Area recognition method based on fingerprint characteristics of wireless channels of converter station inspection robots

Abstract

The present application relates to a method, device, computer equipment, storage medium and computer program product for regional identification based on the fingerprint characteristics of the wireless channel of the inspection robot in the converter station. The method comprises: obtaining the transmission signal of the inspection robot received by the signal receiving end of the converter station; extracting multiple fingerprint feature information corresponding to the wireless channel of the inspection robot according to the transmission signal of the inspection robot; generating a fingerprint feature vector corresponding to the current area of ​​the inspection robot according to the multiple fingerprint feature information; obtaining the fingerprint feature vectors corresponding to each inspection area in the converter station, and comparing the corresponding fingerprint feature vectors of each inspection area with the fingerprint feature vectors corresponding to the current area, and obtaining the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional position information of the current area in the converter station. The present method can accurately identify the inspection area where the inspection robot is currently located.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, computer equipment, storage medium and computer program product for region identification based on fingerprint features of wireless channels of converter station inspection robots. Background Art

[0002] The inspection task of converter stations is the key to ensure the normal operation of the power grid. With the development of computer technology, various inspection methods for converter stations have emerged.

[0003] At present, inspection robots are often used to inspect converter stations, replacing the traditional manual inspection mode. However, due to the complex signal interference in the converter station, when four inspection robots inspect the converter station, they often cannot accurately monitor the current position of the inspection robots, which is inconvenient for inspection management of the inspection robots.

[0004] Therefore, there is a problem in the traditional technology that the inspection area of ​​the inspection robot is not accurately identified. Summary of the invention

[0005] Based on this, it is necessary to provide a region identification method, device, computer equipment, computer-readable storage medium and computer program product based on the wireless channel fingerprint characteristics of the converter station inspection robot, which can more accurately identify the current location of the inspection robot in response to the above technical problems.

[0006] A method for region recognition based on wireless channel fingerprint features of a converter station inspection robot, characterized in that the method comprises:

[0007] Acquire a patrol robot transmission signal received by a signal receiving end of a converter station; the patrol robot transmission signal is a wireless signal transmitted by the patrol robot in the converter station;

[0008] According to the signal transmitted by the inspection robot, multiple fingerprint feature information corresponding to the wireless channel of the inspection robot is extracted; each fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot;

[0009] Generate a fingerprint feature vector corresponding to the current area of ​​the inspection robot based on multiple fingerprint feature information;

[0010] The fingerprint feature vectors corresponding to each inspection area in the converter station are obtained, and the fingerprint feature vectors corresponding to each inspection area are compared with the fingerprint feature vectors corresponding to the current area to obtain the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional position information of the current area in the converter station.

[0011] In one embodiment, based on the signal transmitted by the inspection robot, multiple fingerprint feature information corresponding to the wireless channel of the inspection robot is extracted, including:

[0012] The transmission signal of the inspection robot is input into the wireless channel feature extraction model to obtain the signal feature data corresponding to each channel sample point in the wireless channel of the inspection robot;

[0013] According to the signal characteristic data corresponding to each channel sample point, a signal characteristic data sequence is generated, and fingerprint feature extraction is performed on the signal characteristic data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the inspection robot.

[0014] In one embodiment, the transmission signal of the inspection robot is input into the wireless channel feature extraction model to obtain the signal feature data corresponding to each channel sample point in the wireless channel of the inspection robot, including:

[0015] The transmission signal of the inspection robot is input into the ideal signal feature extraction sub-model to obtain the ideal signal feature data corresponding to each channel sample point in the wireless channel of the inspection robot;

[0016] The ideal signal feature data is input into the actual signal feature extraction sub-model to obtain the signal feature data actually corresponding to each channel sample point in the wireless channel of the inspection robot.

[0017] In one embodiment, a signal feature data sequence is generated according to the signal feature data corresponding to each channel sample point, and fingerprint feature extraction is performed on the signal feature data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the inspection robot, including:

[0018] Generate a signal characteristic data sequence corresponding to the wireless channel of the inspection robot according to the signal characteristic data corresponding to each channel sample point;

[0019] Performing noise reduction processing on the signal characteristic data sequence to obtain a noise-reduced signal characteristic data sequence;

[0020] The fingerprint feature is extracted from the signal feature data sequence after noise reduction to obtain multiple fingerprint feature information corresponding to the wireless channel of the inspection robot.

[0021] In one embodiment, performing noise reduction processing on the signal characteristic data sequence to obtain the signal characteristic data sequence after noise reduction includes:

[0022] Generate a signal feature matrix corresponding to the wireless channel of the inspection robot according to the signal feature data sequence;

[0023] Performing pre-noise reduction processing on the signal feature matrix to obtain a signal feature matrix after pre-noise reduction;

[0024] Extracting elements from the signal feature matrix after pre-noise reduction to generate a signal feature data sequence after noise reduction;

[0025] The signal feature data sequence after pre-denoising is input into the wavelet denoising model to obtain the signal feature data sequence after denoising.

[0026] In one embodiment, a fingerprint feature vector corresponding to each inspection area in the converter station is obtained, and the fingerprint feature vector corresponding to each inspection area is compared with the fingerprint feature vector corresponding to the current area to obtain the area similarity information corresponding to each inspection area, including:

[0027] Obtain the fingerprint feature vectors corresponding to each inspection area in the converter station, compare the fingerprint feature vectors corresponding to each inspection area with the fingerprint feature vectors corresponding to the current area, and obtain the fingerprint feature similarity vectors corresponding to each inspection area;

[0028] According to the weight information corresponding to each fingerprint feature information, a fingerprint feature weight vector is generated, and the regional similarity information corresponding to each inspection area is determined according to the fingerprint feature similarity vector and the fingerprint feature weight vector corresponding to each inspection area; the regional similarity information corresponding to each inspection area represents the regional similarity between each inspection area and the current area.

[0029] A region identification device based on the wireless channel fingerprint feature of a converter station inspection robot, characterized in that the device comprises:

[0030] An acquisition module is used to acquire a signal transmitted by the inspection robot received by a signal receiving end of the converter station; the signal transmitted by the inspection robot is a wireless signal transmitted by the inspection robot in the converter station;

[0031] The extraction module is used to extract multiple fingerprint feature information corresponding to the wireless channel of the inspection robot according to the signal transmitted by the inspection robot; each fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot;

[0032] A generation module is used to generate a fingerprint feature vector corresponding to the current area of ​​the inspection robot based on multiple fingerprint feature information;

[0033] The determination module is used to obtain the fingerprint feature vectors corresponding to each inspection area in the converter station, and compare the corresponding fingerprint feature vectors of each inspection area with the fingerprint feature vectors corresponding to the current area to obtain the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional position information of the current area in the converter station.

[0034] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and is characterized in that the processor implements the steps of the above method when executing the computer program.

[0035] A computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the above method when executed by a processor.

[0036] A computer program product comprises a computer program, wherein the computer program implements the steps of the above method when executed by a processor.

[0037] The above-mentioned area recognition method, device, computer equipment, storage medium and computer program product based on the fingerprint characteristics of the wireless channel of the inspection robot in the converter station obtains the inspection robot transmission signal received by the signal receiving end of the converter station; the inspection robot transmission signal is a wireless signal transmitted by the inspection robot in the converter station; then, based on the inspection robot transmission signal, multiple fingerprint feature information corresponding to the wireless channel of the inspection robot is extracted; each fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot; then, based on the multiple fingerprint feature information, a fingerprint feature vector corresponding to the current area of ​​the inspection robot is generated; then, the fingerprint feature vectors corresponding to the inspection area in the converter station are obtained. The corresponding fingerprint feature vector is obtained, and the corresponding fingerprint feature vector of each inspection area is compared with the fingerprint feature vector corresponding to the current area, so as to obtain the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional position information of the current area in the converter station; in this way, the current regional channel feature corresponding to the inspection task of the inspection robot is extracted, and according to the similarity between the current regional channel feature corresponding to the inspection robot and the regional channel feature corresponding to each inspection area, the position of the inspection robot in the converter station can be accurately determined, and the inspection management of the inspection robot in the converter station can be more efficient, which is conducive to maintaining the safe operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A diagram of an application environment of a wireless channel fingerprint feature area recognition method in one embodiment;

[0039] Figure 2 A schematic diagram of a flow chart of a method for identifying a wireless channel fingerprint feature area in an embodiment;

[0040] Figure 3 A schematic diagram of a transmission process of a signal transmitted by an inspection robot in one embodiment;

[0041] Figure 4 is an analysis flow chart of a wireless channel fingerprint feature analysis method in one embodiment;

[0042] Figure 5 A flow chart of a wavelet denoising method based on SVD singular value decomposition in one embodiment;

[0043] Figure 6 A flow chart of a wavelet denoising method based on a fixed threshold in one embodiment;

[0044] Figure 7 A schematic flow chart of a method for identifying a wireless channel fingerprint feature area in another embodiment;

[0045] Figure 8 It is a structural block diagram of a wireless channel fingerprint feature area recognition device in one embodiment;

[0046] Fig. 9 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] The area recognition method based on the wireless channel fingerprint feature of the converter station inspection robot provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the inspection robot 102 communicates with the server 104 through the network. The server 104 obtains the inspection robot transmission signal received by the signal receiving end of the converter station; the inspection robot transmission signal is a wireless signal transmitted by the inspection robot in the converter station; the server 104 extracts multiple fingerprint feature information corresponding to the wireless channel of the inspection robot according to the inspection robot transmission signal; each fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot; the server 104 generates a fingerprint feature vector corresponding to the current area of ​​the inspection robot according to the multiple fingerprint feature information; the server 104 obtains the fingerprint feature vector corresponding to each inspection area in the converter station, and compares the corresponding fingerprint feature vector of each inspection area with the fingerprint feature vector corresponding to the current area, and obtains the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional location information of the current area in the converter station. Among them, the server 104 can be implemented with an independent server or a server cluster composed of multiple servers.

[0049] In one embodiment, Figure 2 As shown in the figure, a region recognition method based on the wireless channel fingerprint characteristics of the converter station inspection robot is provided. Figure 1 Taking the server 104 in the example as an example, the following steps are included:

[0050] Step S202, obtaining a patrol robot transmission signal received by a signal receiving end of the converter station; the patrol robot transmission signal is a wireless signal transmitted by the patrol robot in the converter station.

[0051] In a specific implementation, the server receives a signal transmitted by the inspection robot through a signal receiving end of the converter station.

[0052] Among them, the transmission process of the inspection robot transmitting the signal is as follows: Figure 3 As shown, first, the central control platform of the inspection robot transmits a radio wave signal, and filters the radio wave signal through a radio frequency filter. Then, the filtered radio wave signal is transmitted to the communication wireless channel, wherein the radio wave signal is affected by the strong electromagnetic noise signal caused by multipath transmission differences, external environmental noise interference and multipath delay in the communication wireless channel. Finally, the radio wave signal affected by the strong electromagnetic noise signal is received by the wireless switch, and the wireless switch processes the radio wave signal to obtain a signal processing result, and sends the signal processing result to the client.

[0053] Step S204: extract multiple fingerprint feature information corresponding to the wireless channel of the inspection robot according to the signal transmitted by the inspection robot; each fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot.

[0054] Among them, the signal fading characteristics can be parameter characteristics obtained by processing the signal data based on methods such as Doppler spread, delay spread, path loss, time domain peak analysis, and frequency domain peak analysis.

[0055] In the specific implementation, the server processes the signal transmitted by the inspection robot according to the signal transmitted by the inspection robot to obtain various fingerprint feature information corresponding to the wireless channel of the inspection robot, such as parameter characteristics obtained by data processing the signal based on methods such as Doppler expansion, delay expansion, path loss, time domain peak analysis and frequency domain peak analysis.

[0056] Step S206: Generate a fingerprint feature vector corresponding to the area where the inspection robot is currently located based on the multiple fingerprint feature information.

[0057] In a specific implementation, the server generates a fingerprint feature vector corresponding to the wireless channel of the area where the inspection robot is currently located based on each fingerprint feature information.

[0058] Step S208, obtain the fingerprint feature vectors corresponding to each inspection area in the converter station, compare the fingerprint feature vectors corresponding to each inspection area with the fingerprint feature vectors corresponding to the current area, and obtain the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional position information of the current area in the converter station.

[0059] The area similarity information may be feature similarity information representing the wireless channel fading characteristics of each inspection area and the current area.

[0060] In the specific implementation, the server obtains the fingerprint feature vectors corresponding to each inspection area in the converter station, and then compares the fingerprint feature vectors corresponding to each inspection area with the fingerprint feature vectors corresponding to the current area of ​​the inspection robot, and obtains the wireless channel similarity information between each inspection area and the current area of ​​the inspection robot as the regional similarity information corresponding to each inspection area. The server compares the similarity corresponding to the regional similarity information, and takes the inspection area with the largest regional similarity as the current area of ​​the inspection robot, thereby completing the positioning and identification of the inspection robot.

[0061] To facilitate the understanding of those skilled in the art, Figure 4 An analysis flow chart of a method for analyzing the wireless channel fingerprint characteristics of a quadruped robot in a converter station is provided as an example. First, a model is established based on the wireless channel, and a wireless channel time-varying model is established by considering the dynamic changes of the quadruped robot. The signal transmitted by the quadruped robot is converted into signal feature data through the wireless channel time-varying model to obtain a signal feature data matrix. Then, the signal feature data matrix is ​​pre-denoised based on the principle of SVD singular value decomposition (a data dimension reduction algorithm for machine learning). Further, the signal feature data after pre-denoising is subjected to secondary denoising by a wavelet denoising method with a fixed threshold. Finally, the basic parameters of the signal feature data are processed to calculate the wireless channel fingerprint feature data, and the inspection area of ​​the quadruped robot is identified according to the wireless channel fingerprint feature data.

[0062] The above-mentioned area recognition method, device, computer equipment, storage medium and computer program product based on the fingerprint characteristics of the wireless channel of the inspection robot in the converter station obtains the inspection robot transmission signal received by the signal receiving end of the converter station; the inspection robot transmission signal is a wireless signal transmitted by the inspection robot in the converter station; then, based on the inspection robot transmission signal, multiple fingerprint feature information corresponding to the wireless channel of the inspection robot is extracted; each fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot; then, based on the multiple fingerprint feature information, a fingerprint feature vector corresponding to the current area of ​​the inspection robot is generated; then, the fingerprint feature vectors corresponding to the inspection area in the converter station are obtained. The corresponding fingerprint feature vector is obtained, and the corresponding fingerprint feature vector of each inspection area is compared with the fingerprint feature vector corresponding to the current area, so as to obtain the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional position information of the current area in the converter station; in this way, the current regional channel feature corresponding to the inspection task of the inspection robot is extracted, and according to the similarity between the current regional channel feature corresponding to the inspection robot and the regional channel feature corresponding to each inspection area, the position of the inspection robot in the converter station can be accurately determined, and the inspection management of the inspection robot in the converter station can be more efficient, which is conducive to maintaining the safe operation of the power grid.

[0063] In another embodiment, based on the signal transmitted by the patrol robot, multiple fingerprint feature information corresponding to the wireless channel of the patrol robot is extracted, including: inputting the signal transmitted by the patrol robot into a wireless channel feature extraction model to obtain signal feature data corresponding to each channel sample point in the wireless channel of the patrol robot; generating a signal feature data sequence based on the signal feature data corresponding to each channel sample point, and performing fingerprint feature extraction on the signal feature data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the patrol robot.

[0064] The wireless channel feature extraction model may be a model for determining signal feature data in a wireless channel.

[0065] The signal characteristic data may be a signal value in a wireless channel.

[0066] In the specific implementation, the server inputs the signal transmitted by the inspection robot into the wireless channel feature extraction model to obtain the signal value corresponding to each channel sample point in the wireless channel of the inspection robot. The server generates a signal feature data sequence based on the signal value corresponding to each channel sample point in the wireless channel of the inspection robot. The server performs fingerprint feature extraction on the signal feature data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the inspection robot, such as Doppler expansion parameter information, delay expansion parameter information, path loss parameter information, time domain peak analysis parameter information, and frequency domain peak analysis parameter information.

[0067] The technical solution of this embodiment is to input the transmission signal of the inspection robot into the wireless channel characteristic model, obtain the signal characteristic data corresponding to each channel sample point in the wireless channel, and generate a signal characteristic data sequence, and then extract the fingerprint features of the signal characteristic data sequence according to the channel fading characteristics, so as to convert the complex and changeable wireless signal into visual data, which is beneficial to the positioning and identification of the inspection robot.

[0068] In another embodiment, the transmission signal of the patrol robot is input into a wireless channel feature extraction model to obtain signal feature data corresponding to each channel sample point in the wireless channel of the patrol robot, including: inputting the transmission signal of the patrol robot into an ideal signal feature extraction sub-model to obtain ideal signal feature data corresponding to each channel sample point in the wireless channel of the patrol robot; inputting the ideal signal feature data into an actual signal feature extraction sub-model to obtain signal feature data actually corresponding to each channel sample point in the wireless channel of the patrol robot.

[0069] The ideal signal feature extraction sub-model may be a model for obtaining an ideal signal value in a wireless channel.

[0070] The actual signal feature extraction sub-model may be a model for converting an ideal signal value in a wireless channel into an actual signal value.

[0071] In the specific implementation, the server inputs the inspection robot's transmission signal into the ideal signal feature extraction sub-model to obtain the ideal signal feature data corresponding to each channel sample point in the inspection robot's wireless channel. The server inputs the ideal signal feature data into the actual signal feature extraction sub-model to obtain the signal feature data actually corresponding to each channel sample point in the inspection robot's wireless channel.

[0072] To facilitate the understanding of those skilled in the art, the following exemplifies a wireless channel modeling method for determining the signal characteristic data of a wireless channel. This method is based on the theory of the flat fading channel model. Due to the multipath transmission differences of the transmitted signal, the transmitted signal has different time delays and different channel coefficients in different transmission paths, wherein the time delay causes the signal to be delayed during the transmission process, and the channel coefficient causes the signal to be attenuated during the transmission process. Since the transmission performance of the signal is the superposition of the signal in all transmission paths, the method first establishes an ideal wireless channel characteristic model, then establishes a basic wireless channel characteristic model, further, constructs the test transmission signal to be selected for the wireless channel characteristic model, and finally introduces the time parameter t to establish a wireless channel time-varying model. The specific steps of this method are as follows:

[0073] Step 1: Establish an ideal channel model:

[0074]

[0075] Since the channel samples are distributed in the channel at a certain time interval, let the sampling time interval be Δκ, then the sampling time interval Δκ is the time T corresponding to the total length of the signal. N Divided by the total number of samples N, the channel samples are sampled from time 0, so the channel sample n refers to the code of the signal value collected at time (n-1)Δκ. In the above formula, L is the total number of signal transmission paths at the current moment; N is the total number of sample point identifiers, that is, the signal length; γ l The channel coefficient of the lth signal transmission path at the current moment reflects the attenuation of the signal transmitted on this path, so its value range is (0,1); τ l is the delay of the lth path at the current moment, τ l Converted to an integer multiple of the sample point, that is, the signal is delayed by τ at this moment l Sample points, τ l The unit is an integer multiple of the number of samples; X(n) is the measurement result of sample point n in an ideal channel. An ideal channel is a propagation channel without the influence of a noise signal. It is an ideal propagation condition and can characterize the ideal transmission characteristics of a wireless channel.

[0076] Step 2: Introduce external noise influence into the ideal channel model and establish the basic wireless channel characteristic model:

[0077]

[0078] Among them, Y(n) is the measurement result of the signal value at sample point n considering the influence of external noise. It is the real signal measurement result and represents the real transmission characteristics of the wireless channel. is the filter length, ξ(μ) is the filter coefficient when the RF filter length corresponds to μ, and is equivalent to the influence of all RF filters in the wireless channel. Since the filter used is known, the filter parameters are all known quantities, and φ(n) is the influence of the electromagnetic noise signal introduced by the sample point n at that moment.

[0079] Step 3: Construct a test transmission signal of the basic wireless channel characteristic model in step 2. Assume that the test transmission signal is a unit step pulse signal:

[0080]

[0081] Step 4: Establish a time-varying model of the wireless channel. Based on the basic wireless channel characteristic model, the time parameter t is introduced. At the same time, according to the actual progressive area division of the two-dimensional inspection plane of the inspection robot, the coordinates (i, j) are used to describe the inspection area where the inspection robot is located in real time to reflect the dynamic changes of the inspection robot's position. Based on the above data and the established basic wireless channel characteristic model, a wireless channel time-varying model that considers the dynamic changes of the quadruped robot is established:

[0082]

[0083]

[0084] Among them, T is the test time corresponding to the sample identification point, is the channel coefficient of the signal in the lth signal transmission path at the regional coordinate (i, j) at time t; is the delay of the signal on the lth path at the regional coordinate (i, j) at time t; φ i,j (n, t) is the noise signal size at sample point n at the region coordinate (i, j) at time t; X i,j (n, t) is the ideal channel signal value at the region coordinate (i, j) at time t, processed at sample point n; Y i,j (n, t) is the actual measured value of the signal at sample point n at the region coordinate (i, j) at time t.

[0085] The technical solution of this embodiment is to input the transmission signal of the inspection robot into the ideal signal feature extraction sub-model, and consider the influence of signal noise, and input the ideal signal feature data into the actual signal feature extraction sub-model to obtain the signal feature data actually corresponding to each channel sample point in the wireless channel of the inspection robot. When fingerprint feature extraction is performed on the signal feature data sequence, more accurate fingerprint feature information can be obtained, which is conducive to more accurate positioning and identification of the inspection robot.

[0086] In another embodiment, a signal characteristic data sequence is generated according to the signal characteristic data corresponding to each channel sample point, and fingerprint feature extraction is performed on the signal characteristic data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the patrol robot, including: generating a signal characteristic data sequence corresponding to the wireless channel of the patrol robot according to the signal characteristic data corresponding to each channel sample point; performing noise reduction processing on the signal characteristic data sequence to obtain a noise-reduced signal characteristic data sequence; performing fingerprint feature extraction on the noise-reduced signal characteristic data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the patrol robot.

[0087] In the specific implementation, the server generates a signal characteristic data sequence corresponding to the wireless channel of the inspection robot based on the signal characteristic data corresponding to each channel sample point of the wireless channel of the inspection robot. The server performs noise reduction processing on the signal characteristic data sequence to obtain the signal characteristic data sequence after noise reduction. The server then performs fingerprint feature extraction on the signal characteristic data sequence after noise reduction to obtain multiple fingerprint feature information corresponding to the wireless channel of the inspection robot, such as Doppler spread parameter information, delay spread parameter information, path loss parameter information, time domain peak analysis parameter information and frequency domain peak analysis parameter information.

[0088] To facilitate the understanding of those skilled in the art, the following exemplifies a fingerprint feature extraction method for the wireless channel of a converter station inspection robot, which is based on the fading characteristics of the wireless channel. The basic parameters of the signal data are first preprocessed, and then the fingerprint feature parameters are calculated based on the basic parameters of the signal data.

[0089] In the process of preprocessing the basic parameters of the signal data, the signal characteristic data Y′ after noise reduction is firstly i,j (n,t) is processed, because Y′ i,j (n, t) is a complex variable signal, so for Y′ i,j The instantaneous amplitude and instantaneous power of (n, t) are calculated according to the following formula:

[0090]

[0091] Calculate Y′ i,j The instantaneous amplitude Ap of (n,t) i,j (n,t), where real(Y′ i,j (n,t)) is the complex signal Y′ i,j The real part of (n,t), imag(Y′ i,j (n,t)) is the complex signal Y′ i,j The imaginary part of (n,t). According to the following formula:

[0092] P i,j (n,t)=[Ap i,j (n,t)] 2

[0093] Calculate Y′ i,j The instantaneous power P of (n,t) i,j (n,t). According to the following formula

[0094] Ap i,j (n,t)→Fourier transform→Af i,j (n,t)

[0095] Y′ i,j The instantaneous amplitude Ap of (n,t) i,j (n, t) is Fourier transformed to obtain the frequency domain analysis data Af i,j (n,t).

[0096] Among them, in the process of calculating the fingerprint characteristic parameters according to the basic parameters of the signal data, it is necessary to calculate the Doppler spread, the delay spread, the path loss and the signal peak analysis.

[0097] Among them, when calculating Doppler spread, it is necessary to determine the maximum frequency shift f m , where f m According to the formula:

[0098]

[0099] Calculated. Where v is the transmission speed of the signal in the channel, which is obtained by dividing the signal transmission distance by the signal transmission time (the maximum value of the instantaneous amplitude of the signal from the signal transmission to the signal reception is approximated as the signal transmission time), c is the propagation speed of electromagnetic waves under ideal conditions, and f c is the transmission frequency of the signal, which can be obtained from the information of the transmitted signal.

[0100] The calculation of Doppler spread is based on the typical Doppler power spectrum formula:

[0101]

[0102] Calculate using the Doppler power spectrum f = f c The Doppler power value at As a characteristic parameter of Doppler spread.

[0103] The calculation of delay spread is based on the calculation method of global average additional delay and global rms delay spread. The specific calculation formula is as follows:

[0104]

[0105]

[0106]

[0107]

[0108] in, is the instantaneous power of the step signal of the input filter at the signal identification point n in the ideal state, Δt is the measurement interval of adjacent samples in the signal sequence, is the signal peak value r t ' op The corresponding delay at [n,t] is is the global additional delay, is the global rms delay spread. is the expected value of the quadratic global additional delay, is the characteristic parameter of delay spread.

[0109] Among them, when calculating the path loss, according to the formula

[0110]

[0111] Calculate the path loss of the inspection robot communication signal in the area. r is the transmitted signal power, is the calculated path loss of the signal at time t. By calculating the path loss at different times, the path loss set of the communication signal of the patrol robot in the area is obtained:

[0112]

[0113] Using the linear least squares method to fit the above set of data, we can get the slope of the fitting line. It is the characteristic parameter of path loss.

[0114] Among them, when performing signal peak analysis, for the noise reduction signal data Y′ i,j , extract the number of signal amplitude peaks in different amplitude ranges from the perspective of time domain analysis and frequency domain analysis Where num1 is the number of peak intervals in time domain analysis, and num2 is the number of peak intervals in frequency domain analysis. i,j (n,t),Af i,j The peak numbers of (n,t) data are counted according to different amplitude intervals.

[0115] In this way, the fingerprint feature information of the wireless channel of the inspection robot in the converter station is determined, including Doppler spread parameter information, delay spread parameter information, path loss parameter information, time domain peak analysis parameter information and frequency domain peak analysis parameter information.

[0116] The technical solution of this embodiment generates a signal feature data sequence from the signal feature data corresponding to each channel sample point in the wireless channel, performs noise reduction on the signal feature data sequence, and then performs fingerprint feature extraction on the signal feature data sequence after noise reduction, thereby converting complex signal data into feature data with practical meaning, thereby realizing feature analysis of the wireless channel, which is beneficial to feature identification of the wireless channel of the inspection robot.

[0117] In another embodiment, a signal feature data sequence is subjected to denoising processing to obtain a signal feature data sequence after denoising, including: generating a signal feature matrix corresponding to a wireless channel of the inspection robot according to the signal feature data sequence; performing pre-denoising processing on the signal feature matrix to obtain a signal feature matrix after pre-denoising; extracting elements in the signal feature matrix after pre-denoising to generate a signal feature data sequence after denoising; and inputting the signal feature data sequence after pre-denoising into a wavelet denoising model to obtain a signal feature data sequence after denoising.

[0118] Among them, the wavelet denoising model can be a model used to perform denoising processing on signal feature data.

[0119] In a specific implementation, the server generates a signal feature matrix corresponding to the wireless channel of the inspection robot based on the signal feature data sequence. The signal feature matrix is ​​a signal matrix containing signal noise information. The server performs pre-noise reduction processing on the signal feature matrix to obtain a pre-noise reduced signal feature matrix. The server extracts elements from the pre-noise reduced signal feature matrix to generate a denoised signal feature data sequence. The server performs wavelet denoising processing on the denoised signal feature data sequence to obtain a denoised signal feature data sequence.

[0120] In order to facilitate the understanding of those skilled in the art, Figure 5 An exemplary flow chart of a wavelet denoising method based on SVD singular value decomposition is provided. The method constructs a noisy matrix according to the original signal data, and performs SVD decomposition on the noisy matrix. After the SVD decomposition is completed, a pre-denoising process is performed, and then the pre-denoising data is subjected to secondary denoising according to a wavelet denoising method based on a fixed threshold, thereby outputting denoised signal data. The specific steps corresponding to the method are as follows:

[0121] Step 1: Construct a noisy matrix based on the signal length. The noisy signal sequence measured by the signal receiving end at a certain time in a certain area is recorded as Y. According to the data in Y, ε is 0.5 times N and is a positive integer, and it is constructed as an ε×(N-ε+1) matrix. The constructed noisy matrix is:

[0122]

[0123] Step 2: Perform SVD decomposition on the constructed noisy matrix R. According to the principle of SVD singular value decomposition, the singular value decomposition expression and related formulas of the noisy matrix R can be obtained as follows:

[0124]

[0125] Among them, ∑ is an N×(N-ε+1) diagonal matrix composed of non-negative diagonal elements arranged in descending order, and U and V are N×N and (N-ε+1)×(N-ε+1) orthogonal matrices respectively. The matrices U, V, and ∑ can be obtained by singular value decomposition.

[0126] Step 3: Pre-denoising after SVD decomposition. Select the appropriate main singular value S va , assign zeros to the matrix ∑ to obtain the matrix ∑′, and then according to the formula R′=U·∑′·V T , the preprocessing matrix R′ can be obtained, and then the corresponding elements in the preprocessing matrix R′ can be extracted to obtain the pre-denoising signal sequence Y′.

[0127] Step 4: Perform secondary denoising based on a wavelet denoising method with a fixed threshold. The wavelet denoising method with a fixed threshold is used for the pre-denoised signal sequence Y′. The specific process of the wavelet denoising method with a fixed threshold is as follows: Figure 6 The specific process of wavelet denoising is as follows: first, the original signal is decomposed by wavelet by selecting the appropriate wavelet type and decomposition layer number to obtain the coefficients of each layer; further, the threshold is set to perform threshold processing on the coefficients of each layer. The threshold setting method here is selected according to The purpose of setting the threshold is to remove the strong noise signal in the signal. Since the SVD singular value can only perform a certain noise reduction effect on the weak but persistent Gaussian signal in the environment, it is not obvious for the strong noise signal in the environment. Therefore, it is necessary to perform secondary noise reduction processing through wavelet denoising. The coefficients of each layer are denoised according to the denoising method based on the fixed threshold. Finally, the processed coefficients of each layer (that is, the quantized coefficients) are reconstructed by wavelet to obtain the denoised signal Y′ i,j (n,t).

[0128] The technical solution of this embodiment performs multiple noise reduction processes on the signal feature data corresponding to the wireless channel of the inspection robot to obtain a noise-reduced signal feature data sequence, thereby reducing the influence of strong electromagnetic interference in the inspection environment of the converter station and the Gaussian noise signal originally existing in the environment on the wireless signal in the wireless channel of the inspection robot, making the acquired fingerprint feature information more accurate, which is conducive to the positioning and identification of the inspection robot.

[0129] In another embodiment, a fingerprint feature vector corresponding to each inspection area in a converter station is obtained, and the fingerprint feature vector corresponding to each inspection area is compared with the fingerprint feature vector corresponding to the current area to obtain regional similarity information corresponding to each inspection area, including: obtaining a fingerprint feature vector corresponding to each inspection area in a converter station, and comparing the fingerprint feature vector corresponding to each inspection area with the fingerprint feature vector corresponding to the current area to obtain a fingerprint feature similarity vector corresponding to each inspection area; generating a fingerprint feature weight vector according to weight information corresponding to each fingerprint feature information, and determining the regional similarity information corresponding to each inspection area according to the fingerprint feature similarity vector and the fingerprint feature weight vector corresponding to each inspection area; the regional similarity information corresponding to each inspection area represents the regional similarity between each inspection area and the current area.

[0130] The elements in the fingerprint feature similarity vector may be the similarities between the fingerprint feature data representing the wireless channels corresponding to the inspection area.

[0131] The elements in the fingerprint feature weight vector may be the importance corresponding to each fingerprint feature representing the wireless channel.

[0132] In the specific implementation, the server obtains the fingerprint feature vectors corresponding to the wireless channels of each inspection area in the converter station, and compares the fingerprint feature vectors corresponding to the wireless channels of each inspection area with the fingerprint feature vectors corresponding to the area where the inspection robot is currently located, and obtains the fingerprint feature similarity vectors of the wireless channels of each inspection area and the wireless channels of the area where the inspection robot is currently located. The server generates a fingerprint feature weight vector based on the weight information corresponding to each fingerprint feature information, and distributes the fingerprint feature similarity vectors and fingerprint feature weight vectors corresponding to each inspection area to obtain the regional similarity between each inspection area and the area where the inspection robot is currently located. The server determines the inspection area with the highest regional similarity as the specific position of the inspection robot in the converter station based on the regional similarity between each inspection area and the area where the inspection robot is currently located.

[0133] In order to facilitate the understanding of those skilled in the art, the following exemplary provides a channel area recognition model based on weight allocation. When the positioning of the inspection robot is unknown, the inspection area is identified by extracting the real-time signal of the inspection robot in combination with the fingerprint characteristics of the existing regional channels. The similarity of the characteristic parameters of Doppler spread, delay spread, path loss, time domain peak analysis and frequency domain peak analysis is calculated according to the following formula:

[0134]

[0135]

[0136]

[0137]

[0138]

[0139] Calculate, where * and (i, j) represent the coordinates of the inspection area to be identified and the coordinates of the inspection area with known channel characteristics, respectively. Input the wireless channel characteristics of the area to be identified into the above five formulas, and you can get five similarity indicators, which are recorded as similarity vectors

[0140]

[0141] Further, according to the above method, the similarity vector between the required identification area and all known channel feature areas is calculated, and then a weight distribution vector Ω = [ω1, ω2, ω3, ω4, ω5] is set according to the following formula:

[0142]

[0143] Calculate the comprehensive similarity. Among them, q vg is the channel feature similarity between the vth inspection area to be identified and the gth inspection area with known channel features, and map is the coordinate set of the inspection area with known channel features. Based on the above calculation, a comprehensive similarity analysis network Q can be established, and q (i,j) ∈Q, the coordinates corresponding to the maximum comprehensive similarity in the network are the coordinates of the inspection area.

[0144] The technical solution of this embodiment calculates the fingerprint feature similarity vector corresponding to each inspection area based on the fingerprint feature vector of each inspection area and the area where the inspection robot is currently located, and then determines the fingerprint feature similarity information corresponding to each inspection area based on the weight information corresponding to each fingerprint feature in the fingerprint feature vector, that is, the regional similarity information corresponding to each inspection area. It comprehensively considers the importance of each fingerprint feature information, and can obtain more accurate regional similarity information, which is conducive to the positioning and identification of the inspection robot.

[0145] In another embodiment, Figure 7 As shown in the figure, a region recognition method based on the wireless channel fingerprint characteristics of the converter station inspection robot is provided. Figure 1 Taking the server 104 in the example as an example, the following steps are included:

[0146] Step S702, obtaining a patrol robot transmission signal received by a signal receiving end of the converter station; the patrol robot transmission signal is a wireless signal transmitted by the patrol robot in the converter station;

[0147] Step S704, inputting the transmission signal of the inspection robot into the wireless channel feature extraction model to obtain signal feature data corresponding to each channel sample point in the wireless channel of the inspection robot;

[0148] Step S706, generating a signal feature data sequence according to the signal feature data corresponding to each channel sample point, and performing fingerprint feature extraction on the signal feature data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the inspection robot; each fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot;

[0149] Step S708, generating a fingerprint feature vector corresponding to the area where the inspection robot is currently located according to the multiple fingerprint feature information;

[0150] Step S710, obtain the fingerprint feature vector corresponding to each inspection area in the converter station, compare the corresponding fingerprint feature vector of each inspection area with the fingerprint feature vector corresponding to the current area, and obtain the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional position information of the current area in the converter station.

[0151] It should be noted that the specific definition of the above steps can be found in the above specific definition of a region identification method based on the wireless channel fingerprint characteristics of a converter station inspection robot, which will not be repeated here.

[0152] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0153] Based on the same inventive concept, the embodiment of the present application also provides a region identification device based on the wireless channel fingerprint feature of the converter station inspection robot for implementing the region identification method based on the wireless channel fingerprint feature of the converter station inspection robot involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more region identification devices based on the wireless channel fingerprint feature of the converter station inspection robot provided below can be referred to the limitations of the region identification method based on the wireless channel fingerprint feature of the converter station inspection robot above, and will not be repeated here.

[0154] In one embodiment, Figure 8 As shown, a region recognition device based on the wireless channel fingerprint feature of a converter station inspection robot is provided, comprising:

[0155] The acquisition module 802 is used to acquire the inspection robot transmission signal received by the signal receiving end of the converter station; the inspection robot transmission signal is a wireless signal transmitted by the inspection robot in the converter station;

[0156] The extraction module 804 is used to extract multiple fingerprint feature information corresponding to the wireless channel of the inspection robot according to the signal transmitted by the inspection robot; each fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot;

[0157] A generation module 806 is used to generate a fingerprint feature vector corresponding to the area where the inspection robot is currently located according to the multiple fingerprint feature information;

[0158] Determination module 808 is used to obtain the fingerprint feature vectors corresponding to each inspection area in the converter station, and compare the corresponding fingerprint feature vectors of each inspection area with the fingerprint feature vectors corresponding to the current area to obtain the regional similarity information corresponding to each inspection area; the regional similarity information is used to determine the regional position information of the current area in the converter station.

[0159] In one of the embodiments, based on the signal transmitted by the patrol robot, multiple fingerprint feature information corresponding to the wireless channel of the patrol robot is extracted. The extraction module 804 is specifically used to input the signal transmitted by the patrol robot into the wireless channel feature extraction model to obtain the signal feature data corresponding to each channel sample point in the wireless channel of the patrol robot; based on the signal feature data corresponding to each channel sample point, a signal feature data sequence is generated, and fingerprint feature extraction is performed on the signal feature data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the patrol robot.

[0160] In one of the embodiments, the transmission signal of the patrol robot is input into the wireless channel feature extraction model to obtain the signal feature data corresponding to each channel sample point in the wireless channel of the patrol robot. The extraction module 804 is specifically used to input the transmission signal of the patrol robot into the ideal signal feature extraction sub-model to obtain the ideal signal feature data corresponding to each channel sample point in the wireless channel of the patrol robot; the ideal signal feature data is input into the actual signal feature extraction sub-model to obtain the signal feature data actually corresponding to each channel sample point in the wireless channel of the patrol robot.

[0161] In one of the embodiments, a signal characteristic data sequence is generated based on the signal characteristic data corresponding to each channel sample point, and fingerprint feature extraction is performed on the signal characteristic data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the patrol robot. The extraction module 804 is specifically used to generate a signal characteristic data sequence corresponding to the wireless channel of the patrol robot based on the signal characteristic data corresponding to each channel sample point; perform noise reduction processing on the signal characteristic data sequence to obtain a noise-reduced signal characteristic data sequence; perform fingerprint feature extraction on the noise-reduced signal characteristic data sequence to obtain multiple fingerprint feature information corresponding to the wireless channel of the patrol robot.

[0162] In one of the embodiments, a signal feature data sequence is subjected to denoising processing to obtain a denoised signal feature data sequence, and an extraction module 804 is specifically used to generate a signal feature matrix corresponding to a wireless channel of the inspection robot based on the signal feature data sequence; the signal feature matrix is ​​subjected to pre-denoising processing to obtain a pre-denoised signal feature matrix; elements in the pre-denoised signal feature matrix are extracted to generate a denoised signal feature data sequence; and the pre-denoised signal feature data sequence is input into a wavelet denoising model to obtain a denoised signal feature data sequence.

[0163] In one of the embodiments, a fingerprint feature vector corresponding to each inspection area in the converter station is obtained, and the fingerprint feature vector corresponding to each inspection area is compared with the fingerprint feature vector corresponding to the current area to obtain regional similarity information corresponding to each inspection area. A determination module is specifically used to obtain the fingerprint feature vector corresponding to each inspection area in the converter station, and the fingerprint feature vector corresponding to each inspection area is compared with the fingerprint feature vector corresponding to the current area to obtain the fingerprint feature similarity vector corresponding to each inspection area; a fingerprint feature weight vector is generated according to the weight information corresponding to each fingerprint feature information, and the regional similarity information corresponding to each inspection area is determined according to the fingerprint feature similarity vector and the fingerprint feature weight vector corresponding to each inspection area; the regional similarity information corresponding to each inspection area represents the regional similarity between each inspection area and the current area.

[0164] Each module in the above-mentioned area recognition device based on the wireless channel fingerprint characteristics of the converter station inspection robot can be fully or partially implemented by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0165] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig. 9 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store regional identification data based on the wireless channel fingerprint characteristics of the converter station inspection robot. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a regional identification method based on the wireless channel fingerprint characteristics of the converter station inspection robot is implemented.

[0166] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is only a block diagram of a part of the structure 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 certain components, or have a different arrangement of components.

[0167] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above-mentioned method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot. Here, the steps of the method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot can be the steps of the method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot in each of the above-mentioned embodiments.

[0168] In one embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the processor executes the steps of the above-mentioned method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot. Here, the steps of the method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot can be the steps of the method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot in each of the above-mentioned embodiments.

[0169] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot. Here, the steps of the method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot can be the steps of the method for identifying a region based on the fingerprint characteristics of the wireless channel of the converter station inspection robot in each of the above-mentioned embodiments.

[0170] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0171] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and 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-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile 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. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0172] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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, they should be considered to be within the scope of this specification.

[0173] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for identifying regions based on the fingerprint characteristics of wireless channels of a converter station inspection robot, characterized in that: The method comprises: Acquire a patrol robot transmission signal received by a signal receiving end of a converter station; the patrol robot transmission signal is a wireless signal transmitted by the patrol robot in the converter station; Extracting multiple fingerprint feature information corresponding to the wireless channel of the inspection robot according to the signal transmitted by the inspection robot; each of the fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot; Generate a fingerprint feature vector corresponding to the area where the inspection robot is currently located according to the multiple fingerprint feature information; Obtain fingerprint feature vectors corresponding to each inspection area in the converter station, compare the fingerprint feature vectors corresponding to each inspection area with the fingerprint feature vectors corresponding to the current area, and obtain regional similarity information corresponding to each inspection area, including: comparing the fingerprint feature vectors corresponding to each inspection area with the fingerprint feature vectors corresponding to the current area, and obtaining fingerprint feature similarity vectors corresponding to each inspection area; generating a fingerprint feature weight vector according to weight information corresponding to each fingerprint feature information, and determining the regional similarity information corresponding to each inspection area according to the fingerprint feature similarity vectors corresponding to each inspection area and the fingerprint feature weight vector; the regional similarity information corresponding to each inspection area represents the regional similarity between each inspection area and the current area; the regional similarity information is used to determine the regional position information of the current area in the converter station.

2. The method according to claim 1, characterized in that The extracting, according to the signal transmitted by the inspection robot, a plurality of fingerprint feature information corresponding to the wireless channel of the inspection robot comprises: Inputting the transmission signal of the inspection robot into a wireless channel feature extraction model to obtain signal feature data corresponding to each channel sample point in the wireless channel of the inspection robot; A signal feature data sequence is generated according to the signal feature data corresponding to each of the channel sample points, and fingerprint feature extraction is performed on the signal feature data sequence to obtain a plurality of fingerprint feature information corresponding to the wireless channel of the inspection robot.

3. The method according to claim 2, characterized in that The wireless channel feature extraction model includes an ideal signal feature extraction sub-model and an actual signal feature extraction sub-model. The signal transmitted by the inspection robot is input into the first wireless channel feature model to obtain signal feature data corresponding to each channel sample point in the wireless channel of the inspection robot, including: Inputting the transmission signal of the inspection robot into the ideal signal feature extraction sub-model to obtain the ideal signal feature data corresponding to each channel sample point in the wireless channel of the inspection robot; The ideal signal feature data is input into the actual signal feature extraction sub-model to obtain the signal feature data actually corresponding to each channel sample point in the wireless channel of the inspection robot.

4. The method according to claim 2, characterized in that: The generating of a signal feature data sequence according to the signal feature data corresponding to each of the channel sample points, and performing fingerprint feature extraction on the signal feature data sequence to obtain a plurality of fingerprint feature information corresponding to the wireless channel of the inspection robot includes: Generate a signal characteristic data sequence corresponding to the wireless channel of the inspection robot according to the signal characteristic data corresponding to each of the channel sample points; Performing noise reduction processing on the signal characteristic data sequence to obtain a noise-reduced signal characteristic data sequence; Fingerprint feature extraction is performed on the signal feature data sequence after noise reduction to obtain multiple fingerprint feature information corresponding to the wireless channel of the inspection robot.

5. The method according to claim 4, characterized in that The performing noise reduction processing on the signal characteristic data sequence to obtain the signal characteristic data sequence after noise reduction comprises: Generate a signal feature matrix corresponding to the wireless channel of the inspection robot according to the signal feature data sequence; Performing pre-noise reduction processing on the signal feature matrix to obtain a signal feature matrix after pre-noise reduction; Extracting elements from the pre-denoised signal feature matrix to generate a denoised signal feature data sequence; The signal characteristic data sequence after pre-denoising is input into a wavelet denoising model to obtain a signal characteristic data sequence after denoising.

6. A region recognition device based on the wireless channel fingerprint characteristics of a converter station inspection robot, characterized in that: The device comprises: An acquisition module, used to acquire a patrol robot transmission signal received by a signal receiving end of a converter station; the patrol robot transmission signal is a wireless signal transmitted by the patrol robot in the converter station; An extraction module, used to extract multiple fingerprint feature information corresponding to the wireless channel of the inspection robot according to the signal transmitted by the inspection robot; each of the fingerprint feature information is used to characterize the corresponding signal fading characteristics in the wireless channel of the inspection robot; A generation module, used to generate a fingerprint feature vector corresponding to the area where the inspection robot is currently located according to the multiple fingerprint feature information; A determination module is used to obtain fingerprint feature vectors corresponding to each inspection area in the converter station, and compare the fingerprint feature vectors corresponding to each inspection area with the fingerprint feature vectors corresponding to the current area to obtain regional similarity information corresponding to each inspection area, including: comparing the fingerprint feature vectors corresponding to each inspection area with the fingerprint feature vectors corresponding to the current area to obtain fingerprint feature similarity vectors corresponding to each inspection area; generating a fingerprint feature weight vector according to weight information corresponding to each fingerprint feature information, and determining the regional similarity information corresponding to each inspection area according to the fingerprint feature similarity vectors corresponding to each inspection area and the fingerprint feature weight vector; the regional similarity information corresponding to each inspection area represents the regional similarity between each inspection area and the current area; the regional similarity information is used to determine the regional position information of the current area in the converter station.

7. A computer device comprising a memory and a processor, wherein the memory stores 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.

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

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

Images