A method, system, terminal and storage medium for predicting life of pot-type insulator

Abstract

The present invention relates to the technical field of maintenance and diagnosis of high-voltage power transmission equipment, and specifically provides a method, system, terminal and storage medium for predicting the life of a pot-type insulator, including: selecting a plurality of pot-type insulators with different service times, and dividing a plurality of measuring points on each pot-type insulator; building a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform based on an ultrasonic wedge with multiple channels and a plurality of groups of ultrasonic probes with different frequencies; obtaining waveform data of ultrasonic critical refraction longitudinal waves of pot-type insulators with different service times at each measuring point and each channel based on the multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform, extracting feature data after dimension reduction of waveform data based on principal component analysis; constructing a remaining life prediction model, inputting the current pot-type insulator feature data into the remaining life prediction model, and outputting the predicted remaining life of the current pot-type insulator. Accurately reveal the evolution process of its internal defects, and realize the accurate evaluation of the pot-type insulator state as a whole.

Description

Technical Field

[0001] The present invention belongs to the technical field of high-voltage power transmission equipment maintenance and diagnosis, and in particular relates to a method, system, terminal and storage medium for predicting the life of a pot-type insulator. Background Art

[0002] The pot insulator is mainly composed of insulating materials and metal accessories. It is usually pot-shaped or disc-shaped, with a hole in the center for installing the conductor. The insulating material is generally made of high-quality composite materials such as epoxy resin. The metal accessories are usually installed around the edge and center hole of the insulator to connect other electrical equipment components or fix the conductor to ensure the stability and reliability of the electrical connection. Pot insulators are widely used in high-voltage transmission systems.

[0003] After long-term service, the performance of the pot insulator may decline due to mechanical stress, environmental factors, etc., which will eventually affect the safe operation of the system. Referring to the invention patent CN105203879B, it is necessary to sample the pot insulator after artificial accelerated aging test, test its resistivity and electric field strength, and then obtain the life prediction model of the pot insulator under the working field strength and evaluate the life of the pot insulator.

[0004] The resistivity and electric field strength are mainly used to evaluate the status of pot-type insulators from the external or overall performance level. This is a macroscopic test and it is difficult to accurately reveal the evolution process of its internal defects, resulting in large errors in the life prediction of pot-type insulators. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method, system, terminal and storage medium for predicting the life of a pot-type insulator to solve the above-mentioned technical problems.

[0006] In a first aspect, the present invention provides a method for predicting the life of a pot-type insulator, comprising:

[0007] S1, select multiple pot-type insulators with different service times, and divide multiple measuring points on each pot-type insulator;

[0008] S2, building a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform based on a multi-channel ultrasonic wedge and multiple sets of ultrasonic probes with different frequencies;

[0009] S3, based on the multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform, the waveform data of ultrasonic critical refraction longitudinal waves of pot-type insulators with different service times at each measuring point and each channel are obtained, and the characteristic data after the waveform data is reduced in dimension is extracted based on the principal component analysis method;

[0010] S4, constructing a remaining life prediction model based on a neural network algorithm, inputting the current basin insulator characteristic data into the remaining life prediction model, and outputting the predicted remaining life of the current basin insulator.

[0011] In an optional embodiment, the multiple groups of ultrasonic probes with different frequencies include 3 ultrasonic transmitting probes and 3 ultrasonic receiving probes, the ultrasonic transmitting probes and the ultrasonic receiving probes correspond to each other one by one, and the frequencies of each group of ultrasonic probes are 0.5 MHz, 2.25 MHz, and 7.5 MHz respectively;

[0012] The ultrasonic wedge is a hexagonal wedge with three channels, which are a 0.5 MHz channel, a 2.25 MHz channel, and a 7.5 MHz channel.

[0013] In an optional embodiment, each set of ultrasound probes includes an ultrasound transmitting probe and an ultrasound receiving probe, and the multi-frequency ultrasound critical refraction longitudinal wave transceiver platform specifically includes:

[0014] A signal generator configured to generate an electrical excitation signal of a preset frequency;

[0015] A plurality of ultrasonic transmitting probes, each of which is configured to receive the electrical excitation signal at a preset frequency and convert the electrical excitation signal into an ultrasonic wave of a preset frequency based on an inverse piezoelectric effect;

[0016] An ultrasonic wedge configured to couple ultrasonic waves of a preset frequency into the pot insulator and generate ultrasonic critical refracted longitudinal waves on the surface of the insulator;

[0017] A plurality of ultrasonic receiving probes are configured to receive critical refracted longitudinal waves and convert the ultrasonic critical refracted longitudinal waves into electrical signals based on the piezoelectric effect;

[0018] an oscilloscope configured to receive the electrical signal to display the waveform of the ultrasonic critical refracted longitudinal wave;

[0019] The PC is configured to collect waveform data displayed by the oscilloscope.

[0020] In an optional implementation, in step S3, acquiring the characteristic data specifically includes:

[0021] Determine the amplitude measurement point in the waveform data and calculate the amplitude, determine the start reference point and the end reference point in the waveform in the waveform data and calculate the propagation time;

[0022] Perform Fourier transform on the waveform data to obtain the spectrum energy distribution;

[0023] Construct the original characteristic parameter data set based on amplitude, propagation time and spectral energy distribution;

[0024] Based on the principal component analysis method, the original feature parameter data set is reduced in dimension to obtain feature data.

[0025] In an optional embodiment, in step S4, the remaining life prediction model is constructed by using a multi-layer perceptron neural network structure, and the specific steps include:

[0026] Obtain historical feature data and the corresponding historical remaining life and construct a data set, and divide the data set into a training set, a validation set, and a test set, wherein the historical feature data is used as an input vector, and the historical remaining life is used as an output vector. The input layer is determined according to the data in the input vector, and the output layer is determined according to the data in the output vector.

[0027] For each sample in the training set, the input vector is fed into the remaining life prediction model, and the predicted output vector of the model is calculated by forward propagation in the hidden layer;

[0028] The loss value between the predicted output vector and the actual output feature vector is calculated according to the loss function, and the gradient of the loss function to the model weights and biases is calculated using the back-propagation algorithm, and the weights and biases of the model are updated using the gradient information;

[0029] The preset number of training times per interval is used to calculate the evaluation index based on the validation set combined with the root mean square error function, and the model structure is adjusted based on the changing trend of the evaluation index;

[0030] The trained remaining life prediction model is finally evaluated based on the test set.

[0031] In an optional embodiment, the predicted output vector of the model calculated by forward propagation in the hidden layer specifically includes:

[0032] Each node in the hidden layer performs a weighted summation of the output information of the previous layer:

[0033]

[0034] Where n is the number of nodes in the previous layer, is the output information of the i-th node in the previous layer, is the weight connecting the i-th node in the previous layer and the j-th node in the current layer, is the bias of the jth node in the current layer;

[0035] The result of the weighted sum is transformed nonlinearly through the activation function to obtain the output of the node;

[0036] After multiple hidden layers are processed layer by layer, the output layer receives the output of the last hidden layer to obtain the predicted remaining lifespan.

[0037] In an optional implementation, step S1 specifically includes: dividing the area below the convex side insert of the pot insulator into multiple sub-areas, the sub-areas are distributed in a ring shape around the center of the pot insulator, and selecting multiple measuring points on the circumference corresponding to each sub-area.

[0038] In a second aspect, the present invention provides a pot-type insulator life prediction system, which executes the above-mentioned pot-type insulator life prediction method when the system is implemented, and the system includes:

[0039] The measuring point division module selects multiple pot-type insulators with different service times and divides multiple measuring points on each pot-type insulator;

[0040] A platform building module, which builds a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform based on a multi-channel ultrasonic wedge and multiple sets of ultrasonic probes with different frequencies;

[0041] The feature extraction module obtains the waveform data of ultrasonic critical refraction longitudinal waves of pot-type insulators with different service times at each measuring point and each channel based on the multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform, and extracts the feature data after the waveform data is reduced in dimension based on the principal component analysis method;

[0042] The life prediction module builds a remaining life prediction model based on a neural network algorithm, inputs the current pot-type insulator characteristic data into the remaining life prediction model, and outputs the predicted remaining life of the current pot-type insulator.

[0043] In a third aspect, a terminal is provided, including:

[0044] processor, memory, wherein:

[0045] The memory is used to store computer programs.

[0046] The processor is used to call and run the computer program from the memory, so that the terminal executes the above-mentioned terminal method.

[0047] In a fourth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer executes the methods described in the above aspects.

[0048] The beneficial effect of the present invention lies in that the life prediction method, system, terminal and storage medium of the pot-type insulator provided by the present invention can effectively realize the prediction of the remaining life of the pot-type insulator by selecting pot-type insulators with different service times and dividing the measuring points, building a multi-frequency ultrasonic transceiver platform to obtain waveform data and extract feature data, and then using a neural network algorithm to build a prediction model, so as to provide strong support for the maintenance of high-voltage power transmission equipment, comprehensively evaluate the different service states and position information of insulators, enhance the detection accuracy and depth resolution capability by multi-frequency ultrasonic detection, improve the data processing efficiency and model performance by principal component analysis dimensionality reduction, and accurately establish a life prediction model by a neural network algorithm, which is helpful to ensure the safe and stable operation of the high-voltage power transmission system, accurately reveal the evolution process of its internal defects, discover possible safety hazards in advance, and avoid system failures caused by insulator performance degradation.

[0049] In addition, the invention has a reliable design principle, a simple structure and a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0051] Figure 1 It is a schematic flow chart of a method for predicting the life of a pot-type insulator according to an embodiment of the present invention.

[0052] Figure 2 It is a structural diagram of a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform according to an embodiment of the present invention.

[0053] Figure 3 The figure is a structural view of an ultrasonic wedge and a probe according to an embodiment of the present invention.

[0054] Figure 4 FIG. 4 is another structural view of an ultrasonic wedge and a probe according to an embodiment of the present invention.

[0055] Figure 5 It is a schematic block diagram of a pot-type insulator life prediction system according to an embodiment of the present invention.

[0056] Figure 6 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions 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 should fall within the scope of protection of the present invention.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0059] The pot-type insulator life prediction method provided in the embodiment of the present invention is executed by a computer device, and accordingly, the pot-type insulator life prediction system runs in the computer device.

[0060] Figure 1 FIG. 1 is a schematic flow chart of a method for predicting the life of a pot-type insulator according to an embodiment of the present invention. Figure 1 The execution subject may be a basin insulator life prediction system. According to different requirements, the order of the steps in the flow chart may be changed, and some may be omitted.

[0061] like Figure 1 As shown, the method includes:

[0062] S1, select multiple pot-type insulators with different service times, and divide multiple measuring points on each pot-type insulator;

[0063] Selecting pot-type insulators with different service times can cover all stages of the insulator from its initial state to the possible performance degradation as the service time increases; since the mechanical stress and environmental influences on different parts of the insulator may vary during operation, multiple measuring points can comprehensively detect the status of all directions of the insulator, avoiding the one-sidedness of local detection, thereby more accurately evaluating the overall health of the insulator.

[0064] S2, building a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform based on a multi-channel ultrasonic wedge and multiple sets of ultrasonic probes with different frequencies;

[0065] Using multiple sets of ultrasonic probes with different frequencies, the ultrasonic critical refracted longitudinal waves of different frequencies have different penetration depths in the insulator. The low-frequency ultrasonic penetration depth is relatively deep, and it can detect information at deeper locations inside the insulator, and can be used to detect possible deep defects; the high-frequency ultrasonic penetration depth is shallow, and it is more sensitive to tiny surface and near-surface defects, and can more accurately detect early surface damage.

[0066] S3, based on the multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform, the waveform data of ultrasonic critical refraction longitudinal waves of pot-type insulators with different service times at each measuring point and each channel are obtained, and the characteristic data after the waveform data is reduced in dimension is extracted based on the principal component analysis method;

[0067] It can reflect the changes in the internal structure of the insulator from multiple dimensions, such as material aging, the generation and development of microscopic defects, etc., which helps to deeply explore the internal mechanism of insulator performance degradation.

[0068] S4, constructing a remaining life prediction model based on a neural network algorithm, inputting the current basin insulator characteristic data into the remaining life prediction model, and outputting the current basin insulator predicted remaining life;

[0069] The characteristic data of the current pot-type insulator is input into the prediction model in real time and the remaining life prediction results are quickly output, so that the operation and maintenance personnel can understand the health status of the insulator in a timely manner. It helps to accurately reveal the evolution process of its internal defects during the operation of the high-voltage transmission system, discover possible safety hazards in advance, and avoid system failures caused by the degradation of insulator performance.

[0070] Optionally, as an embodiment of the present invention, step S1 specifically includes: since the area below the convex side insert is a key part of the insulator, its internal structure and electric field distribution and other factors have an important influence on the performance of the insulator, the area below the convex side insert of the pot-type insulator is divided into four sub-areas, and the sub-areas are distributed in a ring shape around the center of the pot-type insulator, and the same number of measuring points are selected on the circumference corresponding to each sub-area.

[0071] Optionally, as an embodiment of the present invention, refer to Figure 2 Each set of ultrasonic probes includes an ultrasonic transmitting probe and an ultrasonic receiving probe. The multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform specifically includes:

[0072] A signal generator is configured to generate an electrical excitation signal of a preset frequency (0.5 MHz, 2.25 MHz or 7.5 MHz). The electrical excitation signal is the starting point of the entire signal transmission process, and its frequency determines the frequency of the ultrasonic wave generated by the subsequent ultrasonic transmitting probe;

[0073] The high-voltage amplifier is connected to the signal generator and is configured to increase the voltage or power of the signal to a level sufficient to drive the ultrasonic transmitting probe to generate ultrasonic waves;

[0074] A multiplexer 1 is used to distribute the amplified electrical excitation signal output by the high-voltage amplifier to ultrasonic transmitting probes of different frequencies;

[0075] A plurality of ultrasonic transmitting probes, each of which is configured to receive the amplified electrical excitation signal at a frequency corresponding to the preset frequency of the electrical excitation signal, and convert the electrical excitation signal into an ultrasonic wave of a preset frequency based on the inverse piezoelectric effect. For example, when a 0.5 MHz amplified electrical signal reaches a 0.5 MHz ultrasonic transmitting probe, the piezoelectric crystal material of the probe will deform to generate an ultrasonic wave of a frequency of 0.5 MHz.

[0076] The ultrasonic wedge is configured to adjust the incident angle of the ultrasonic wave and couple the ultrasonic wave of the preset frequency into the pot insulator, so as to generate an ultrasonic critical refraction longitudinal wave on the surface of the insulator and propagate along the surface of the insulator within a certain depth range. During the propagation process, the ultrasonic wave will interact with the internal structure of the insulator, and its propagation characteristics (such as amplitude, propagation time, etc.) will be affected by the material properties, microstructure and possible defects of the internal insulator;

[0077] A plurality of ultrasonic receiving probes are configured to receive critical refracted longitudinal waves and convert the ultrasonic critical refracted longitudinal waves into electrical signals based on the piezoelectric effect;

[0078] A multiplexer 2 is used to distribute the electrical signals received by different ultrasound receiving probes to the signal amplifiers;

[0079] Signal amplifier: Since the electrical signal converted by the ultrasonic receiving probe is usually weak, the signal amplifier increases the strength of the received signal to a level suitable for processing by devices such as oscilloscopes and PCs;

[0080] An oscilloscope, configured to receive electrical signals to display the waveform of the ultrasonic critical refracted longitudinal wave, through which the shape, amplitude, period and other characteristics of the waveform can be visually observed to preliminarily determine the situation inside the insulator;

[0081] The PC is configured to collect waveform data displayed by the oscilloscope. The PC can perform denoising, feature extraction (including amplitude, propagation time, spectrum characteristics, etc.) and subsequent construction of a life prediction model based on a neural network. By analyzing the waveform data of insulators at a large number of different measuring points, different frequencies and different service times, the PC can mine information related to the remaining service life of the insulators, thereby realizing the prediction of the life of pot-type insulators.

[0082] Optionally, as an embodiment of the present invention, the multiple groups of ultrasonic probes with different frequencies include 3 ultrasonic transmitting probes and 3 ultrasonic receiving probes, the ultrasonic transmitting probes and the ultrasonic receiving probes correspond to each other one by one, and the frequencies of each group of ultrasonic probes are 0.5 MHz, 2.25 MHz, and 7.5 MHz respectively;

[0083] Ultrasonic waves of different frequencies have different propagation characteristics in insulators. The penetration depth of 0.5MHz ultrasonic waves is relatively deep, and it can detect information at deeper locations inside the insulator; the penetration depth and sensitivity to tiny defects of 2.25MHz ultrasonic waves are between 0.5MHz and 7.5MHz; the penetration depth of 7.5MHz ultrasonic waves is shallow, but it is more sensitive to tiny defects on the surface and near the surface. By using a combination of probes with these three frequencies, structural information at different depths inside the insulator can be obtained, thereby more comprehensively evaluating the internal condition of the insulator.

[0084] Reference Figure 3 and Figure 4 ,The ultrasonic wedge is a hexagonal wedge with three channels, which are 0.5MHz channel, 2.25MHz channel, and 7.5MHz channel;

[0085] A channel is formed between the two parallel ends of the hexagonal wedge, and a set of ultrasonic transmitting probes and ultrasonic receiving probes are respectively close to one of the two parallel ends. When testing, a set of ultrasonic transmitting probes and ultrasonic receiving probes are respectively close to one of the two parallel ends, so that the ultrasonic wave of the preset frequency generated by the transmitting probe can enter the insulator through the corresponding wedge channel, and after propagating in the insulator, the reflected or refracted ultrasonic wave can be received by the receiving probe through the same channel.

[0086] Optionally, as an embodiment of the present invention, the above-mentioned multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform is used to place the ultrasonic wedge on the convex surface of the pot insulator, and alcohol is used as a coupling agent. The signal generator generates a 0.5MHz excitation signal, selects the 0.5MHz ultrasonic transmitting probe and the channel corresponding to the wedge, and the electrical signal is amplified and converted into an acoustic signal, propagates in the insulator, and is picked up by the 0.5MHz ultrasonic receiving probe after a certain period of time.

[0087] Subsequently, the frequencies of the excitation signals generated by the signal generator were changed to 2.25 MHz and 7.5 MHz in turn, and the probes and channels of the corresponding frequencies were selected accordingly.

[0088] The waveform of the ultrasonic critical refracted longitudinal wave with a frequency of j MHz received at the i-th measuring point of a pot-type insulator that has been in service for k months is: , where k = 0, 1, …, N-1; i = 0, 1, …M; j = 0.5, 2.25, 7.5; and t is time.

[0089] Optionally, as an embodiment of the present invention, in step S3, acquiring the characteristic data specifically includes:

[0090] Use wavelet transform to transform waveform signal Perform denoising to obtain denoised waveform data .

[0091] Determine amplitude measurement points in waveform data and calculate amplitude , determine the starting and ending reference points in the waveform data and calculate the propagation time :

[0092] Select the peak or trough in the waveform as the measurement point, measure the vertical distance between the selected peak and trough and calculate it as the amplitude;

[0093] Select the starting characteristic point of the waveform corresponding to the moment when the transmitting probe emits ultrasonic waves as the starting reference point, such as the point where the waveform starts to rise after the transmitting signal is triggered; select the point where the receiving probe receives the end of the main signal part of the critical refracted longitudinal wave as the ending reference point, such as the point where the waveform drops close to the baseline and remains stable; calculate the time interval from the starting reference point to the ending reference point as the propagation time.

[0094] Perform Fourier transform on waveform data to obtain spectrum energy distribution , calculated as:

[0095]

[0096] Where t is the time, is the frequency, Represents the Fourier transform operation.

[0097] Construct original characteristic parameter data set based on amplitude, propagation time and spectral energy distribution X ;

[0098] Based on the principal component analysis method, the original feature parameter data set is reduced in dimension to obtain the feature data, which is calculated as:

[0099]

[0100] in, W is the eigenvector matrix; Z is the feature parameter after dimensionality reduction, the specific form is , n is the number of features of the original data, m is the number of features after dimensionality reduction;

[0101] Combine the reduced feature parameters to construct a feature data set D :

[0102] .

[0103] Optionally, as an embodiment of the present invention, in step S4, the remaining life prediction model is constructed by adopting a multi-layer perceptron neural network structure, and the specific steps include:

[0104] Obtain historical feature data and the corresponding historical remaining life and construct a data set, and divide the data set into a training set, a validation set, and a test set, wherein the historical feature data is used as an input vector, and the historical remaining life is used as an output vector. The input layer is determined according to the data in the input vector, and the output layer is determined according to the data in the output vector.

[0105] For each sample in the training set, the input vector is fed into the remaining life prediction model, and the predicted output vector of the model is calculated by forward propagation in the hidden layer;

[0106] The loss value between the predicted output vector and the actual output feature vector is calculated according to the loss function, and the gradient of the loss function to the model weights and biases is calculated using the back-propagation algorithm, and the weights and biases of the model are updated using the gradient information;

[0107] The preset number of training times per interval is used to calculate the evaluation index based on the validation set combined with the root mean square error function, and the model structure is adjusted based on the changing trend of the evaluation index;

[0108] The trained remaining life prediction model is finally evaluated based on the test set.

[0109] Optionally, as an embodiment of the present invention, the predicted output vector of the model calculated by forward propagation in the hidden layer specifically includes:

[0110] Each node in the hidden layer performs a weighted summation of the output information of the previous layer:

[0111]

[0112] Where n is the number of nodes in the previous layer, is the output information of the i-th node in the previous layer, is the weight connecting the i-th node in the previous layer and the j-th node in the current layer, is the bias of the jth node in the current layer;

[0113] The result of the weighted sum is transformed nonlinearly through the activation function to obtain the output of the node;

[0114] After multiple hidden layers are processed layer by layer, the output layer receives the output of the last hidden layer to obtain the predicted remaining lifespan.

[0115] By adopting the above technical solution, the life prediction of pot-type insulators is realized. The propagation characteristics of ultrasonic critical refraction longitudinal waves of different frequencies can more accurately capture the internal change information of insulators. Based on the waveform data of multi-frequency ultrasonic critical refraction longitudinal waves, the characteristic quantity is extracted and combined with the neural network algorithm, the life of pot-type insulators can be predicted.

[0116] Build a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform. The designed multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform is flexible and easy to disassemble to support the use of ultrasonic waves of multiple frequencies, which enhances the adaptability and practicality of the experiment.

[0117] Innovation in extracting characteristic parameters. Through data denoising, waveform and spectrum feature extraction, combined with the PCA algorithm, the health and life status of insulators can be more comprehensively understood, providing more valuable reference life prediction results.

[0118] Reduced cost and complexity: Compared with traditional monitoring methods, this method has lower equipment cost and installation complexity, and is suitable for application in a variety of situations.

[0119] In some embodiments, the pot-type insulator life prediction system may include multiple functional modules composed of computer program segments. The computer program of each program segment in the pot-type insulator life prediction system may be stored in the memory of a computer device and executed by at least one processor to perform (see Figure 1 Description) Function of life prediction of pot insulator.

[0120] In this embodiment, the pot insulator life prediction system can be divided into multiple functional modules according to the functions it performs, such as Figure 5 As shown. The functional modules of the system may include: a measurement point division module, a platform construction module, a feature extraction module, and a life prediction module. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments. The system includes:

[0121] The measuring point division module selects multiple pot-type insulators with different service times and divides multiple measuring points on each pot-type insulator;

[0122] A platform building module, which builds a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform based on a multi-channel ultrasonic wedge and multiple sets of ultrasonic probes with different frequencies;

[0123] The feature extraction module obtains the waveform data of ultrasonic critical refraction longitudinal waves of pot-type insulators with different service times at each measuring point and each channel based on the multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform, and extracts the feature data after the waveform data is reduced in dimension based on the principal component analysis method;

[0124] The life prediction module builds a remaining life prediction model based on a neural network algorithm, inputs the current pot-type insulator characteristic data into the remaining life prediction model, and outputs the predicted remaining life of the current pot-type insulator.

[0125] The measurement point division module comprehensively covers different service stages and detects insulators in multiple directions; the platform construction module uses multi-frequency ultrasound and multi-channel wedges to improve detection accuracy and adaptability; the feature extraction module obtains rich data and optimizes the processing through principal component analysis; the life prediction module uses neural network algorithms to accurately predict the life and accurately reveal the evolution process of its internal defects, thereby achieving an accurate assessment of the status of pot-type insulators as a whole, effectively ensuring the safe and stable operation of high-voltage transmission equipment and improving operation and maintenance efficiency.

[0126] Figure 6 A schematic diagram of the structure of a terminal provided in an embodiment of the present invention, wherein the terminal can be used to execute the method for predicting the life of a pot-type insulator provided in an embodiment of the present invention.

[0127] The terminal may include: a processor, a memory and a communication unit. These components communicate via one or more buses. Those skilled in the art will appreciate that the server structure shown in the figure does not limit the present invention. It may be a bus structure or a star structure, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0128] The memory can be used to store the execution instructions of the processor, and the memory can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory are executed by the processor, the terminal is enabled to execute some or all of the steps in the following method embodiments.

[0129] The processor is the control center of the storage terminal, which uses various interfaces and lines to connect various parts of the entire electronic terminal, and executes various functions and / or processes data of the electronic terminal by running or executing software programs and / or modules stored in the memory, and calling data stored in the memory. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor can only include a central processing unit (CPU). In the embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.

[0130] The communication unit is used to establish a communication channel so that the storage terminal can communicate with other terminals, receive user data sent by other terminals or send user data to other terminals.

[0131] The present invention also provides a computer-readable storage medium, wherein the computer storage medium may store a program, and when the program is executed, the program may include some or all of the steps in each embodiment provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0132] Therefore, the technical effects that can be achieved by this embodiment can be found in the description above and will not be repeated here.

[0133] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention, in essence or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes, including several instructions for enabling a computer terminal (which can be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention.

[0134] In this specification, the same or similar parts between the various embodiments can be referred to each other. In particular, for the terminal embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method embodiment.

[0135] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules 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 systems or modules, which can be electrical, mechanical or other forms.

[0136] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0137] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0138] Although the present invention has been described in detail with reference to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person of ordinary skill in the art may easily think of changes or substitutions within the technical scope disclosed by the present invention, and these shall be within the scope of protection of the present invention.

Claims

1. A method for predicting the life of a pot-type insulator, characterized in that: The following steps are involved: S1, select multiple pot-type insulators with different service times, and divide multiple measuring points on each pot-type insulator; S2, building a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform based on a multi-channel ultrasonic wedge and multiple sets of ultrasonic probes with different frequencies; S3, based on the multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform, the waveform data of ultrasonic critical refraction longitudinal waves of pot-type insulators with different service times at each measuring point and each channel are obtained, and the characteristic data after the waveform data dimension reduction is extracted based on the principal component analysis method; In step S3, the acquisition of feature data specifically includes: Determine the amplitude measurement point in the waveform data and calculate the amplitude, determine the start reference point and the end reference point in the waveform in the waveform data and calculate the propagation time; Perform Fourier transform on the waveform data to obtain the spectrum energy distribution; Construct the original characteristic parameter data set based on amplitude, propagation time and spectral energy distribution; Based on the principal component analysis method, the original feature parameter data set is processed by dimension reduction to obtain feature data; S4, constructing a remaining life prediction model based on a neural network algorithm, inputting the current basin insulator characteristic data into the remaining life prediction model, and outputting the predicted remaining life of the current basin insulator.

2. The method for predicting the life of a pot-type insulator according to claim 1, characterized in that: Multiple groups of ultrasonic probes with different frequencies include 3 ultrasonic transmitting probes and 3 ultrasonic receiving probes. The ultrasonic transmitting probes correspond to the ultrasonic receiving probes one by one. The frequencies of each group of ultrasonic probes are 0.5MHz, 2.25MHz, and 7.5MHz respectively. The ultrasonic wedge is a hexagonal wedge with three channels, which are a 0.5 MHz channel, a 2.25 MHz channel, and a 7.5 MHz channel.

3. The method for predicting the life of a pot-type insulator according to claim 1, characterized in that: Each set of ultrasonic probes includes an ultrasonic transmitting probe and an ultrasonic receiving probe. The multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform specifically includes: A signal generator configured to generate an electrical excitation signal of a preset frequency; A plurality of ultrasonic transmitting probes, each of which is configured to receive the electrical excitation signal at a preset frequency and convert the electrical excitation signal into an ultrasonic wave of a preset frequency based on an inverse piezoelectric effect; An ultrasonic wedge configured to couple ultrasonic waves of a preset frequency into the pot insulator and generate ultrasonic critical refracted longitudinal waves on the surface of the insulator; A plurality of ultrasonic receiving probes are configured to receive critical refracted longitudinal waves and convert the ultrasonic critical refracted longitudinal waves into electrical signals based on the piezoelectric effect; an oscilloscope configured to receive the electrical signal to display the waveform of the ultrasonic critical refracted longitudinal wave; The PC is configured to collect waveform data displayed by the oscilloscope.

4. The method for predicting the life of a pot-type insulator according to claim 1, characterized in that: In step S4, the remaining life prediction model is constructed by using a multi-layer perceptron neural network structure, and the specific steps include: Obtain historical feature data and the corresponding historical remaining life and construct a data set, and divide the data set into a training set, a validation set, and a test set, wherein the historical feature data is used as an input vector, and the historical remaining life is used as an output vector. The input layer is determined according to the data in the input vector, and the output layer is determined according to the data in the output vector. For each sample in the training set, the input vector is fed into the remaining life prediction model, and the predicted output vector of the model is calculated by forward propagation in the hidden layer; The loss value between the predicted output vector and the actual output feature vector is calculated according to the loss function, and the gradient of the loss function to the model weights and biases is calculated using the back-propagation algorithm, and the weights and biases of the model are updated using the gradient information; The preset number of training times per interval is used to calculate the evaluation index based on the validation set combined with the root mean square error function, and the model structure is adjusted based on the changing trend of the evaluation index; The trained remaining life prediction model is finally evaluated based on the test set.

5. The method for predicting the life of a pot-type insulator according to claim 4, characterized in that: The predicted output vector of the model calculated by forward propagation in the hidden layer specifically includes: Each node in the hidden layer performs a weighted summation of the output information of the previous layer: Where n is the number of nodes in the previous layer, is the output information of the i-th node in the previous layer, is the weight connecting the i-th node in the previous layer and the j-th node in the current layer, is the bias of the jth node in the current layer; The result of the weighted sum is transformed nonlinearly through the activation function to obtain the output of the node; After multiple hidden layers are processed layer by layer, the output layer receives the output of the last hidden layer to obtain the predicted remaining lifespan.

6. The method for predicting the life of a pot-type insulator according to claim 1, characterized in that: Step S1 specifically includes: dividing the area below the convex side insert of the pot insulator into a plurality of sub-areas, wherein the sub-areas are distributed in a ring shape around the center of the pot insulator, and selecting a plurality of measuring points on the circumference corresponding to each sub-area.

7. A method for predicting the life of a pot-type insulator, characterized in that: When the system is implemented, the method for predicting the life of a pot-type insulator according to any one of claims 1 to 6 is executed, and the system comprises: The measuring point division module selects multiple pot-type insulators with different service times and divides multiple measuring points on each pot-type insulator; A platform building module, which builds a multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform based on a multi-channel ultrasonic wedge and multiple sets of ultrasonic probes with different frequencies; The feature extraction module obtains the waveform data of ultrasonic critical refraction longitudinal waves of pot-type insulators with different service times at each measuring point and each channel based on the multi-frequency ultrasonic critical refraction longitudinal wave transceiver platform, and extracts the feature data after the waveform data is reduced in dimension based on the principal component analysis method; The life prediction module builds a remaining life prediction model based on a neural network algorithm, inputs the current pot-type insulator characteristic data into the remaining life prediction model, and outputs the predicted remaining life of the current pot-type insulator.

8. A terminal, characterized in that: include: A memory, used for storing a life prediction program for a pot-type insulator; A processor is used to implement the steps of the pot-type insulator life prediction method as described in any one of claims 1 to 6 when executing the pot-type insulator life prediction program.

9. A computer-readable storage medium, characterized in that: The readable storage medium stores a pot-type insulator life prediction program, and when the pot-type insulator life prediction program is executed by a processor, the steps of the pot-type insulator life prediction method according to any one of claims 1 to 6 are implemented.

Images