicle weight recognition model, and use the corresponding sample vehicle weights as the training labels to train the vehicle weight recognition model.

[0164] In one embodiment, when the processor executes the computer program, the following steps are further implemented: for each iteration process, obtain the target population in the current iteration process, the target population includes at least one population individual, and the population individual is used to represent the model parameters of the vehicle weight recognition model; for each population individual, determine the predicted vehicle weight obtained by identifying different training sample data through the vehicle recognition model under the model parameters corresponding to the population individual; determine the error mean value between each predicted vehicle weight and the corresponding sample vehicle weight; when the iteration termination condition is not satisfied, according to each error mean value, screen out at least one target population individual from the target population, and perform crossover and mutation on the target population individual to obtain the target population in the next iteration process; when the iteration termination condition is satisfied, use the population individual corresponding to the minimum error mean value among each error mean value as the model parameters of the vehicle weight recognition model.

[0165] In one embodiment, when the processor executes the computer program, the following steps are further implemented: determine the target difference between the actual weight of the target vehicle and the vehicle weight; when the target difference is greater than the preset difference, adjust the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight and the loss function of the vehicle weight recognition model.

[0166] In one embodiment, when the processor executes the computer program, the following steps are further implemented: reconstruct the sound signal according to the low-frequency component and the high-frequency component of the sound signal to obtain at least one band signal; for each band signal, determine the feature vector of the band signal; use the feature vectors of different band signals as the characteristic data of the sound signal.

[0167] In one embodiment, when the processor executes the computer program, the following steps are further implemented: obtaining an initial sound signal of the range to be monitored where the transmission line is located; converting the initial sound signal into an initial frequency-domain signal; segmenting the initial frequency-domain signal to obtain at least one sound signal.

[0168] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0169] Obtaining a sound signal of the range to be monitored where the transmission line is located;

[0170] When the signal intensity of the sound signal is greater than a signal intensity threshold, determining characteristic data of the sound signal;

[0171] Determining the vehicle weight of the target vehicle running in the range to be monitored according to the characteristic data;

[0172] When the vehicle weight is greater than a preset weight threshold, obtaining a monitoring video of the range to be monitored to monitor the target vehicle.

[0173] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: inputting the characteristic data into a vehicle weight recognition model to obtain the vehicle weight of the target vehicle; wherein, the vehicle weight recognition model is obtained through the following steps: obtaining different training sample data and the sample vehicle weight corresponding to the training sample data; the training sample data includes sample sound signals of the sample vehicle in the running state; using different training sample data as input data of the vehicle weight recognition model, and using the corresponding sample vehicle weight as a training label to train the vehicle weight recognition model.

[0174] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: for each iteration process, obtaining a target population in the current iteration process, the target population includes at least one population individual, and the population individual is used to represent the model parameters of the vehicle weight recognition model; for each population individual, determining the predicted vehicle weight obtained by the vehicle recognition model for different training sample data under the model parameters corresponding to the population individual; determining the mean error between each predicted vehicle weight and the corresponding sample vehicle weight; when the iteration termination condition is not satisfied, screening out at least one target population individual from the target population according to each mean error, and performing crossover and mutation on the target population individual to obtain the target population in the next iteration process; when the iteration termination condition is satisfied, using the population individual corresponding to the minimum mean error among each mean error as the model parameters of the vehicle weight recognition model.

[0175] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: determining a target difference between the actual weight of a target vehicle and the vehicle weight; and when the target difference is greater than a preset difference, adjusting the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight, and the loss function of the vehicle weight recognition model.

[0176] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: reconstructing a sound signal to obtain at least one band signal according to the low-frequency component and the high-frequency component of the sound signal; for each band signal, determining a feature vector of the band signal; and using the feature vectors of different band signals as the feature data of the sound signal.

[0177] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: obtaining an initial sound signal in a range to be monitored where a power transmission line is located; converting the initial sound signal into an initial frequency domain signal; and segmenting the initial frequency domain signal to obtain at least one sound signal.

[0178] In one embodiment, a computer program product is provided, including a computer program, which when executed by a processor, implements the following steps:

[0179] obtaining a sound signal in a range to be monitored where a power transmission line is located;

[0180] when the signal intensity of the sound signal is greater than a signal intensity threshold, determining the feature data of the sound signal;

[0181] determining the vehicle weight of a target vehicle operating in the range to be monitored according to the feature data;

[0182] when the vehicle weight is greater than a preset weight threshold, obtaining a monitoring video of the range to be monitored to monitor the target vehicle.

[0183] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: inputting the feature data into a vehicle weight recognition model to obtain the vehicle weight of the target vehicle; wherein, the vehicle weight recognition model is obtained through the following steps: obtaining different training sample data and the corresponding sample vehicle weights; the training sample data includes sample sound signals of a sample vehicle in an operating state; using different training sample data as input data of the vehicle weight recognition model, and using the corresponding sample vehicle weights as training labels to train the vehicle weight recognition model.

[0184] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: for each iteration process, obtain the target population under the current iteration process, where the target population includes at least one population individual, and the population individual is used to represent the model parameters of the vehicle weight recognition model; for each population individual, determine the predicted vehicle weight obtained by the vehicle recognition model for different training sample data under the model parameters corresponding to the population individual; determine the mean error between each predicted vehicle weight and the corresponding sample vehicle weight; when the iteration termination condition is not met, screen out at least one target population individual from the target population according to each mean error, and perform crossover and mutation on the target population individual to obtain the target population under the next iteration process; when the iteration termination condition is met, use the population individual corresponding to the minimum mean error among each mean error as the model parameters of the vehicle weight recognition model.

[0185] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: determine the target difference between the actual weight of the target vehicle and the vehicle weight; when the target difference is greater than the preset difference, adjust the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight, and the loss function of the vehicle weight recognition model.

[0186] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: reconstruct the sound signal according to the low-frequency component and the high-frequency component of the sound signal to obtain at least one band signal; for each band signal, determine the feature vector of the band signal; use the feature vectors of different band signals as the feature data of the sound signal.

[0187] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented: obtain the initial sound signal of the monitoring range where the transmission line is located; convert the initial sound signal into an initial frequency domain signal; segment the initial frequency domain signal to obtain at least one sound signal.

[0188] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include Read-Only Memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, Resistive Random Access Memory (ReRAM), Magnetoresistive Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene memory, etc. Volatile memory can include Random Access Memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, Artificial Intelligence (AI) processors, etc., without limitation.

[0189] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in the present application.

[0190] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A monitoring method for a transmission line, characterized in that, The method includes: Obtaining a sound signal of a to-be-monitored range where a transmission line is located; Determining characteristic data of the sound signal when the signal intensity of the sound signal is greater than a signal intensity threshold; Inputting the characteristic data into a vehicle weight recognition model to obtain the vehicle weight of a target vehicle operating in the to-be-monitored range; Obtaining a surveillance video of the to-be-monitored range to monitor the target vehicle when the vehicle weight is greater than a preset weight threshold; The vehicle weight recognition model is obtained through the following steps: Obtaining different training sample data and the corresponding sample vehicle weights of the training sample data; the training sample data includes sample sound signals of a sample vehicle in an operating state; Using different training sample data as input data of the vehicle weight recognition model and using the corresponding sample vehicle weights as training labels to train the vehicle weight recognition model.

2. The method according to claim 1, characterized in that, The training of the vehicle weight recognition model includes: For each iteration process, obtaining a target population in the current iteration process, where the target population includes at least one population individual, and the population individual is used to represent the model parameters of the vehicle weight recognition model; For each population individual, determining the predicted vehicle weight obtained by recognizing different training sample data through the vehicle recognition model under the model parameters corresponding to the population individual; Determining the mean error between each predicted vehicle weight and the corresponding sample vehicle weight; When the iteration termination condition is not satisfied, screening at least one target population individual from the target population according to each mean error, and performing crossover and mutation on the target population individual to obtain a target population in the next iteration process; When the iteration termination condition is satisfied, using the population individual corresponding to the minimum mean error among each mean error as the model parameters of the vehicle weight recognition model.

3. The method according to claim 1, characterized in that, The method further includes: Determining a target difference between the actual weight of the target vehicle and the vehicle weight; When the target difference is greater than a preset difference, adjusting the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight, and the loss function of the vehicle weight recognition model.

4. The method according to any one of claims 1 to 3, characterized in that, The determination of the characteristic data of the sound signal includes: Reconstructing the sound signal according to the low-frequency component and the high-frequency component of the sound signal to obtain at least one band signal; For each band signal, determining the feature vector of the band signal; Using the feature vectors of different band signals as the characteristic data of the sound signal.

5. The method according to any one of claims 1 to 3, characterized in that The obtaining of the sound signal of the to-be-monitored range where the transmission line is located includes: Obtaining an initial sound signal of the to-be-monitored range where the transmission line is located; Converting the initial sound signal into an initial frequency domain signal; Segmenting the initial frequency domain signal to obtain at least one of the sound signals.

6. A monitoring device for a transmission line, characterized in that, The device includes: A first obtaining module, configured to obtain a sound signal of a to-be-monitored range where a transmission line is located; A first determining module, configured to determine the characteristic data of the sound signal when the signal intensity of the sound signal is greater than a signal intensity threshold; A second determination module, configured to input the feature data into a vehicle weight recognition model to obtain the vehicle weight of the target vehicle running within the range to be monitored; A second acquisition module, configured to acquire a monitoring video of the range to be monitored to monitor the target vehicle when the vehicle weight is greater than a preset weight threshold; Wherein, the vehicle weight recognition model is obtained through the following steps: obtaining different training sample data and the sample vehicle weight corresponding to the training sample data; the training sample data includes the sample sound signal of the sample vehicle in the running state; using different training sample data as the input data of the vehicle weight recognition model, and using the corresponding sample vehicle weight as the training label to train the vehicle weight recognition model.

7. The device according to claim 6, wherein It further includes: A third determination module, configured to determine the target difference between the actual weight of the target vehicle and the vehicle weight; An adjustment module, configured to adjust the model parameters of the vehicle weight recognition model according to the actual weight, the vehicle weight, and the loss function of the vehicle weight recognition model when the target difference is greater than a preset difference.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images