Anti-interference method for power line carrier communication

Abstract

The invention relates to the technical field of communication anti-interference, in particular to an anti-interference method for power line carrier communication, which comprises the following steps: acquiring a first power line carrier signal on a target power line in real time through a signal collector, and inputting the first power line carrier signal into a first interference detection model to obtain an interference detection result of the first power line carrier signal; judging whether interference exists in the first power line carrier signal according to the interference detection result; if the interference exists, inputting the first power carrier signal into a second interference prediction model, and judging whether an interference risk exists in the future; and if the risk exists, matching a first interference control strategy corresponding to an interference detection result according to the interference detection result of the first power line carrier signal and an interference control strategy library so as to control the first power line carrier signal in real time and reduce the risk that the signal is interfered. The method has the beneficial effects that the accurate identification and automatic response of the interference source are realized, and the accuracy and reliability of data transmission are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of communication anti-interference, and in particular to an anti-interference method for power line carrier communication. Background Art

[0002] Power carrier communication technology uses existing power lines as a communication medium to achieve data information transmission. It uses power lines (such as low-voltage distribution networks) as communication channels, loads information signals onto power lines through modulation technology for transmission, and restores the original information signals at the receiving end through demodulation technology. This technology has the advantages of low cost and easy deployment, so it has been widely used in smart grids, smart homes and other fields. However, due to the complex environment in which the power lines themselves are located, power carrier communication faces serious interference problems, such as lightning interference, electrical interference, weather interference, etc. Interference signals will have a serious impact on power carrier signals, thereby reducing the quality and reliability of signal transmission.

[0003] In summary, the existing carrier communication technology is severely affected by obstacles such as weather and buildings, has poor signal transmission possibility, has unavoidable technical bottlenecks in terms of noise and impedance, and has poor anti-interference ability. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides an anti-interference method for power carrier communication, which solves the technical problems that the existing carrier communication technology is easily affected by the external environment and has poor anti-interference ability.

[0005] (II) Technical solution In order to achieve the above object, the main technical solutions adopted by the present invention include: In a first aspect, an embodiment of the present invention provides an anti-interference method for power line carrier communication, comprising: S11, acquiring a first power carrier signal on a target power line in real time through a preset signal collector, inputting the first power carrier signal into a preset first interference detection model, and obtaining an interference detection result of the first power carrier signal; S12, judging whether there is interference in the first power carrier signal according to the interference detection result of the first power carrier signal; If there is interference in the first power carrier signal, the first power carrier signal is input into a preset second interference prediction model to determine whether there is a risk of interference in the future; S13. If there is an interference risk in the future, the first interference control strategy corresponding to the interference detection result is matched according to the interference detection result of the first power carrier signal and a pre-set interference control strategy library, so as to control the first power carrier signal in real time and reduce the risk of signal interference.

[0006] Optionally, before S11, the step further includes: S10, acquiring the communication signal on the target power line in real time, and converting the communication signal into a corresponding digital signal after amplifying and filtering the acquired communication signal on the target power line; Based on a preset digital signal processing algorithm and an adaptive equalization algorithm, the digital signal is frequency-converted and compensated to obtain a first power carrier signal corresponding to the communication signal.

[0007] Optionally, the first interference detection model includes a decision tree sub-model and an interference source position detection sub-model; Then the S11 includes: S11-1, performing baseband modulation on the received first power carrier signal based on a preset baseband modulation algorithm to obtain a corresponding time domain signal; According to the time domain signal corresponding to the first power carrier signal and the preset formula 1, the frequency domain signal corresponding to the first power carrier signal is obtained; the formula 1 is: X freq =Re(FFT(X time ,dim=2)); Among them, X freq is the frequency domain signal corresponding to the first power carrier signal, FFT is fast Fourier transform, Re represents the real part of the complex number, dim=2 represents the operation in the second dimension, X time is a time domain signal corresponding to the first power carrier signal; S11-2, extracting features from a frequency domain signal and a time domain signal corresponding to the first power carrier signal to obtain feature parameters corresponding to the first power carrier signal; S11-3, inputting characteristic parameters corresponding to the first power carrier signal into a preset decision tree sub-model to obtain interference type and interference intensity corresponding to the first power carrier signal; S11-4. Input the interference type and interference intensity corresponding to the first power carrier signal into a preset interference source position detection sub-model to obtain an interference detection result corresponding to the first power carrier signal.

[0008] Optionally, the S11-3 includes: Obtaining a high-dimensional feature vector data set of the first power carrier signal according to the feature parameters corresponding to the first power carrier signal; Inputting a high-dimensional feature vector data set corresponding to the first power carrier signal into a pre-set decision tree sub-model to obtain an interference type and interference intensity corresponding to the first power carrier signal; Wherein, the characteristic parameters include one or more of the following: time domain skewness, time domain kurtosis, normalized instantaneous amplitude absolute value standard deviation, normalized instantaneous frequency absolute value standard deviation, time domain envelope fluctuation, fast pulse modulation coefficient, box dimension, information dimension, frequency domain skewness, frequency domain kurtosis, frequency domain envelope fluctuation, additive Gaussian white noise factor, carrier factor, spectrum information entropy, spectrum index entropy, spectrum norm entropy, low-frequency approximate coefficient energy proportion, high-frequency detail coefficient energy proportion, wavelet energy spectrum entropy, wavelet variance, bispectral entropy, contour-integrated bispectral mean, contour-integrated bispectral variance and bispectral waveform entropy; Then the pre-set decision tree sub-model includes: a model obtained by training the decision tree sub-model using a first sample training set; the first sample training set includes: a high-dimensional feature vector data set corresponding to at least one second power carrier signal, and an interference type and interference intensity corresponding to each high-dimensional feature vector data set; The second power carrier signal is a historical power carrier signal of the target power line.

[0009] Optionally, the S11-4 includes: According to the interference intensity corresponding to the first power carrier signal and the preset formula 2, the propagation distance of the interference signal corresponding to the first power carrier signal is obtained; the formula 2 is: d×A=20log 10 (d)+20log 10 (f)+20log 10 (4π / c); Wherein, d is the propagation distance of the interference signal corresponding to the first power carrier signal, A is the preset attenuation, f is the signal frequency, and c is the speed of light; Obtaining the interference source location range according to the propagation distance of the interference signal corresponding to the first power carrier signal and the preset position coordinates of the signal collector; Based on the pre-set power line network topology map, all interference sources within the interference source location range are obtained; According to the interference type corresponding to the first power carrier signal, all interference sources within the interference source position range are screened to obtain the interference source position corresponding to the first power carrier signal; The interference detection result includes interference type, interference intensity and interference source location.

[0010] Optionally, the S12 includes: Determining whether there is interference in the first power carrier signal according to the interference detection result of the first power carrier signal; If there is interference in the first power carrier signal, the first power carrier signal is input into a preset second interference prediction model to determine whether there is a risk of interference in the future; During the process, the characteristic parameters are classified according to the characteristic attributes of each characteristic vector in the characteristic parameters corresponding to the first power carrier signal to obtain at least one characteristic block; the characteristic block includes one or more of the following: a time domain characteristic block, a fractal domain characteristic block, a frequency domain characteristic block, a wavelet domain characteristic block or a dual spectrum domain characteristic block; Based on the feature attribute corresponding to each feature block and the first linear transformation matrix preset for each feature attribute, and the preset formula three, each feature block is mapped to the target dimension; the formula three is: E i =α i ×patch i ; Where E is the feature block after linear transformation, patch is the feature block, α is the first linear transformation matrix pre-set for each feature attribute, and i is the index of the feature block; According to the preset attention mechanism, the attention matrix corresponding to each feature block is obtained; According to the attention matrix corresponding to each feature block, it is determined whether there is interference risk in the future.

[0011] Optionally, the step S12, obtaining an attention matrix corresponding to each feature block according to a preset attention mechanism, includes: According to the preset attention mechanism and the preset attention allocation weight of each feature attribute, and the preset formula 4, the attention matrix corresponding to each feature block is obtained; the formula 4 is: ; Among them, Q is the query vector, K is the key vector, and V is the value vector. is a preset scaling factor, T is a preset transpose, i is the index of the feature block, and β is the attention allocation weight corresponding to the feature attribute.

[0012] Optionally, the S12, judging whether there is interference risk in the future according to the attention matrix corresponding to each feature block, includes: The attention matrix corresponding to each feature block is concatenated with the original features corresponding to the feature block to obtain the fused features corresponding to each feature block; According to the fusion features corresponding to each feature block and the preset attention score weight matrix, and the preset formula 5, the attention score corresponding to each feature block is obtained; the formula 5 is: S i =C i γ score; Among them, S is the attention score corresponding to the feature block, C is the fusion feature corresponding to the feature block, γ score is the attention score weight matrix; The attention scores corresponding to each feature block are normalized by a preset softmax function to generate the attention weights corresponding to each feature block; According to the attention weight and fusion feature corresponding to each feature block, and the preset formula 6, the global feature corresponding to the first power carrier signal is obtained; the formula 6 is: ; Among them, C is the fusion feature corresponding to the feature block, B is the attention weight corresponding to the feature block, c is the global feature corresponding to the first power carrier signal, n is the number of feature blocks, and i is the index of the feature block; Whether there is an interference risk in the future is determined according to the global characteristics corresponding to the first power carrier signal.

[0013] Optionally, before S11, the step further includes: S01. Collect a second power carrier signal corresponding to a target power line, and extract features of the second power carrier signal to obtain feature parameters corresponding to the second power carrier signal; Obtaining the interference type and interference intensity corresponding to the second power carrier signal according to the characteristic parameters corresponding to the second power carrier signal; According to the interference type and interference intensity corresponding to the second power carrier signal, match the second interference control strategy corresponding to each preset interference type and / or interference intensity, and establish an interference control strategy library; The second power carrier signal is a historical power carrier signal of the target power line.

[0014] Optionally, after S13, the step further includes: S14, inputting the interference detection result corresponding to the first power carrier signal into a preset interference control compensation model to obtain a compensation result corresponding to the first power carrier signal; updating the first interference control strategy based on the compensation result so as to control the first power carrier signal of the power supply in real time and reduce the risk of signal interference; The preset interference control compensation model includes: a compensation model obtained by training the interference control compensation model using a second sample training set; the second sample training set includes: an interference detection result corresponding to at least one second power carrier signal, and a compensation result corresponding to each interference detection result; The second power carrier signal is a historical power carrier signal of the target power line.

[0015] (III) Beneficial effects The beneficial effects of the present invention are as follows: an anti-interference method for power carrier communication of the present invention can automatically identify whether interference exists according to the power carrier signal and match the corresponding interference control strategy. Compared with the prior art, it realizes accurate identification and automatic response of interference sources, and improves the accuracy and reliability of data transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of an anti-interference method for power line carrier communication provided by an embodiment of the present invention; Figure 2 A structural flow chart of an anti-interference method for power carrier communication provided by an embodiment of the present invention; Figure 3 A flowchart of obtaining the location of an interference source in an anti-interference method for power line carrier communication provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0018] An anti-interference method for power carrier communication proposed in an embodiment of the present invention can automatically identify whether interference exists according to the power carrier signal and match the corresponding interference control strategy. Compared with the existing technology, it realizes accurate identification and automatic response of interference sources, and improves the accuracy and reliability of data transmission.

[0019] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0020] Example 1

[0021] This embodiment provides an anti-interference method for power line carrier communication, such as Figure 1 As shown, including: S11, acquiring a first power carrier signal on a target power line in real time through a preset signal collector, inputting the first power carrier signal into a preset first interference detection model, and obtaining an interference detection result of the first power carrier signal; S12, judging whether there is interference in the first power carrier signal according to the interference detection result of the first power carrier signal; If there is interference in the first power carrier signal, the first power carrier signal is input into a preset second interference prediction model to determine whether there is a risk of interference in the future; S13. If there is an interference risk in the future, the first interference control strategy corresponding to the interference detection result is matched according to the interference detection result of the first power carrier signal and a pre-set interference control strategy library, so as to control the first power carrier signal in real time and reduce the risk of signal interference.

[0022] The present embodiment provides an anti-interference method for power carrier communication, which solves the problem of interference to power lines and serious impact on power carrier communication signals by performing interference detection and interference prediction on power carrier signals, realizes accurate identification of interference sources, and improves the accuracy and reliability of data transmission.

[0023] Example 2

[0024] This embodiment provides an anti-interference method for power line carrier communication, including: S10, acquiring the communication signal on the target power line in real time, and converting the communication signal into a corresponding digital signal after amplifying and filtering the acquired communication signal on the target power line; Based on a preset digital signal processing algorithm and an adaptive equalization algorithm, the digital signal is frequency-converted and compensated to obtain a first power carrier signal corresponding to the communication signal.

[0025] S11, performing baseband modulation on the received first power carrier signal according to a preset baseband modulation algorithm to obtain a corresponding time domain signal; According to the time domain signal corresponding to the first power carrier signal and the preset formula 1, the frequency domain signal corresponding to the first power carrier signal is obtained; the formula 1 is: X freq =Re(FFT(X time ,dim=2)); Among them, X freq is the frequency domain signal corresponding to the first power carrier signal, FFT is fast Fourier transform, Re represents the real part of the complex number, dim=2 represents the operation in the second dimension, X time is a time domain signal corresponding to the first power carrier signal; Extracting features of a frequency domain signal and a time domain signal corresponding to the first power carrier signal to obtain feature parameters corresponding to the first power carrier signal; Obtaining a high-dimensional feature vector data set of the first power carrier signal according to the feature parameters corresponding to the first power carrier signal; The high-dimensional feature vector data set corresponding to the first power carrier signal is input into a pre-set decision tree sub-model to obtain the interference type and interference intensity corresponding to the first power carrier signal; if the decision tree sub-model outputs that the interference type corresponding to the first power carrier signal is no interference and the interference intensity is zero, then there is no need for the interference source position detection sub-model to detect the interference source position.

[0026] Wherein, the characteristic parameters include one or more of the following: time domain skewness, time domain kurtosis, normalized instantaneous amplitude absolute value standard deviation, normalized instantaneous frequency absolute value standard deviation, time domain envelope fluctuation, fast pulse modulation coefficient, box dimension, information dimension, frequency domain skewness, frequency domain kurtosis, frequency domain envelope fluctuation, additive Gaussian white noise factor, carrier factor, spectrum information entropy, spectrum index entropy, spectrum norm entropy, low-frequency approximate coefficient energy proportion, high-frequency detail coefficient energy proportion, wavelet energy spectrum entropy, wavelet variance, bispectral entropy, contour-integrated bispectral mean, contour-integrated bispectral variance and bispectral waveform entropy; Then the pre-set decision tree sub-model includes: a model obtained by training the decision tree sub-model using a first sample training set; the first sample training set includes: a high-dimensional feature vector data set corresponding to at least one second power carrier signal and an interference type and interference intensity corresponding to each high-dimensional feature vector data set; The second power carrier signal is a historical power carrier signal of the target power line.

[0027] According to the interference intensity corresponding to the first power carrier signal and the preset formula 2, the propagation distance of the interference signal corresponding to the first power carrier signal is obtained; the formula 2 is: d×A=20log 10 (d)+20log 10 (f)+20log 10 (4π / c); Wherein, d is the propagation distance of the interference signal corresponding to the first power carrier signal, A is the preset attenuation, f is the signal frequency, and c is the speed of light; Obtaining the interference source location range according to the propagation distance of the interference signal corresponding to the first power carrier signal and the preset position coordinates of the signal collector; Based on the pre-set power line network topology map, all interference sources within the interference source location range are obtained; According to the interference type corresponding to the first power carrier signal, all interference sources within the interference source position range are screened to obtain the interference source position corresponding to the first power carrier signal; The interference detection result includes the interference type, interference intensity and interference source location. The interference type includes interference or no interference, and when there is interference, the interference type is a specific type (such as white noise interference, narrowband interference, broadband interference, multipath interference, inter-code interference, phase noise, pulse interference, etc.). When the decision tree submodel determines that the interference type of the first power carrier signal is no interference and the interference intensity is zero, there is no need to detect the location of the interference source through the interference source location detection submodel, and the interference detection result is directly output. At this time, the interference detection result includes the interference type (no interference) and the interference intensity (zero).

[0028] This embodiment introduces a decision tree sub-model to directly classify the interference type of the power carrier signal by analyzing the characteristic parameters of the power carrier signal; and the interference source location detection sub-model is based on a pre-set power line network topology diagram, and determines the interference source location by analyzing the interference type and interference intensity.

[0029] S12, judging whether there is interference in the first power carrier signal according to the interference detection result of the first power carrier signal; If there is interference in the first power carrier signal, the first power carrier signal is input into a preset second interference prediction model to determine whether there is a risk of interference in the future; During the process, the characteristic parameters are classified according to the characteristic attributes of each characteristic vector in the characteristic parameters corresponding to the first power carrier signal to obtain at least one characteristic block; the characteristic block includes one or more of the following: a time domain characteristic block, a fractal domain characteristic block, a frequency domain characteristic block, a wavelet domain characteristic block or a dual spectrum domain characteristic block; Based on the feature attribute corresponding to each feature block and the first linear transformation matrix preset for each feature attribute, and the preset formula three, each feature block is mapped to the target dimension; the formula three is: E i =α i ×patch i ; Where E is the feature block after linear transformation, patch is the feature block, α is the first linear transformation matrix pre-set for each feature attribute, and i is the index of the feature block; According to the preset attention mechanism and the preset attention allocation weight of each feature attribute, and the preset formula 4, the attention matrix corresponding to each feature block is obtained; the formula 4 is: ; Among them, Q is the query vector, K is the key vector, and V is the value vector. is a preset scaling factor, T is a preset transpose, i is the index of the feature block, β is the attention allocation weight corresponding to the feature attribute, and E is the feature block after linear transformation.

[0030] The attention matrix corresponding to each feature block is concatenated with the original features corresponding to the feature block to obtain the fused features corresponding to each feature block; According to the fusion features corresponding to each feature block and the preset attention score weight matrix, and the preset formula 5, the attention score corresponding to each feature block is obtained; the formula 5 is: S i =C i γ score ; Among them, S is the attention score corresponding to the feature block, C is the fusion feature corresponding to the feature block, γ score is the attention score weight matrix; The attention scores corresponding to each feature block are normalized by a preset softmax function to generate the attention weights corresponding to each feature block; According to the attention weight and fusion feature corresponding to each feature block, and the preset formula 6, the global feature corresponding to the first power carrier signal is obtained; the formula 6 is: ; Among them, C is the fusion feature corresponding to the feature block, B is the attention weight corresponding to the feature block, c is the global feature corresponding to the first power carrier signal, n is the number of feature blocks, and i is the index of the feature block; Whether there is an interference risk in the future is determined according to the global characteristics corresponding to the first power carrier signal.

[0031] Furthermore, before S11, it also includes: S01. Collect a second power carrier signal corresponding to a target power line, and extract features of the second power carrier signal to obtain feature parameters corresponding to the second power carrier signal; Obtaining the interference type and interference intensity corresponding to the second power carrier signal according to the characteristic parameters corresponding to the second power carrier signal; According to the interference type and interference intensity corresponding to the second power carrier signal, match the second interference control strategy corresponding to each preset interference type and / or interference intensity, and establish an interference control strategy library; The second power carrier signal is a historical power carrier signal of the target power line.

[0032] Furthermore, after S13, it also includes: S14, inputting the interference detection result corresponding to the first power carrier signal into a preset interference control compensation model to obtain a compensation result corresponding to the first power carrier signal; updating the first interference control strategy based on the compensation result so as to control the first power carrier signal of the power supply in real time and reduce the risk of signal interference; The preset interference control compensation model includes: a compensation model obtained by training the interference control compensation model using a second sample training set; the second sample training set includes: an interference detection result corresponding to at least one second power carrier signal, and a compensation result corresponding to each interference detection result.

[0033] Specifically, this embodiment can be applied to data exchange between various devices in the smart grid, such as communication between smart meters and concentrators; or in home / commercial buildings, to realize intelligent functions such as home appliance control, lighting management, and security systems through power lines; or in a factory environment, PLC monitors and controls production processes; urban infrastructure management, such as traffic light control, street light management system, etc.

[0034] This embodiment provides an anti-interference method for power carrier communication, which outputs interference detection results and interference prediction results by building a first interference detection model and a second interference prediction model; establishes an interference control strategy library based on historical data, traverses the interference control strategy library based on the interference detection results, and outputs the interference control strategy; builds an interference control compensation model, outputs the compensation result based on the interference detection result, and updates the interference control strategy based on the compensation result to obtain the updated strategy to adjust the power carrier communication system. It solves the problem that power lines are susceptible to environmental interference, which in turn has a serious impact on power carrier communication, thereby reducing the quality of signal transmission. Through the detection and prediction of power carrier signals, accurate identification of interference sources is achieved, the anti-interference control strategy is optimized, and the accuracy and reliability of data transmission are improved.

[0035] Example 3 This embodiment provides an anti-interference method for power line carrier communication, such as Figure 2 As shown, including: The communication signal is collected and preprocessed to obtain a preprocessed power carrier signal.

[0036] Power carrier communication uses transmission lines as the transmission medium of carrier signals to realize data transmission and information exchange. It uses power lines (such as low-voltage distribution networks) as communication channels and loads information signals onto power lines for transmission through modulation technology. At the receiving end, the original information signal is restored through demodulation technology. Power carrier communication is a communication technology based on power line transmission, which has the characteristics of economy, reliability, and wide application. By adopting advanced modulation technology and anti-interference means, power carrier communication plays an important role in the fields of power systems and smart homes. Among them, anti-interference refers to a device or technology used to counteract any interference in communication or radar operation. Its main purpose is to protect the reception of useful signals and prevent useful signals from being damaged. Anti-interference is academically defined as combining the characteristics of the circuit to minimize interference, and the device can prevent electromagnetic interference from passing through the antenna input, the housing of the device, and acting on the device along the power line. This embodiment provides an anti-interference method for power carrier communication, through which the stability of communication is improved, the anti-interference ability is enhanced, the communication efficiency is improved, and the maintenance cost is reduced, which helps power carrier communication to achieve efficient and stable data transmission services in complex electromagnetic environments.

[0037] To implement the method proposed in this embodiment, it is necessary to select a suitable signal collector. These devices usually include signal acquisition cards or signal receivers, etc., which can capture the power carrier signal (carrier communication signal) on the power line. According to the communication protocol and frequency band, select the signal frequency band to be collected to ensure that the captured power carrier signal is valid. Set a suitable sampling rate according to the signal bandwidth and the required accuracy. A higher sampling rate can restore the signal waveform more accurately, but it will also increase the amount of data processing. Since the signal of the power line may be weak, it is necessary to amplify the signal through a signal amplifier to improve the signal-to-noise ratio. Use a bandpass filter to remove high-frequency noise and low-frequency interference in the signal and retain the signal within the effective frequency band. It helps to improve the purity and readability of the signal. Convert the amplified and filtered analog signal into a digital signal for subsequent digital signal processing. The accuracy and speed of analog-to-digital conversion have a great influence on the signal quality, and convert the high-speed digital signal into a low-speed baseband signal for subsequent signal processing and analysis. Digital down-conversion can be achieved through digital signal processing algorithms, such as fast Fourier transform (FFT). Since the transmission characteristics of the power line may cause signal distortion, signal equalization technology is needed to compensate for these distortions. Signal equalization can be achieved through digital filters or adaptive equalization algorithms. After the above steps, the preprocessed communication signals can be obtained. These signals have removed most of the noise and interference, have higher signal-to-noise ratio and readability, and can be used for subsequent signal analysis, decoding and transmission operations. Select appropriate acquisition equipment and preprocessing algorithms, and set and optimize parameters according to specific application scenarios and communication protocols to obtain high-quality preprocessed power carrier signals.

[0038] Build a first interference detection model and a second interference prediction model, input the preprocessed power carrier signal into the interference detection model and the interference prediction model, and output the interference detection result and the interference prediction result, wherein the interference detection result includes the interference type, interference intensity and interference source location.

[0039] Clarify what types of interference (interference types) need to be detected, determine the accuracy requirements for detecting the location of the interference source, and select a hardware platform suitable for processing power carrier signals, such as DSP (digital signal processor), FPGA (field programmable gate array) or high-performance computer, to ensure that the hardware platform has sufficient processing power to analyze the preprocessed power carrier signal in real time. Design an algorithm to extract key features from the preprocessed power carrier signal, which can be used to distinguish different types of interference. Use the extracted features (i.e., feature parameters) combined with threshold judgment or machine learning models to detect whether there is interference and determine the type of interference. According to the transmission characteristics of the signal on the power line (such as attenuation, reflection, etc.), combined with the network topology, determine the location of the interference source, and design an algorithm to quantify the intensity of the interference, such as by measuring the signal-to-noise ratio or bit error rate. Based on historical data and machine learning models, predict possible interference situations in the future. Use simulated data or actual collected data for testing to ensure that the module can accurately detect interference and output correct results.

[0040] Select hardware equipment, design and make circuit boards, and connect the selected hardware equipment with communication interfaces and storage devices; design interference detection algorithms and prediction algorithms based on the characteristics of the hardware platform, and design the first interference detection model and the second interference prediction model based on the interference detection algorithms and prediction algorithms.

[0041] According to the characteristics of the power carrier signal, select a suitable signal acquisition card or receiver to ensure that the communication signal can be captured and digitized. Select a high-performance DSP, FPGA or microprocessor to perform complex signal processing and analysis tasks. Select an appropriate storage device (such as SD card, solid-state hard disk, etc.) to store historical data, model parameters and detection results. Interfaces for communication with communication systems and other devices, such as Ethernet, serial port, etc., are suitable. According to the specifications and interface requirements of the selected hardware equipment, design the schematic diagram of the circuit board, including the connection relationship between signal acquisition, processing, communication and storage parts. Based on the schematic diagram, carry out PCB (printed circuit board) layout design to ensure that the connection between components is reasonable, the wiring is compact, and the electromagnetic compatibility is good. Send the PCB layout file to the circuit board manufacturer for production, and then accept and test it after production. According to the circuit board design, connect the signal acquisition device, processor, communication interface and storage device together to form a complete interference detection system. Ensure that the connection between the interfaces is correct and use appropriate connection cables and connectors. According to the characteristics of the hardware platform and the complexity of the algorithm, select a suitable programming language for algorithm implementation. C / C++ is a common choice because they have high execution efficiency and good portability. Design the overall framework and module division of the software to ensure that the algorithm can run efficiently and interact with other modules. Use historical data and simulation data to train the model, optimize the parameters and structure of the model, and improve the accuracy and generalization ability of the model. Integrate the trained first interference detection model and the second prediction model into the interference detection system to ensure that the system can receive the pre-processed power carrier signal in real time, perform interference detection and interference prediction tasks, and output detection results and prediction results.

[0042] like Figure 3 As shown, relevant information of characteristic parameters is extracted, including signal spectrum, time domain characteristics and frequency domain characteristics; characteristic parameters are classified and identified based on the decision tree sub-model and stored in the interference detection system; the power carrier signal is input into the interference position detection sub-model for positioning analysis to obtain the interference source position.

[0043] Extract features of the power carrier signal, use fast Fourier transform (FFT) or similar algorithms to calculate the signal spectrum, and extract key parameters from the spectrum, such as main frequency, bandwidth, power density, etc. Analyze the waveform of the signal, extract features such as amplitude, period, pulse width, etc., to obtain the statistical characteristics of the signal, such as mean variance, skewness, kurtosis, etc. Use short-time Fourier transform (STFT) or wavelet transform to simultaneously analyze the features of the signal in the time domain and frequency domain, and extract spectrum parameters that change over time, such as frequency change rate, spectrum center of gravity, etc. Normalize or standardize the extracted features to ensure that different features are comparable in value.

[0044] The decision tree model is trained using the first sample training set marked with labels (including power carrier signals of consistent interference types). The decision tree recursively divides the data set into subsets to minimize the classification error rate or maximize the information gain. At each internal node, the decision tree divides the data set into two or more subsets based on the value of a certain feature. Each leaf node represents a category (i.e., interference type).

[0045] In addition to classification and identification, the first interference detection model also needs to be able to locate the interference source. This may require combining information such as network topology, signal attenuation and reflection characteristics. Positioning algorithms based on signal arrival time (TOA), arrival time difference (TDOA) or arrival angle (AOA) can be used.

[0046] The preprocessed power carrier signal is input into the first interference detection model, and the pre-trained decision tree sub-model in the first interference detection model is used to extract features and classify the input signal. If interference is detected, the interference location detection sub-model is used to estimate the location of the interference source based on the characteristics of signal propagation in the network. At the same time, the type of interference source is determined in combination with the classification results of the decision tree. The analysis results of the first interference detection model (including the location and type of the interference source) are output in an appropriate manner, such as displaying on the monitoring interface, sending alarm information, etc.

[0047] Specifically, the trained first interference detection model is used to determine whether there is interference in the preprocessed power carrier signal. If it is determined that there is no interference in the signal, the location type information (which is a normal identification of the power carrier signal, not the specific location of the interference source) is directly output. If there is interference, the power carrier communication signal needs to be monitored in real time. This can be achieved through a loop or a timed task to continuously capture and analyze the power carrier signal. During the real-time detection process, the power carrier signal is continuously captured and the captured signal is preprocessed to meet the input requirements of the second interference prediction model. The preprocessed power carrier signal is input into the second interference prediction model. The second interference prediction model analyzes based on the input signal and outputs the comment result. These comment results generally include two states: predicted interference and non-interference. The comment result is a prediction of the future state of the real-time power carrier signal, that is, whether the signal may be interfered in a certain period of time in the future. In order to distinguish between interference and interference states in the prediction results, one or more safety thresholds need to be established. These thresholds may be based on the output score, probability or other indicators of the model. According to the output of the prediction model and the set safety threshold, it is determined whether the comment result meets the non-interference condition. If it is satisfied, it is considered that the current power carrier signal is in a safe state; otherwise, it is considered that there may be interference risk.

[0048] An interference control strategy library is established based on historical data, and the interference control strategy library is traversed according to the interference detection results to output the interference control strategy.

[0049] Collect historical power carrier signal data, including interference-free signals, various types of interference signals and corresponding interference control strategies, and organize the collected data, including data cleaning, format unification, and annotation. Extract interference-related features from the organized historical data, such as signal spectrum, time domain features, time-frequency features, etc. According to the extracted features, classify the interference signal to determine the type and intensity of the interference. Based on the processing experience of different interference types and intensities in historical data, formulate corresponding interference control strategies. These strategies may include signal filtering, retransmission mechanism, frequency switching, power adjustment, etc. The formulated interference control strategy is stored in the interference control strategy library according to certain rules (such as interference type and intensity). Use the first interference detection model to perform interference detection on the preprocessed power carrier signal, and traverse the interference control strategy library according to the interference detection results to find the matching interference control strategy. Output the matched interference control strategy to the corresponding execution module or system to control the power carrier communication signal in real time.

[0050] Collect historical data, clean and annotate the historical data, and obtain the data stream of interference detection result-control strategy; integrate the data stream and establish an interference control strategy library; use the interference control strategy library to match interference strategies and extract interference control strategies; output the interference control strategies.

[0051] Determine the sources for collecting historical data, which may include log files, database records, experimental data, etc., and use appropriate tools and methods to collect data related to power carrier communication interference from historical data sources. Remove duplicate, incomplete or invalid data. Correct obvious errors in the data, such as incorrect dates, timestamps, values, etc. Extract interference-related features from the cleaned data, such as signal strength, spectrum characteristics, time domain characteristics, etc. Based on feature analysis, mark the interference situation of each data point or data segment, such as interference type, interference intensity, etc. For known interference control strategies, mark them with the corresponding interference detection results to form data pairs of interference detection results-interference control strategies. Design the structure of the database, including tables, fields, indexes, etc., to store data pairs of interference detection results-interference control strategies. Integrate the marked data into the database to form an interference control strategy library. Verify the data in the interference control strategy library to ensure the accuracy and consistency of the data. When a new interference detection result is detected, the characteristic parameters corresponding to the interference detection result are input into the interference control strategy library, and historical data similar to the input interference detection result is queried. The characteristics of the historical data and the current interference detection result are compared to find the most matching interference detection result-interference control strategy data pair. The corresponding interference control strategy is extracted from the matching data pair. The evaluated interference control strategy is output to the execution system or operator.

[0052] Build an interference control compensation model, input the interference detection result into the interference control compensation model, output the compensation result, update the interference control strategy according to the compensation result, obtain the updated control strategy, and adjust the power carrier communication system according to the updated control strategy.

[0053] Prepare a second sample training set for training the compensation model, including historical interference detection results, corresponding compensation measures, and compensated interference detection system performance data. Use the prepared second sample training set to train the interference control compensation model so that it can learn the mapping relationship from interference detection results to compensation measures. Monitor the operating status of the carrier communication system in real time, and obtain the current interference detection results through the first interference detection model. Perform necessary preprocessing on the interference detection results, such as data cleaning, format conversion, etc., to meet the input requirements of the compensation model. Input the preprocessed interference detection results into the trained interference control compensation model for inference calculation. Obtain the output compensation results from the compensation model, which may include system parameters that need to be adjusted, recommended compensation measures, etc. According to the compensation results, analyze the impact of the current interference on the power carrier communication system and the possible effects of the compensation measures. Update the original interference control strategy based on the strategy analysis and compensation results. This may include modifying control parameters, adjusting control logic, adding new control strategies, etc. Verify the updated interference control strategy to ensure that it can effectively cope with the current interference situation in practical applications. Store the updated control strategy that has passed the verification in the interference control strategy library for subsequent use. According to the updated interference control strategy, the relevant parameters and configurations of the power carrier communication system are adjusted to cope with the current interference situation. During and after the adjustment process, the system performance and interference situation are continuously monitored and feedback data is collected to further optimize the interference control strategy.

[0054] The second sample training set is used to divide the data, and the interference control compensation model is supervised and trained. When the interference control compensation model converges, the accuracy result is calculated. The accuracy index is preset, and when the accuracy result meets the accuracy index, the interference control compensation model is obtained.

[0055] Furthermore, historical interference data is collected from the logs, monitoring data or other sources of the power carrier communication system to ensure that the collected data contains sufficient interference detection results and corresponding system response information. The collected historical interference data is cleaned to remove erroneous, duplicated or invalid data. The data is transformed and standardized as necessary so that the neural network can better process the data. Features related to interference, such as interference type, intensity, duration, etc., and related parameters of system response are extracted. The preprocessed data is divided into input features (i.e., interference detection results) and output labels (i.e., expected compensation results). The data set is divided into training sets, validation sets and test sets for subsequent model training and evaluation. An initial interference control compensation model is built based on a neural network (such as a multilayer perceptron, a convolutional neural network, a recurrent neural network, etc.). The structure of the input layer, hidden layer and output layer of the model, as well as the number of neurons in each layer, the activation function and other parameters are determined. The initial interference control compensation model is trained using the second sample training set. The model output is calculated by forward propagation, and the difference between the model output and the true label is measured using a loss function (such as mean square error, cross entropy, etc.). Use the back-propagation algorithm to calculate the gradient and update the weights and bias parameters of the model to minimize the loss function. Repeat the training process until the performance of the model on the validation set reaches a stable or preset maximum number of iterations. Evaluate the performance of the trained model on the test set and calculate indicators such as accuracy, recall, and F1 score. Adjust the model's hyperparameters (such as learning rate, batch size, number of iterations, etc.) based on the evaluation results to optimize the model performance. Preset a reasonable accuracy indicator based on actual needs and application scenarios. This indicator should be able to reflect the performance requirements of the model in actual applications. When the accuracy of the model on the test set meets the preset accuracy indicator, the model training is considered successful. Save the trained model as an interference control compensation model for subsequent deployment and application in the power carrier communication system.

[0056] Obtain compensation results, update the interference control strategy according to the compensation results, and obtain an updated control strategy; implement the updated control strategy on the power carrier communication system, enable a monitoring system to monitor the power carrier communication system, and obtain monitoring results; and determine whether the updated control strategy is effective based on the monitoring results.

[0057] Obtain compensation results from the interference control compensation model. These compensation results are usually suggestions for adjusting certain parameters or configurations of the power carrier communication system. Based on the compensation results, update the existing interference control strategy. This may involve modifying the control logic, adjusting control parameters, or introducing a new control strategy. Ensure that the updated control strategy is compatible with the rest of the power carrier communication system and follows relevant standards and specifications. Verify and test the updated interference control strategy to ensure that it can operate normally and achieve the expected results in the actual environment. If the verification passes, store the updated control strategy as a formal version and prepare to implement it in the power carrier communication system. Apply the updated control strategy in the power carrier communication system, which may involve modifying the system configuration, updating software or hardware, etc. Before implementing the updated control strategy, ensure that the key data and configuration of the system have been backed up so that they can be restored when needed. Enable the monitoring system of the power carrier communication system to monitor the status, performance, and interference of the system in real time. Ensure that the monitoring system can accurately collect and analyze data and can generate clear reports and alarms. Obtain the monitoring results of the power carrier communication system from the monitoring system, including system performance data, interference detection data, and any related alarm or error messages. Based on the monitoring results, analyze whether the updated control strategy has a positive impact on the performance and anti-interference ability of the power carrier communication system. If the updated control strategy is effective, continue to monitor the operation of the system and make further optimizations and adjustments as needed. If the updated control strategy is ineffective or ineffective, re-evaluate the compensation results and update the control strategy, and consider taking other measures to solve the problem.

[0058] A preset period is set to check the power carrier communication system for potential safety hazards. If any potential safety hazards exist, the potential safety hazards are sent to the staff for processing.

[0059] According to the actual operation of the power carrier communication system, determine a suitable preset cycle for troubleshooting safety hazards. This cycle can be set according to the stability, importance and frequency of historical problems of the system, such as once a week, month or quarter. According to the preset cycle, formulate a detailed safety hazard troubleshooting plan. The plan should clearly specify the time, scope, content, method and responsible person of the troubleshooting, and the troubleshooting content may include the status of hardware equipment, the security of software system, the stability of network connection, the reliability of data transmission and other aspects. According to the troubleshooting plan, conduct a comprehensive safety hazard troubleshooting of the power carrier communication system. Professional safety detection tools and technical means can be used to carefully check all aspects of the system. Record all problems found during the troubleshooting process, including the specific manifestations, possible causes and potential impacts of safety hazards. Evaluate the safety hazards found during the troubleshooting to determine their severity and urgency. The evaluation can be based on factors such as the possible consequences, scope of impact and difficulty of handling of safety hazards. Sort the safety hazards according to priority so that targeted measures can be taken in subsequent processing. Once a safety hazard is found, send the relevant information to the responsible staff or security team immediately. The notification content should include the description, location, evaluation results and handling suggestions of the safety hazard. Ensure that staff are notified in a timely manner and understand the severity and urgency of safety hazards.

[0060] A preset period is set to conduct safety hazard inspection and risk point improvement re-inspection on the power carrier communication system, wherein the safety hazard includes management hazard and technical hazard; if there is a safety hazard, the hazard type is specifically analyzed and countermeasures are taken; and the hazard is handled based on the countermeasures.

[0061] The preset cycle should be set according to the complexity and importance of the system. Usually, a comprehensive safety hazard inspection is conducted every quarter, a special inspection of key areas or key equipment is conducted every month, and the improved risk points are re-inspected in a timely manner. Safety hazard inspection should cover two aspects: management hazards and technical hazards. Management hazards include checking document integrity, inspection and regular inspection, spare fiber core detection, etc. Technical hazards include electromagnetic leakage protection, checking whether the system has taken electromagnetic leakage protection measures to ensure information security; electromagnetic radiation management, for equipment that may generate large electromagnetic radiation such as long-distance carrier telephones, check whether its radiation management is proper to avoid information leakage; equipment status, check the status of hardware equipment such as communication equipment and cables to ensure that they are operating normally. After discovering safety hazards, a specific analysis of the hazard type should be conducted. For example, management hazards may involve improper document management, unreasonable inspection cycles, lax password management, etc.; technical hazards may involve electromagnetic leakage, excessive electromagnetic radiation, equipment failure, etc. Based on the response measures, the discovered hazards are dealt with in a timely manner. For management hazards, they can be improved by improving systems and strengthening training; for technical hazards, they can be solved by replacing equipment and adjusting parameters. At the same time, timely re-inspections should be conducted for the improved risk points to ensure that the hidden dangers are completely eliminated. Safety hidden danger detection and hidden danger treatment is a continuous process. During the detection and treatment process, we should constantly summarize experience and lessons, improve the detection standards and response measures, and improve the efficiency and quality of detection and treatment.

[0062] This embodiment provides an anti-interference method for power carrier communication, which outputs interference detection results and interference prediction results by building a first interference detection model and a second interference prediction model; establishes an interference control strategy library based on historical data, traverses the interference control strategy library based on the interference detection results, and outputs the interference control strategy; builds an interference control compensation model, outputs the compensation result based on the interference detection result, and updates the interference control strategy based on the compensation result to obtain the updated strategy to adjust the power carrier communication system. It solves the problem that power lines are susceptible to environmental interference, which in turn has a serious impact on power carrier communication, thereby reducing the quality of signal transmission. Through the detection and prediction of power carrier signals, accurate identification of interference sources is achieved, the anti-interference control strategy is optimized, and the accuracy and reliability of data transmission are improved.

[0063] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0064] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0065] In the present invention, unless otherwise clearly specified and limited, when a first feature is “on” or “below” a second feature, it may be that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Moreover, when a first feature is “above”, “above” or “above” a second feature, it may be that the first feature is directly above or obliquely above the second feature, or it may simply mean that the first feature is higher in level than the second feature. When a first feature is “below”, “below” or “below” a second feature, it may be that the first feature is directly below or obliquely below the second feature, or it may simply mean that the first feature is lower in level than the second feature.

[0066] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0067] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may alter, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. An anti-interference method for power carrier communication, characterized in that: include: S11, acquiring a first power carrier signal on a target power line in real time through a preset signal collector, inputting the first power carrier signal into a preset first interference detection model, and obtaining an interference detection result of the first power carrier signal; S12, judging whether there is interference in the first power carrier signal according to the interference detection result of the first power carrier signal; If there is interference in the first power carrier signal, the first power carrier signal is input into a preset second interference prediction model to determine whether there is a risk of interference in the future; S13. If there is an interference risk in the future, the first interference control strategy corresponding to the interference detection result is matched according to the interference detection result of the first power carrier signal and a pre-set interference control strategy library, so as to control the first power carrier signal in real time and reduce the risk of signal interference.

2. The anti-interference method for power carrier communication according to claim 1, characterized in that: The S11 and before also include: S10, acquiring the communication signal on the target power line in real time, performing signal amplification and signal filtering processing on the acquired communication signal on the target power line, and converting the communication signal into a corresponding digital signal; Based on a preset digital signal processing algorithm and an adaptive equalization algorithm, the digital signal is frequency-converted and compensated to obtain a first power carrier signal corresponding to the communication signal.

3. The anti-interference method for power carrier communication according to claim 1, characterized in that: The first interference detection model includes a decision tree sub-model and an interference source position detection sub-model; Then the S11 includes: S11-1, performing baseband modulation on the received first power carrier signal based on a preset baseband modulation algorithm to obtain a corresponding time domain signal; According to the time domain signal corresponding to the first power carrier signal and the preset formula 1, the frequency domain signal corresponding to the first power carrier signal is obtained; the formula 1 is: X freq =Re(FFT(X time ,dim=2)); Among them, X freq is the frequency domain signal corresponding to the first power carrier signal, FFT is fast Fourier transform, Re represents the real part of the complex number, dim=2 represents the operation in the second dimension, X time is a time domain signal corresponding to the first power carrier signal; S11-2, extracting features from a frequency domain signal and a time domain signal corresponding to the first power carrier signal to obtain feature parameters corresponding to the first power carrier signal; S11-3, inputting characteristic parameters corresponding to the first power carrier signal into a preset decision tree sub-model to obtain interference type and interference intensity corresponding to the first power carrier signal; S11-4. Input the interference type and interference intensity corresponding to the first power carrier signal into a preset interference source position detection sub-model to obtain an interference detection result corresponding to the first power carrier signal.

4. The anti-interference method for power carrier communication according to claim 3, characterized in that: The S11-3 includes: Obtaining a high-dimensional feature vector data set of the first power carrier signal according to the feature parameters corresponding to the first power carrier signal; Inputting a high-dimensional feature vector data set corresponding to the first power carrier signal into a pre-set decision tree sub-model to obtain an interference type and interference intensity corresponding to the first power carrier signal; Wherein, the characteristic parameters include one or more of the following: time domain skewness, time domain kurtosis, normalized instantaneous amplitude absolute value standard deviation, normalized instantaneous frequency absolute value standard deviation, time domain envelope fluctuation, fast pulse modulation coefficient, box dimension, information dimension, frequency domain skewness, frequency domain kurtosis, frequency domain envelope fluctuation, additive Gaussian white noise factor, carrier factor, spectrum information entropy, spectrum index entropy, spectrum norm entropy, low-frequency approximate coefficient energy proportion, high-frequency detail coefficient energy proportion, wavelet energy spectrum entropy, wavelet variance, bispectral entropy, contour-integrated bispectral mean, contour-integrated bispectral variance and bispectral waveform entropy; Then the pre-set decision tree sub-model includes: a model obtained by training the decision tree sub-model using a first sample training set; the first sample training set includes: a high-dimensional feature vector data set corresponding to at least one second power carrier signal, and an interference type and interference intensity corresponding to each high-dimensional feature vector data set; The second power carrier signal is a historical power carrier signal of the target power line.

5. The anti-interference method for power carrier communication according to claim 3, characterized in that: The S11-4 includes: According to the interference intensity corresponding to the first power carrier signal and the preset formula 2, the propagation distance of the interference signal corresponding to the first power carrier signal is obtained; the formula 2 is: d×A=20log 10 (d)+20log 10 (f)+20log 10 (4π / c); Wherein, d is the propagation distance of the interference signal corresponding to the first power carrier signal, A is the preset attenuation, f is the signal frequency, and c is the speed of light; Obtaining the interference source location range according to the propagation distance of the interference signal corresponding to the first power carrier signal and the preset position coordinates of the signal collector; Based on the pre-set power line network topology map, all interference sources within the interference source location range are obtained; According to the interference type corresponding to the first power carrier signal, all interference sources within the interference source position range are screened to obtain the interference source position corresponding to the first power carrier signal; The interference detection result includes interference type, interference intensity and interference source location.

6. The anti-interference method for power carrier communication according to claim 3, characterized in that: The S12 includes: Determining whether there is interference in the first power carrier signal according to the interference detection result of the first power carrier signal; If there is interference in the first power carrier signal, the characteristic parameters are classified according to the characteristic attributes of each characteristic vector in the characteristic parameters corresponding to the first power carrier signal to obtain at least one characteristic block; the characteristic block includes one or more of the following: a time domain characteristic block, a fractal domain characteristic block, a frequency domain characteristic block, a wavelet domain characteristic block or a dual spectrum domain characteristic block; Based on the feature attribute corresponding to each feature block and the first linear transformation matrix preset for each feature attribute, and the preset formula three, each feature block is mapped to the target dimension; the formula three is: E i =a i ×patch i ; Where E is the feature block after linear transformation, patch is the feature block, α is the first linear transformation matrix pre-set for each feature attribute, and i is the index of the feature block; According to the preset attention mechanism, the attention matrix corresponding to each feature block is obtained; According to the attention matrix corresponding to each feature block, it is determined whether there is interference risk in the future.

7. The anti-interference method for power carrier communication according to claim 6, characterized in that: The step S12, according to a preset attention mechanism, obtains an attention matrix corresponding to each feature block, including: According to the preset attention mechanism and the preset attention allocation weight of each feature attribute, and the preset formula 4, the attention matrix corresponding to each feature block is obtained; the formula 4 is: ； Among them, Q is the query vector, K is the key vector, and V is the value vector. is a preset scaling factor, T is a preset transpose, i is the index of the feature block, and β is the attention allocation weight corresponding to the feature attribute.

8. The anti-interference method for power carrier communication according to claim 6, characterized in that: S12, judging whether there is interference risk in the future according to the attention matrix corresponding to each feature block, includes: The attention matrix corresponding to each feature block is concatenated with the original features corresponding to the feature block to obtain the fused features corresponding to each feature block; According to the fusion features corresponding to each feature block and the preset attention score weight matrix, and the preset formula 5, the attention score corresponding to each feature block is obtained; the formula 5 is: S i =C i c score ; Among them, S is the attention score corresponding to the feature block, C is the fusion feature corresponding to the feature block, γ score is the attention score weight matrix; The attention scores corresponding to each feature block are normalized by a preset softmax function to generate the attention weights corresponding to each feature block; According to the attention weight and fusion feature corresponding to each feature block, and the preset formula 6, the global feature corresponding to the first power carrier signal is obtained; the formula 6 is: ； Among them, C is the fusion feature corresponding to the feature block, B is the attention weight corresponding to the feature block, c is the global feature corresponding to the first power carrier signal, n is the number of feature blocks, and i is the index of the feature block; Whether there is an interference risk in the future is determined according to the global characteristics corresponding to the first power carrier signal.

9. The anti-interference method for power carrier communication according to claim 1, characterized in that: The S11 and before also include: S01. Collect a second power carrier signal corresponding to a target power line, and extract features of the second power carrier signal to obtain feature parameters corresponding to the second power carrier signal; Obtaining the interference type and interference intensity corresponding to the second power carrier signal according to the characteristic parameters corresponding to the second power carrier signal; According to the interference type and interference intensity corresponding to the second power carrier signal, match the second interference control strategy corresponding to each preset interference type and / or interference intensity, and establish an interference control strategy library; The second power carrier signal is a historical power carrier signal of the target power line.

10. The anti-interference method for power carrier communication according to claim 1, characterized in that: The S13 further includes: S14, inputting the interference detection result corresponding to the first power carrier signal into a preset interference control compensation model to obtain a compensation result corresponding to the first power carrier signal; updating the first interference control strategy based on the compensation result so as to control the first power carrier signal of the power supply in real time and reduce the risk of signal interference; The preset interference control compensation model includes: a compensation model obtained by training the interference control compensation model using a second sample training set; the second sample training set includes: an interference detection result corresponding to at least one second power carrier signal, and a compensation result corresponding to each interference detection result; The second power carrier signal is a historical power carrier signal of the target power line.

Images