A carrier communication method and device based on noise stripping optimization

Abstract

The application discloses a carrier communication method and device based on noise stripping optimization, which extracts information flow fingerprints and energy flow fingerprints of a carrier to be processed to obtain the two kinds of fingerprints, calculates the similarity coefficient of the carrier to be processed and each standard noise in a standard noise library according to the two kinds of fingerprints, and determines the noise source in the carrier according to the similarity coefficient; a noise stripping network is generated according to the two kinds of fingerprints to strip the noise in the carrier, a prediction model is generated according to the noise source to predict the noise, and the carrier communication equipment is optimized according to the prediction result. The application determines the noise source in the carrier signal and strips the noise through the cooperation of information flow fingerprints and energy flow fingerprints, improves the accuracy of the judgment result and the stripping result, optimizes the carrier communication equipment based on the two kinds of fingerprints, realizes the cooperative optimization of information flow-energy flow of the carrier communication equipment, and improves the improvement effect of the health state of the carrier communication equipment.

Description

Technical Field

[0001] This invention relates to the field of carrier communication technology, and in particular to a carrier communication method and apparatus based on noise stripping optimization. Background Technology

[0002] Power line carrier (PLC) communication, as a crucial component of modern power system communication networks, has been widely adopted due to its economic efficiency and reliability. However, with the increasing use of power electronic equipment, noise interference during PLC signal transmission has become increasingly severe. Therefore, accurately identifying the types of equipment sources causing PLC noise, predicting their health status in a targeted manner, and optimizing PLC communication methods to ensure high-quality PLC communication and the safe and reliable operation of the power grid are urgent problems that need to be addressed.

[0003] Existing methods for identifying noise sources in power line carrier signals primarily involve signal processing of the power line carrier signal and noise testing. However, these methods do not fully consider the impact of both the information flow and energy flow of the power line carrier signal, nor do they address the issue of subsequent noise reduction. Summary of the Invention

[0004] This invention provides a carrier communication method and apparatus based on noise stripping optimization, which achieves the technical effect of determining the noise source of power line carrier signals from both the perspectives of information flow and energy flow, and simultaneously reducing the noise of the signals.

[0005] To address the aforementioned technical problems, embodiments of the present invention provide a carrier communication method based on noise stripping optimization, comprising the following steps:

[0006] A carrier to be processed is acquired, and information flow fingerprint extraction and energy flow fingerprint extraction are performed on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed.

[0007] Calculate the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard noises in a preset standard noise library, and determine the noise source in the carrier to be processed based on the similarity coefficients;

[0008] A corresponding noise stripping network is generated based on the information flow fingerprint and the energy flow fingerprint. The noise stripping network performs noise stripping on the carrier to be processed. At the same time, a corresponding prediction model is generated based on the noise source to perform noise prediction and obtain the corresponding prediction result. The corresponding carrier communication equipment is optimized based on the prediction result.

[0009] The carrier communication method provided by this invention, after acquiring the carrier to be processed, first extracts fingerprints of both information flow and energy flow from the carrier, thereby obtaining the information flow fingerprint and energy flow fingerprint of the carrier. On the one hand, this provides data support for subsequently identifying the source device of noise in the carrier to be processed; on the other hand, the extraction of these two fingerprint information also enriches the reference data for the system's subsequent judgment of noise sources in the carrier to be processed, improving the accuracy, completeness, and scientific nature of noise source judgment.

[0010] After obtaining the two fingerprints mentioned above, the system filters and matches them in the standard fingerprint database stored in the system to determine the source of noise in the carrier to be processed. By calculating the similarity coefficient between the two fingerprints and the standard noise corresponding to each device in the standard fingerprint database, the source of noise in the carrier to be processed is determined based on the calculated similarity coefficient. This provides a data basis for subsequent noise stripping and communication equipment optimization for the carrier.

[0011] After identifying the noise source, the system first generates a noise stripping network based on two fingerprints to strip the noise from the carrier to be processed. At the same time, it generates a prediction model based on the identified noise source to predict the noise of the carrier signal and obtain the corresponding prediction results. Based on the prediction results, the system optimizes the equipment corresponding to the noise source in the carrier communication equipment, thereby improving the communication efficiency and quality of the carrier communication equipment and enhancing the health status of the carrier communication equipment.

[0012] As a preferred example, the step of performing information flow fingerprint extraction processing and energy flow fingerprint extraction processing on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed specifically includes:

[0013] The carrier to be processed is subjected to differential operation by calling a preset differential constellation trajectory diagram, and the obtained differential operation result is used as the information flow fingerprint;

[0014] Simultaneously, a preset Hilbert transform algorithm is invoked to obtain the current analysis signal of the carrier to be processed, and the current analysis signal is subjected to feature extraction of amplitude, phase and frequency respectively, and the feature extraction results are used as the energy flow fingerprint.

[0015] To improve the accuracy of noise source identification in the carrier wave to be processed, the carrier communication method provided by this invention extracts the information flow fingerprint and energy flow fingerprint in the carrier wave to be processed. The system uses a differential constellation trajectory diagram to perform differential operations on the carrier signal to be processed for information flow extraction. The result of the differential operation is the extracted information flow fingerprint. In this way, the information flow fingerprint, which is difficult to measure directly, can be transformed into a differential constellation trajectory diagram that can be intuitively represented, so that the system can perform further calculations on the information flow fingerprint data.

[0016] For energy flow fingerprint extraction of the carrier wave to be processed, this invention obtains the current analytical signal of the carrier wave through the Hilbert transform algorithm, and extracts features in three aspects: amplitude, phase, and frequency. The extracted result is the energy flow fingerprint. The energy flow fingerprint obtained by the above method provides a data foundation for subsequent noise signal stripping in the carrier wave.

[0017] As a preferred example, the step of calculating the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard noises in a preset standard noise library, and determining the noise source in the carrier to be processed based on the similarity coefficients, specifically involves:

[0018] The differences between the information flow fingerprint and the energy flow fingerprint and the information flow standard fingerprint and energy flow standard fingerprint corresponding to the several types of standard noise in the standard noise library are calculated respectively, and then several pairs of information flow differences and energy flow differences are obtained;

[0019] The sum of squares of the information flow difference and energy flow difference for each pair is calculated, and the reciprocal of the calculated sum of squares is used as the similarity coefficient of the corresponding standard noise. Then, based on the calculated similarity coefficient, a preset maximum likelihood estimation method is called to determine the noise source of the carrier to be processed, and the noise source is obtained.

[0020] To further improve the accuracy of noise source identification in the carrier under test, the carrier communication method provided by the present invention calculates the difference between the information flow fingerprint and energy flow fingerprint extracted from the carrier under test and the information flow standard fingerprint and energy flow standard fingerprint corresponding to each standard noise in the standard noise library, thereby obtaining several pairs of differences. Each pair of differences includes an information flow difference and an energy flow difference. Each pair of differences also corresponds to a standard noise, and each standard noise also corresponds to a device that generates the noise, i.e., the noise source.

[0021] After determining several pairs of differences, the system performs a series of calculations on each pair to determine the similarity coefficient between the carrier to be processed and each standard noise. Then, using the maximum likelihood estimation method, it determines the specific device from which the noise in the carrier to be processed originates, i.e., it identifies the noise source. The accuracy and reliability of identifying the noise source by combining the information flow data and energy flow data in the aforementioned carrier signal are effectively improved. Furthermore, the collaborative analysis of noise sources using these two types of extracted information further enhances the completeness and accuracy of subsequent noise stripping from the carrier.

[0022] As a preferred example, the step of generating a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, so that the noise stripping network performs noise stripping processing on the carrier to be processed, specifically involves:

[0023] Feature extraction and splicing are performed on the standard noise in the preset standard noise library, the information flow fingerprint, and the energy flow fingerprint. The splicing result is then used as input data to the preset initial feature processing network, so that the initial feature processing network is updated to the noise stripping network.

[0024] The noise stripping network is controlled to perform feature processing and analysis on the input data, output corresponding sequence data as a noise stripping sequence, and perform noise stripping processing on the carrier to be processed according to the noise stripping sequence.

[0025] To further improve the integrity and accuracy of noise stripping, the carrier communication method provided by this invention first performs feature extraction and splicing on the standard noise in the standard noise library and the extracted information flow fingerprint and energy flow fingerprint when stripping noise, and inputs the splicing result into the feature processing network preset by the system, so that the feature processing network is updated into a noise stripping network.

[0026] After the network update, the system controls the updated noise stripping network to perform feature processing and analysis on the feature splicing results (i.e., the input data), and outputs corresponding noise stripping sequence data. Once this sequence data is obtained, the system can perform noise stripping on the carrier to be processed based on it. By performing noise stripping on the carrier to be processed in this way, the accuracy of the stripping and the completeness of the stripped noise can be improved. Furthermore, the noise stripping sequence data obtained through this method can also serve as reference data for subsequent system optimization of the carrier communication equipment.

[0027] As a preferred example, the step of generating a corresponding prediction model based on the noise source to perform noise prediction and obtain the corresponding prediction result specifically involves:

[0028] The preset equipment digital model is updated according to the noise source to obtain the corresponding prediction model, and the equipment operating noise is predicted according to the prediction model to obtain the corresponding noise prediction sequence.

[0029] Based on the noise prediction sequence and the noise stripping sequence, the health status of the device is predicted by the prediction model to obtain the prediction result of the health status of the corresponding carrier communication device.

[0030] In order to improve the future operation and communication quality of carrier communication equipment, the carrier communication method provided by the present invention, after removing noise from the carrier signal, will update the preset digital model in the system according to the determined noise source to obtain a prediction model for predicting noise in the carrier, and predict the operating noise of the carrier communication equipment according to the prediction model to obtain the corresponding noise prediction sequence data.

[0031] After obtaining the noise prediction sequence, the system predicts the health status of the carrier communication equipment based on the noise prediction sequence and the noise stripping sequence, and obtains the corresponding prediction results, providing reference data for subsequent optimization of the carrier communication equipment.

[0032] As a preferred example, optimizing the corresponding carrier communication equipment based on the prediction result specifically includes:

[0033] The prediction results are compared with the preset first threshold and the second threshold respectively to obtain the corresponding comparison results;

[0034] When the comparison result is that the predicted result is greater than or equal to the second threshold, the health status of the carrier communication device is determined to be excellent, and the status log of the carrier communication device is updated according to the health status.

[0035] When the comparison result is that the predicted result is greater than the first threshold and less than the second threshold, the health status of the carrier communication device is determined to be good. The carrier communication device is then adjusted to online maintenance status, and the communication line of the carrier communication device is optimized.

[0036] When the comparison result is less than or equal to the first threshold, the health status of the carrier communication device is determined to be extremely poor. The carrier communication device is then adjusted to an offline maintenance state, and the communication line of the carrier communication device is optimized.

[0037] To further improve the timeliness and accuracy of optimization for carrier communication devices, the carrier communication method provided by this invention compares the prediction results obtained from the above preferred examples with the first threshold and the second threshold respectively to obtain the corresponding comparison results. Based on the corresponding comparison results, the method selects the optimization method that is more in line with the real-time status of the device to optimize the communication line of the device. This allows the device to improve the communication quality and efficiency of the carrier after the communication line optimization, and also improves the health status of the carrier communication device while improving the carrier communication quality.

[0038] Accordingly, the present invention also provides a carrier communication device based on noise stripping optimization, the carrier communication device including a fingerprint extraction module, a source determination module, a noise stripping module and a device optimization module;

[0039] The fingerprint extraction module is used to acquire the carrier to be processed and to perform information flow fingerprint extraction and energy flow fingerprint extraction processing on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed.

[0040] The source determination module is used to calculate the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard fingerprints in a preset standard fingerprint library, and to determine the noise source in the carrier to be processed based on the similarity coefficients;

[0041] The noise stripping module is used to generate a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, so that the noise stripping network performs noise stripping processing on the carrier to be processed.

[0042] The device optimization module is used to generate a corresponding prediction model based on the noise source, perform noise prediction, obtain the corresponding prediction results, and optimize the corresponding carrier communication equipment based on the prediction results.

[0043] As a preferred example, the fingerprint extraction module performs information flow fingerprint extraction processing and energy flow fingerprint extraction processing on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed, specifically including:

[0044] The carrier to be processed is subjected to differential operation by calling a preset differential constellation trajectory diagram, and the obtained differential operation result is used as the information flow fingerprint;

[0045] Simultaneously, a preset Hilbert transform algorithm is invoked to obtain the current analysis signal of the carrier to be processed, and the current analysis signal is subjected to feature extraction of amplitude, phase and frequency respectively, and the feature extraction results are used as the energy flow fingerprint.

[0046] As a preferred example, the source determination module calculates the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard fingerprints in a preset standard fingerprint library, and determines the noise source in the carrier to be processed based on the similarity coefficients, specifically as follows:

[0047] The differences between the information flow fingerprint and the energy flow fingerprint and the information flow standard fingerprint and the energy flow standard fingerprint corresponding to the several types of standard fingerprints in the standard fingerprint library are calculated respectively, and then several pairs of information flow differences and energy flow differences are obtained;

[0048] The sum of squares of the information flow difference and energy flow difference for each pair is calculated, and the reciprocal of the calculated sum of squares is used as the similarity coefficient of the corresponding standard fingerprint. Then, based on the calculated similarity coefficient, a preset maximum likelihood estimation method is called to determine the noise source of the carrier to be processed, and the noise source is obtained.

[0049] As a preferred example, the noise stripping module generates a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, enabling the noise stripping network to perform noise stripping processing on the carrier to be processed, specifically as follows:

[0050] Feature extraction and splicing are performed on the standard noise in the preset standard noise library, the information flow fingerprint, and the energy flow fingerprint. The splicing result is then used as input data to the preset initial feature processing network, so that the initial feature processing network is updated to the noise stripping network.

[0051] The noise stripping network is controlled to perform feature processing and analysis on the input data, output corresponding sequence data as a noise stripping sequence, and perform noise stripping processing on the carrier to be processed according to the noise stripping sequence. Attached Figure Description

[0052] Figure 1 : A flowchart illustrating an embodiment of the carrier communication method based on noise stripping optimization provided by the present invention;

[0053] Figure 2 : This is a schematic diagram of an embodiment of the noise stripping network for power line carrier provided by the present invention;

[0054] Figure 3 : A schematic diagram of an embodiment of the carrier communication device based on noise stripping optimization provided by the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Example 1

[0057] Please refer to Figure 1 This is a flowchart illustrating an embodiment of the carrier communication method based on noise stripping optimization provided by the present invention, including steps 101 to 104, each step specifically including:

[0058] Step 101: Obtain the carrier to be processed, and perform information flow fingerprint extraction and energy flow fingerprint extraction on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed.

[0059] The carrier communication method provided in this invention, after acquiring the carrier to be processed, first extracts fingerprints of both information flow and energy flow from the carrier, thereby obtaining the information flow fingerprint and energy flow fingerprint of the carrier. On the one hand, this provides data support for subsequently identifying the source device of noise in the carrier to be processed; on the other hand, the extraction of these two fingerprint information also enriches the reference data for the system's subsequent judgment of the noise source in the carrier to be processed, improving the accuracy, completeness, and scientific nature of the noise source judgment.

[0060] In this embodiment, because the existing technology does not fully consider the fusion of power line carrier information flow and energy flow, errors occur in noise stripping and noise source identification of power line carriers, resulting in poor accuracy. Therefore, to improve the accuracy of noise stripping and noise source identification for power line carriers, this embodiment extracts information flow fingerprints and energy flow fingerprints upon receiving the power line carrier to be processed. Noise source identification and noise stripping are then performed by comprehensively considering the extracted information flow fingerprints and energy flow fingerprints for collaborative analysis and identification.

[0061] For example, the information flow fingerprint extraction and energy flow fingerprint extraction processes performed on the carrier to be processed in this embodiment to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed specifically include:

[0062] The carrier to be processed is subjected to differential operation by calling a preset differential constellation trajectory diagram, and the obtained differential operation result is used as the information flow fingerprint;

[0063] Simultaneously, a preset Hilbert transform algorithm is invoked to obtain the current analysis signal of the carrier to be processed, and the current analysis signal is subjected to feature extraction of amplitude, phase and frequency respectively, and the feature extraction results are used as the energy flow fingerprint.

[0064] To improve the accuracy of noise source identification in the carrier wave to be processed, the carrier communication method provided in this embodiment of the invention extracts the information flow fingerprint and energy flow fingerprint in the carrier wave to be processed. The system uses a differential constellation trajectory diagram to perform differential operations on the carrier wave signal to be processed. The result of the differential operation is the extracted information flow fingerprint. In this way, the information flow fingerprint, which is difficult to measure directly, can be transformed into a differential constellation trajectory diagram that can be intuitively represented, so that the system can perform further calculations on the information flow fingerprint data.

[0065] In this embodiment, a differential constellation trajectory diagram is used to perform differential operations on the power line carrier signal to obtain the information flow fingerprint at any given time. The specific calculation formula used is as follows:

[0066]

[0067] Where, d i I(t) is the information flow fingerprint of the power line carrier signal at time t; I(t) is the power line carrier received signal at time t; y(t) is the power line carrier transmitted signal at time t. denoted as the sampling interval for differential operations; l is the index of the channel multipath component; Δ is the carrier frequency difference between the transmitter and receiver; W is the carrier signal bandwidth; r is the carrier channel response; (·) * For conjugate operations on complex numbers.

[0068] As can be seen from the above formula, the result of the signal differential operation mainly contains two components: the result of the differential operation of the transmitted signal. and phase rotation factor For power line carrier signals, the results of the above differential operations are equivalent to a finite number of discrete values, corresponding to a finite number of positions on the complex plane. However, because power line carrier signals are affected by noise interference from power electronic devices (i.e., carrier communication equipment) during transmission, the differential operation results will exhibit some dispersion around their theoretical positions. Therefore, by plotting the differential operation results of the above calculation formula on the complex plane, and then discretizing the complex two-dimensional plane into a series of pixels, and representing the grayscale value of a pixel region by the number of differential operation results falling into that region, a corresponding differential constellation trajectory map can be formed. In this way, the information flow fingerprint, which is difficult to measure directly, can be transformed into a differential constellation trajectory map that can be intuitively represented. This allows users to more intuitively view the information flow fingerprint extracted by the system, and also facilitates further calculations by the system based on the extracted information flow fingerprint.

[0069] To extract the energy flow fingerprint of the carrier wave to be processed, this embodiment of the invention obtains the current analytical signal of the carrier wave through the Hilbert transform algorithm, and then extracts features in three aspects: amplitude, phase, and frequency. The extracted result is the energy flow fingerprint. The energy flow fingerprint obtained by the above method provides a data foundation for subsequent noise signal stripping in the carrier wave.

[0070] In this embodiment, in order to make full use of the current complex frequency domain information, the specific calculation formula for obtaining the current analytical signal at time t using the Hilbert transform algorithm is as follows:

[0071]

[0072] in, Let i(t) be the current analytical signal at time t; i(t) is the electric field line current signal at time t.

[0073] After obtaining the current analysis signal of the power line carrier at time t, the system will extract the energy flow fingerprint at time t based on the obtained current analysis signal, using three dimensions: amplitude, phase, and frequency. Specifically, the amplitude is measured by the magnitude of the current analysis signal, the phase by the complex frequency domain angle of the current analysis signal, and the frequency by the rate of change of the imaginary part of the current analysis signal in radians. The specific calculation formulas are as follows:

[0074]

[0075] Where, d e (t) represents the energy flow fingerprint of the power line carrier signal at time t; the first term of the formula represents the instantaneous energy amplitude characteristic of the current analytical signal in the time domain, the second term represents the instantaneous energy phase characteristic in the time domain, and the third term represents the instantaneous energy frequency characteristic in the time domain. Similarly, the above calculation formula can be simplified to:

[0076]

[0077] The simplified formula above has the same meaning as the original formula.

[0078] Step 102: Calculate the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard noises in the preset standard noise library, and determine the noise source in the carrier to be processed based on the similarity coefficients.

[0079] After obtaining the two fingerprints mentioned above, the system filters and matches them in the standard fingerprint database stored in the system to determine the source of noise in the carrier to be processed. By calculating the similarity coefficient between the two fingerprints and the standard noise corresponding to each device in the standard fingerprint database, the source of noise in the carrier to be processed is determined based on the calculated similarity coefficient. This provides a data basis for subsequent noise stripping and communication equipment optimization for the carrier.

[0080] For example, in this embodiment, the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard noises in a preset standard noise library are calculated respectively, and the noise source in the carrier to be processed is determined based on the similarity coefficients. Specifically:

[0081] The differences between the information flow fingerprint and the energy flow fingerprint and the information flow standard fingerprint and energy flow standard fingerprint corresponding to the several types of standard noise in the standard noise library are calculated respectively, and then several pairs of information flow differences and energy flow differences are obtained;

[0082] The sum of squares of the information flow difference and energy flow difference for each pair is calculated, and the reciprocal of the calculated sum of squares is used as the similarity coefficient of the corresponding standard noise. Then, based on the calculated similarity coefficient, a preset maximum likelihood estimation method is called to determine the noise source of the carrier to be processed, and the noise source is obtained.

[0083] To further improve the accuracy of noise source identification in the carrier under test, the carrier communication method provided by the present invention calculates the difference between the information flow fingerprint and energy flow fingerprint extracted from the carrier under test and the information flow standard fingerprint and energy flow standard fingerprint corresponding to each standard noise in the standard noise library, thereby obtaining several pairs of differences. Each pair of differences includes an information flow difference and an energy flow difference. Each pair of differences also corresponds to a standard noise, and each standard noise also corresponds to a device that generates the noise, i.e., the noise source.

[0084] In this embodiment, after the system extracts the information flow fingerprint and energy flow fingerprint of the carrier to be processed, it calculates the similarity coefficient between these fingerprints and the standard fingerprints of J types of power electronic devices (i.e., carrier communication devices) in the power electronic device fingerprint database, i.e., the standard fingerprint database. The similarity coefficient is defined as the reciprocal of the sum of the squared differences between the extracted information flow fingerprint, energy flow fingerprint, and the standard fingerprint of the j-th device at time t. If the similarity coefficient with the standard fingerprint of the j-th device in the power electronic device fingerprint database is set to δ... j If (t), then the formula for calculating the similarity coefficient is as follows:

[0085]

[0086] in, These are the standard information flow fingerprint and energy flow fingerprint of the j-th device in the power electronic equipment fingerprint database, respectively.

[0087] After determining several pairs of differences, the system performs a series of calculations on each pair to determine the similarity coefficient between the carrier to be processed and each standard noise. Then, using the maximum likelihood estimation method, it determines the specific device from which the noise in the carrier to be processed originates, i.e., it identifies the noise source. The accuracy and reliability of identifying the noise source by combining the information flow data and energy flow data in the aforementioned carrier signal are effectively improved. Furthermore, the collaborative analysis of noise sources using these two types of extracted information further enhances the completeness and accuracy of subsequent noise stripping from the carrier.

[0088] In this embodiment, after calculating the similarity coefficient between each standard fingerprint and the extracted information flow fingerprint and energy flow fingerprint, the system can determine the noise source device based on the maximum likelihood estimation method. The specific implementation formula is as follows:

[0089]

[0090] Where, j * Indicates the determination result of the noise source device; δ j (t) represents the similarity coefficient between the power electronic device and the standard fingerprint of the j-th device in the fingerprint database. After completing the above calculation, the noise source device in the carrier wave to be processed can be determined.

[0091] Step 103: Generate a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, so that the noise stripping network performs noise stripping processing on the carrier to be processed.

[0092] After identifying the noise source, the system first uses two fingerprint-generated noise stripping networks to strip noise from the carrier to be processed.

[0093] For example, in this embodiment, generating a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, and enabling the noise stripping network to perform noise stripping processing on the carrier to be processed, specifically involves:

[0094] Feature extraction and splicing are performed on the standard noise in the preset standard noise library, the information flow fingerprint, and the energy flow fingerprint. The splicing result is then used as input data to the preset initial feature processing network, so that the initial feature processing network is updated to the noise stripping network.

[0095] The noise stripping network is controlled to perform feature processing and analysis on the input data, output corresponding sequence data as a noise stripping sequence, and perform noise stripping processing on the carrier to be processed according to the noise stripping sequence.

[0096] To further improve the integrity and accuracy of noise stripping, the carrier communication method provided in this embodiment of the invention first performs feature extraction and splicing on the standard noise in the standard noise library and the extracted information flow fingerprint and energy flow fingerprint when stripping noise, and inputs the splicing result into the feature processing network preset by the system, so that the feature processing network is updated into a noise stripping network.

[0097] See also Figure 2 , Figure 2 This is a schematic diagram of an embodiment of the noise stripping network for power line carrier provided by the present invention. Figure 2 As shown, the power line carrier noise stripping network comprises three parts: an input layer, a fingerprint and noise feature processing layer, and an output layer. The input layer receives network input data, including the extracted information flow fingerprint, energy flow fingerprint, and a standard noise library. These three types of input information are concatenated after feature extraction and then input into the power line carrier noise stripping network. The fingerprint and noise feature processing layer is responsible for performing feature analysis and processing on the input information flow fingerprint, energy flow fingerprint, and standard noise library, and transmitting the analysis results to the output layer.

[0098] After the network update, the system controls the updated noise stripping network to perform feature processing and analysis on the feature splicing results (i.e., the input data), and outputs corresponding noise stripping sequence data. Once this sequence data is obtained, the system can perform noise stripping on the carrier to be processed based on it. By performing noise stripping on the carrier to be processed in this way, the accuracy of the stripping and the completeness of the stripped noise can be improved. Furthermore, the noise stripping sequence data obtained through this method can also serve as reference data for subsequent system optimization of the carrier communication equipment.

[0099] In this embodiment, the input data is analyzed and processed using power line carrier wave analysis to obtain the model fingerprint and features. The resulting data, output through the output layer, is the noise stripping sequence data, specifically an N-point noise stripping sequence. After the above processing, noise removal can be achieved in the carrier signal to be processed.

[0100] Step 104: Simultaneously generate a corresponding prediction model based on the noise source to perform noise prediction and obtain the corresponding prediction results, and optimize the corresponding carrier communication equipment based on the prediction results.

[0101] After removing noise from the carrier wave to be processed, the system also generates a prediction model based on the identified noise sources to predict the noise in the carrier signal and obtain the corresponding prediction results. Based on the prediction results, the system optimizes the equipment corresponding to the noise sources in the carrier communication equipment, thereby improving the communication efficiency and quality of the carrier communication equipment and enhancing the health status of the carrier communication equipment.

[0102] In this embodiment, after the system completes the noise stripping and noise source analysis based on the information flow fingerprint-energy flow fingerprint collaborative discrimination, i.e. the power electronic equipment type analysis corresponding to the noise, it can combine the corresponding noise source equipment discrimination results, the stripped noise sequence, and the corresponding prediction model generated based on the noise source to achieve collaborative optimization of the information flow-energy flow line of the carrier communication equipment.

[0103] For example, in this embodiment, generating a corresponding prediction model based on the noise source to perform noise prediction and obtain the corresponding prediction result specifically involves:

[0104] The preset equipment digital model is updated according to the noise source to obtain the corresponding prediction model, and the equipment operating noise is predicted according to the prediction model to obtain the corresponding noise prediction sequence.

[0105] Based on the noise prediction sequence and the noise stripping sequence, the health status of the device is predicted by the prediction model to obtain the prediction result of the health status of the corresponding carrier communication device.

[0106] In order to improve the future operation and communication quality of carrier communication equipment, the carrier communication method provided in this embodiment of the invention, after removing noise from the carrier signal, will update the preset digital model in the system according to the determined noise source to obtain a prediction model for predicting noise in the carrier, and predict the operating noise of the carrier communication equipment according to the prediction model to obtain the corresponding noise prediction sequence data.

[0107] In this embodiment, the system's preset device digital model is specifically a digital twin model of power electronic equipment constructed by integrating the physical parameters and operating characteristics of power electronic equipment, i.e., carrier communication equipment. j (j = 1, L, J), where J is the number of power electronic devices in the complete carrier communication device. The system determines the device based on the power line carrier noise source identification result j obtained in step 102. * The analysis results of the device, i.e., the power electronic equipment j, are called upon. * The corresponding digital twin model, namely the prediction model described in this embodiment, predicts the future operating noise of the power electronic equipment corresponding to the equipment assessment result, thereby obtaining the noise prediction sequence corresponding to the assessed equipment, i.e., the noise prediction sequence of N points.

[0108] After obtaining the noise prediction sequence, the system predicts the health status of the carrier communication equipment based on the noise prediction sequence and the noise stripping sequence, and obtains the corresponding prediction results, providing reference data for subsequent optimization of the carrier communication equipment.

[0109] In this embodiment, the system will predict the noise sequence based on the determined N points. And the stripped noise sequence originating from the N points of the noise source device. The prediction model is used to analyze the carrier communication device, i.e., the power electronic device j. * The prediction formula and prediction results for predicting the health status of [the individual] are shown below:

[0110]

[0111] in, For noise source equipment j * health status, The larger the value, the better the health status of the equipment; the first term of the formula is used to measure the overall similarity between the N-point noise prediction sequence and the stripped noise sequence, and the larger the value, the higher the similarity between the two; the second term of the formula is used to measure the difference between the N-point noise prediction sequence and the stripped noise sequence, and the smaller the value, the smaller the difference between the two.

[0112] Furthermore, the optimization of the corresponding carrier communication equipment based on the prediction results described in this embodiment specifically includes:

[0113] The prediction results are compared with the preset first threshold and the second threshold respectively to obtain the corresponding comparison results;

[0114] When the comparison result is that the predicted result is greater than or equal to the second threshold, the health status of the carrier communication device is determined to be excellent, and the status log of the carrier communication device is updated according to the health status.

[0115] When the comparison result is that the predicted result is greater than the first threshold and less than the second threshold, the health status of the carrier communication device is determined to be good. The carrier communication device is then adjusted to online maintenance status, and the communication line of the carrier communication device is optimized.

[0116] When the comparison result is less than or equal to the first threshold, the health status of the carrier communication device is determined to be extremely poor. The carrier communication device is then adjusted to an offline maintenance state, and the communication line of the carrier communication device is optimized.

[0117] To further improve the timeliness and accuracy of optimization for carrier communication devices, the carrier communication method provided in this embodiment of the invention compares the prediction results obtained in the above preferred examples with the first threshold and the second threshold respectively to obtain the corresponding comparison results. Based on the corresponding comparison results, the method selects the corresponding optimization method that is more in line with the real-time status of the device to optimize the communication line of the device. This allows the device to improve the communication quality and efficiency of the carrier after the communication line optimization, and also improves the health status of the carrier communication device while improving the carrier communication quality.

[0118] In this embodiment, based on the prediction results obtained from the prediction model, the system can further determine the corresponding power electronic equipment j. * That is, the maintenance strategy for the carrier communication equipment and whether to trigger information flow-energy flow collaborative optimization.

[0119] The system will predict the health status of the equipment. Compared with the first preset threshold Y in the system th,1 Second threshold Y th,2 (Y th,2 >Y th,1 The comparisons are performed separately, and the maintenance strategy for the carrier communication equipment is determined based on the comparison results, and it is determined whether to trigger the information flow-energy flow collaborative optimization.

[0120] If the comparison result indicates the health status of the equipment Higher than or equal to the second threshold Y th,2 At that time, that is If the device health status prediction result is "excellent", then the information flow and energy flow collaborative optimization will not be triggered.

[0121] If the comparison result indicates the health status of the equipment Below the second threshold Y th,2 And higher than the first threshold Yth,1 At that time, that is If the equipment health status prediction result is determined to be "good", the system will reduce the equipment's operating cycle and perform online maintenance, while triggering the equipment's information flow-energy flow collaborative optimization.

[0122] If the comparison result indicates the health status of the equipment Less than or equal to threshold Y th,1 At that time, that is If the predicted health status of the equipment is determined to be "extremely poor", the system will shut down the equipment and perform offline maintenance, while simultaneously triggering the collaborative optimization of the equipment's information flow and energy flow.

[0123] Accordingly, in this embodiment, the system stores the health status information of all power electronic devices in the network, that is, the health status information of all carrier communication devices. The system determines the health status information of the devices. Then, based on the determined health status information The system updates the health status information stored within it, and after determining that the information flow-energy flow collaborative optimization of the device has been triggered, it optimizes the power line carrier information flow-energy flow transmission path of the device, thereby achieving information flow-energy flow collaborative optimization.

[0124] If there are J power electronic devices in the network, forming M power line carrier information flow and energy flow transmission paths, the health state of the m-th transmission path is defined as the weighted difference between the sum of the health states of all power electronic devices on that path and the variance of the health states of all power electronic devices. The corresponding calculation formula is as follows:

[0125]

[0126] In this formula, the first term is the sum of the health status of all power electronic devices along the path, the second term is the variance of the health status of all power electronic devices, and υ1 and υ2 are the corresponding weight coefficients. The health status of the m-th power line carrier transmission path; Ω m Let m be the set of power electronic devices included in the m-th power line carrier transmission path. The health status of all power line carrier transmission paths can be calculated using the above formula.

[0127] After determining the health status of all transmission paths, the system will select the transmission path with the optimal health status based on the health status optimization method for information and energy flow transmission. The final information-energy flow co-optimization result can then be determined as follows:

[0128]

[0129] in, ζ represents the health status of the m-th transmission path; ζ represents the final information flow-energy flow collaborative optimization result. After collaboratively optimizing the device's information flow-energy flow transmission path using the above method, the system will determine whether the device has reached its maximum optimization period. If it is determined that the device has reached its maximum optimization period, the device optimization process is completed; if it is determined that the maximum optimization period has not been reached, the system will return to the power line carrier signal acquisition step and repeat the above optimization process for the carrier communication device.

[0130] To better illustrate the working principle and steps of the carrier communication method and apparatus based on noise stripping optimization of the present invention, please refer to the relevant description above, but not limited to.

[0131] Accordingly, see Figure 3 , Figure 3 This is a schematic diagram of an embodiment of the carrier communication device based on noise stripping optimization provided by the present invention. Figure 3 As shown, the carrier communication device includes a fingerprint extraction module 301, a source determination module 302, a noise stripping module 303, and a device optimization module 304.

[0132] The fingerprint extraction module 301 is used to acquire the carrier to be processed and to perform information flow fingerprint extraction processing and energy flow fingerprint extraction processing on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed.

[0133] Furthermore, the fingerprint extraction module 301 performs information flow fingerprint extraction processing and energy flow fingerprint extraction processing on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed, specifically including:

[0134] A preset differential constellation trajectory diagram is invoked to perform differential operations on the carrier to be processed, and the obtained differential operation result is used as the information flow fingerprint; at the same time, a preset Hilbert transform algorithm is invoked to obtain the current analysis signal of the carrier to be processed, and the amplitude, phase and frequency features of the current analysis signal are extracted respectively, and the feature extraction results are used as the energy flow fingerprint.

[0135] The source determination module 302 is used to calculate the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard fingerprints in a preset standard fingerprint library, and to determine the noise source in the carrier to be processed based on the similarity coefficients.

[0136] Furthermore, the source determination module 302 calculates the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard fingerprints in a preset standard fingerprint library, and determines the noise source in the carrier to be processed based on the similarity coefficients, specifically as follows:

[0137] The differences between the information flow fingerprint and the energy flow fingerprint and the corresponding information flow standard fingerprint and energy flow standard fingerprint in the standard fingerprint library are calculated respectively, thereby obtaining several pairs of information flow differences and energy flow differences; the sum of squares of each pair of information flow differences and energy flow differences is calculated, and the reciprocal of the calculated sum of squares is used as the similarity coefficient of the corresponding standard fingerprint. Then, based on the calculated similarity coefficient, a preset maximum likelihood estimation method is called to determine the noise source of the carrier to be processed, thereby obtaining the noise source.

[0138] The noise stripping module 303 is used to generate a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, so that the noise stripping network performs noise stripping processing on the carrier to be processed.

[0139] Furthermore, the noise stripping module 303 generates a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, enabling the noise stripping network to perform noise stripping processing on the carrier to be processed, specifically as follows:

[0140] The standard noise in the preset standard noise library, the information flow fingerprint, and the energy flow fingerprint are feature extracted and concatenated. The concatenation result is used as input data to the preset initial feature processing network, so that the initial feature processing network is updated to the noise stripping network. The noise stripping network is controlled to perform feature processing and analysis on the input data, output corresponding sequence data as noise stripping sequence, and perform noise stripping processing on the carrier to be processed according to the noise stripping sequence.

[0141] The device optimization module 304 is used to generate a corresponding prediction model based on the noise source, perform noise prediction, obtain the corresponding prediction result, and optimize the corresponding carrier communication device based on the prediction result.

[0142] Furthermore, the equipment optimization module 304 generates a corresponding prediction model based on the noise source to perform noise prediction and obtain the corresponding prediction results, specifically as follows:

[0143] The preset device digital model is updated according to the noise source to obtain the corresponding prediction model, and the device operating noise is predicted according to the prediction model to obtain the corresponding noise prediction sequence; the device health status is predicted according to the noise prediction sequence and the noise stripping sequence through the prediction model to obtain the prediction result of the corresponding carrier communication device health status.

[0144] Furthermore, the device optimization module 304 optimizes the corresponding carrier communication device based on the prediction result, specifically including:

[0145] The prediction results are compared with the preset first threshold and the second threshold respectively to obtain the corresponding comparison results;

[0146] When the comparison result is that the predicted result is greater than or equal to the second threshold, the health status of the carrier communication device is determined to be excellent, and the status log of the carrier communication device is updated according to the health status.

[0147] When the comparison result is that the predicted result is greater than the first threshold and less than the second threshold, the health status of the carrier communication device is determined to be good. The carrier communication device is then adjusted to online maintenance status, and the communication line of the carrier communication device is optimized.

[0148] When the comparison result is less than or equal to the first threshold, the health status of the carrier communication device is determined to be extremely poor. The carrier communication device is then adjusted to an offline maintenance state, and the communication line of the carrier communication device is optimized.

[0149] In summary, this invention provides a carrier communication method and apparatus based on noise stripping optimization. It extracts information flow fingerprints and energy flow fingerprints from the carrier to be processed, and calculates the similarity coefficients between these fingerprints and each standard noise in a standard noise library. Based on these similarity coefficients, the noise sources in the carrier are determined. After determination, a noise stripping network is generated based on the information flow fingerprints and energy flow fingerprints to strip noise from the carrier. Simultaneously, a prediction model is generated based on the noise sources to predict noise, and the carrier communication equipment is optimized based on the prediction results. This invention improves the accuracy of the judgment results by collaboratively analyzing the information flow fingerprints and energy flow fingerprints to determine the noise sources in the carrier signal. Furthermore, the noise stripping based on these two fingerprints also improves the accuracy and completeness of the stripping process. Finally, the optimization of the carrier communication equipment based on these two fingerprints achieves collaborative optimization of the information flow and energy flow of the power line carrier communication equipment, effectively improving the health status of the carrier communication equipment.

[0150] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A carrier communication method based on noise stripping optimization, characterized in that, Includes the following steps: A carrier to be processed is acquired, and information flow fingerprint extraction and energy flow fingerprint extraction are performed on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed. Calculate the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard noises in a preset standard noise library, and determine the noise source in the carrier to be processed based on the similarity coefficients; A corresponding noise stripping network is generated based on the information flow fingerprint and the energy flow fingerprint, so that the noise stripping network performs noise stripping processing on the carrier to be processed. At the same time, a corresponding prediction model is generated based on the noise source to perform noise prediction and obtain the corresponding prediction result. The corresponding carrier communication equipment is optimized based on the prediction result. Specifically, the step of performing information flow fingerprint extraction and energy flow fingerprint extraction on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed includes: The carrier to be processed is subjected to differential operation by calling a preset differential constellation trajectory diagram, and the obtained differential operation result is used as the information flow fingerprint; Simultaneously, a preset Hilbert transform algorithm is invoked to obtain the current analysis signal of the carrier to be processed, and the current analysis signal is subjected to feature extraction of amplitude, phase and frequency respectively, and the feature extraction results are used as the energy flow fingerprint; Specifically, the step of generating a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, so that the noise stripping network performs noise stripping processing on the carrier to be processed, involves: Feature extraction and splicing are performed on the standard noise in the preset standard noise library, the information flow fingerprint, and the energy flow fingerprint. The splicing result is then used as input data to the preset initial feature processing network, so that the initial feature processing network is updated to the noise stripping network. The noise stripping network is controlled to perform feature processing and analysis on the input data, output corresponding sequence data as a noise stripping sequence, and perform noise stripping processing on the carrier to be processed according to the noise stripping sequence.

2. The carrier communication method based on noise stripping optimization as described in claim 1, characterized in that, The step of calculating the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard noises in a preset standard noise library, and determining the noise source in the carrier to be processed based on the similarity coefficients, specifically involves: The differences between the information flow fingerprint and the energy flow fingerprint and the information flow standard fingerprint and energy flow standard fingerprint corresponding to the several types of standard noise in the standard noise library are calculated respectively, and then several pairs of information flow differences and energy flow differences are obtained; The sum of squares of the information flow difference and energy flow difference for each pair is calculated, and the reciprocal of the calculated sum of squares is used as the similarity coefficient of the corresponding standard noise. Then, based on the calculated similarity coefficient, a preset maximum likelihood estimation method is called to determine the noise source of the carrier to be processed, and the noise source is obtained.

3. The carrier communication method based on noise stripping optimization as described in claim 1, characterized in that, The step of generating a corresponding prediction model based on the noise source to perform noise prediction and obtain the corresponding prediction results is as follows: The preset equipment digital model is updated according to the noise source to obtain the corresponding prediction model, and the equipment operating noise is predicted according to the prediction model to obtain the corresponding noise prediction sequence. Based on the noise prediction sequence and the noise stripping sequence, the health status of the device is predicted by the prediction model to obtain the prediction result of the health status of the corresponding carrier communication device.

4. The carrier communication method based on noise stripping optimization as described in claim 1, characterized in that, The optimization of the corresponding carrier communication equipment based on the prediction results specifically includes: The prediction results are compared with the preset first threshold and the second threshold respectively to obtain the corresponding comparison results; When the comparison result is that the predicted result is greater than or equal to the second threshold, the health status of the carrier communication device is determined to be excellent, and the status log of the carrier communication device is updated according to the health status. When the comparison result is that the predicted result is greater than the first threshold and less than the second threshold, the health status of the carrier communication device is determined to be good. The carrier communication device is then adjusted to online maintenance status, and the communication line of the carrier communication device is optimized. When the comparison result is less than or equal to the first threshold, the health status of the carrier communication device is determined to be extremely poor. The carrier communication device is then adjusted to an offline maintenance state, and the communication line of the carrier communication device is optimized.

5. A carrier communication device based on noise stripping optimization, characterized in that, The carrier communication device includes a fingerprint extraction module, a source determination module, a noise stripping module, and a device optimization module; The fingerprint extraction module is used to acquire the carrier to be processed and to perform information flow fingerprint extraction and energy flow fingerprint extraction processing on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed. The source determination module is used to calculate the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard fingerprints in a preset standard fingerprint library, and to determine the noise source in the carrier to be processed based on the similarity coefficients; The noise stripping module is used to generate a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, so that the noise stripping network performs noise stripping processing on the carrier to be processed. The device optimization module is used to generate a corresponding prediction model based on the noise source, perform noise prediction, obtain the corresponding prediction result, and optimize the corresponding carrier communication device based on the prediction result. The fingerprint extraction module performs information flow fingerprint extraction and energy flow fingerprint extraction on the carrier to be processed to obtain the information flow fingerprint and energy flow fingerprint of the carrier to be processed, specifically including: The carrier to be processed is subjected to differential operation by calling a preset differential constellation trajectory diagram, and the obtained differential operation result is used as the information flow fingerprint; Simultaneously, a preset Hilbert transform algorithm is invoked to obtain the current analysis signal of the carrier to be processed, and the current analysis signal is subjected to feature extraction of amplitude, phase and frequency respectively, and the feature extraction results are used as the energy flow fingerprint; Specifically, the noise stripping module generates a corresponding noise stripping network based on the information flow fingerprint and the energy flow fingerprint, enabling the noise stripping network to perform noise stripping processing on the carrier to be processed. Feature extraction and splicing are performed on the standard noise in the preset standard noise library, the information flow fingerprint, and the energy flow fingerprint. The splicing result is then used as input data to the preset initial feature processing network, so that the initial feature processing network is updated to the noise stripping network. The noise stripping network is controlled to perform feature processing and analysis on the input data, output corresponding sequence data as a noise stripping sequence, and perform noise stripping processing on the carrier to be processed according to the noise stripping sequence.

6. A carrier communication device based on noise stripping optimization as described in claim 5, characterized in that, The source determination module calculates the similarity coefficients between the information flow fingerprint and the energy flow fingerprint and several standard fingerprints in a preset standard fingerprint library, and determines the noise source in the carrier to be processed based on the similarity coefficients, specifically as follows: The differences between the information flow fingerprint and the energy flow fingerprint and the information flow standard fingerprint and the energy flow standard fingerprint corresponding to the several types of standard fingerprints in the standard fingerprint library are calculated respectively, and then several pairs of information flow differences and energy flow differences are obtained; The sum of squares of the information flow difference and energy flow difference for each pair is calculated, and the reciprocal of the calculated sum of squares is used as the similarity coefficient of the corresponding standard fingerprint. Then, based on the calculated similarity coefficient, a preset maximum likelihood estimation method is called to determine the noise source of the carrier to be processed, and the noise source is obtained.

Images