Power line carrier data anomaly detection method and device

Abstract

The invention relates to the technical field of power communication, and provides a power carrier data anomaly detection method and device, and the method comprises the steps: obtaining a power carrier signal sequence and environment parameters; extracting time domain and frequency domain characteristic parameters of the signal; generating an abnormal feature sequence based on the feature parameters; generating an abnormal feature interval set in combination with the environmental parameters; and generating an abnormal information sequence by using a pre-trained classification model. Specifically, a signal feature difference sequence is generated by analyzing the difference between adjacent feature parameters, and then an abnormal feature sequence and a feature change interval are determined. And abnormal time intervals are combined according to abnormal types, abnormal / normal signal fragments are divided, abnormal characteristics are analyzed in combination with environmental parameters, finally, detection results are spliced, and abnormal information and tracks are visualized. According to the invention, the accuracy and traceability of abnormal recognition can be improved.

Description

Technical Field

[0001] The present invention relates to the field of power communication technology, and more particularly, to a method and device for detecting power carrier data anomaly. Background Art

[0002] With the continuous development of power systems, power carrier communication (PLC), as an important communication method, has been widely used in areas such as monitoring, protection, and automated control. Power carrier signals are transmitted through power lines, enabling remote data transmission and communication between devices. However, PLC signals are susceptible to various factors during transmission, such as electromagnetic interference, line aging, and equipment failure, which can cause signal anomalies. These abnormal signals can affect the normal operation of the power system and even cause safety incidents. Therefore, timely and accurate detection of anomalies in PLC data is crucial.

[0003] Existing methods for detecting anomalies in power carrier data are mainly based on time-domain or frequency-domain analysis of the signal. Time-domain analysis methods monitor the signal's amplitude, period, and other characteristics to determine whether the signal is abnormal. Frequency-domain analysis methods use techniques such as Fourier transform to convert the signal from the time domain to the frequency domain, analyzing characteristics such as the signal's frequency distribution and power spectrum to detect anomalies. However, these methods can usually only analyze the signal from a single time-domain or frequency-domain perspective and cannot fully reflect the abnormal characteristics of the signal. In addition, existing detection methods often ignore the impact of the power carrier transmission environment on the signal, resulting in limited accuracy and reliability of the detection results.

[0004] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the existing technology: the existing methods are unable to comprehensively consider the time domain and frequency domain characteristics of the power carrier signal, and it is difficult to comprehensively detect signal abnormalities; at the same time, the existing methods do not fully combine the power carrier transmission environment parameters, and cannot accurately reflect the relationship between signal anomalies and the transmission environment, thereby affecting the accuracy of anomaly detection. Summary of the Invention

[0005] The present invention provides a method and device for detecting abnormality in power line carrier data.

[0006] In a first aspect of the present invention, a method for detecting abnormality in power line carrier data is provided, comprising: Obtaining power carrier signal sequence and power carrier transmission environment parameters; Extracting time domain characteristic parameters and frequency domain characteristic parameters according to the power carrier signal sequence; generating an abnormal characteristic sequence corresponding to the power carrier signal based on the time domain characteristic parameters and the frequency domain characteristic parameters; generating an abnormal feature interval set according to the abnormal feature sequence and the power carrier transmission environment parameter; Based on the abnormal feature interval set and the pre-trained abnormal classification model, a power line carrier data abnormal information sequence is generated.

[0007] Furthermore, extracting time domain characteristic parameters and frequency domain characteristic parameters according to the power carrier signal sequence includes: Performing time domain analysis on each signal segment in the power carrier signal sequence to determine each time domain characteristic parameter characterizing the time domain characteristics of the signal; Determining the determined time domain characteristic parameters characterizing the time domain characteristics of the signal as time domain characteristic parameters; Frequency domain analysis is performed on each signal segment in the power carrier signal sequence to obtain frequency domain characteristic parameters, wherein characteristic values in the frequency domain characteristic parameters correspond to signal segments in the power carrier signal sequence.

[0008] Furthermore, the generating of the abnormal characteristic sequence corresponding to the power carrier signal based on the time domain characteristic parameter and the frequency domain characteristic parameter includes: generating a signal feature difference sequence according to the time domain feature parameters and the frequency domain feature parameters; generating an abnormal feature sequence according to the signal feature difference sequence; According to the abnormal feature sequence, a set of abnormal feature intervals corresponding to the power carrier signal is generated.

[0009] Furthermore, generating a signal feature difference sequence according to the time domain feature parameters and the frequency domain feature parameters includes: For each two adjacent characteristic parameters of the time domain characteristic parameters and the frequency domain characteristic parameters, the following steps are performed: determining a difference between the two characteristic parameters as a characteristic difference value; Determining whether the feature difference value is greater than a preset difference threshold; In response to determining that the feature difference value is greater than the preset difference threshold, determining the feature difference value as an abnormal feature difference value; generating a characteristic change rate based on the abnormal characteristic difference value and the signal time difference corresponding to the power carrier signal sequence; Determine the generated individual feature change rates as a feature change rate set; For each feature change rate in the feature change rate set, perform the following steps: Determining two characteristic parameters corresponding to the characteristic change rate as a characteristic point sequence; determining two signal segments corresponding to the characteristic point sequence as a characteristic signal sequence; Determining the acquisition time corresponding to the first signal segment in the characteristic signal sequence as the first time point; Determining the acquisition time corresponding to the second signal segment in the characteristic signal sequence as the second time point; combining the first time point and the second time point into a feature change interval; The combined feature change intervals are determined as a feature change interval set.

[0010] Furthermore, generating an abnormal feature interval set according to the abnormal feature sequence includes: Determining the abnormality type corresponding to each abnormal feature in the abnormal feature sequence; Determining the first two abnormal features in the abnormal feature sequence as an initial abnormal feature group; Based on the initial set of anomaly features, the following loop steps are performed: Determine the initial abnormal feature of the corresponding abnormal type characterizing the mutation in the initial abnormal feature group as the mutation feature; Determine the initial abnormal feature of the drift corresponding to the abnormal type in the initial abnormal feature group as the drift feature; Determining the signal acquisition time corresponding to the mutation feature as the third time point; Determining the signal acquisition time corresponding to the drift feature as a fourth time point; combining the third time point and the fourth time point into an abnormal time interval; In response to determining that the initial abnormal feature group is not the last two abnormal features in the abnormal feature sequence, determining two abnormal features arranged after the initial abnormal feature group in the abnormal feature sequence as the initial abnormal feature group to update the initial abnormal feature group, and executing the loop step again based on the updated initial abnormal feature group; The combined abnormal time intervals are determined as an abnormal feature interval set.

[0011] Furthermore, generating the abnormal feature interval set according to the abnormal feature interval set and the power carrier transmission environment parameter includes: Dividing the power carrier signal according to the abnormal feature interval set to obtain an abnormal signal segment set and a normal signal segment set; For each abnormal signal segment in the abnormal signal segment set, performing feature analysis in combination with the power carrier transmission environment parameter to obtain an abnormal feature analysis result; The obtained abnormal feature analysis results and the normal signal segment set are spliced in chronological order to obtain an abnormality detection result.

[0012] Furthermore, the method further comprises: Performing visualization processing on each abnormal information in the power carrier data abnormal information sequence according to each abnormal level included in the power carrier data abnormal information sequence; Based on the abnormal feature interval set, the abnormal trajectory is superimposed and displayed in the target monitoring space, wherein the target monitoring space corresponds to the transmission scenario when the power carrier signal is acquired; According to the abnormal feature interval set and the abnormal information sequence, the abnormal information sequence is visualized in the target monitoring space.

[0013] In a second aspect of the present invention, a power line carrier data anomaly detection device is provided, comprising: an acquisition unit configured to acquire a power carrier signal sequence and a power carrier transmission environment parameter; a feature extraction unit configured to extract time domain feature parameters and frequency domain feature parameters according to the power carrier signal sequence; a feature generating unit configured to generate an abnormal feature sequence corresponding to the power carrier signal based on the time domain feature parameter and the frequency domain feature parameter; a detection unit configured to generate an abnormal feature interval set according to the abnormal feature sequence and the power carrier transmission environment parameter; The information generating unit is configured to generate a power carrier data abnormality information sequence based on the abnormal feature interval set and a pre-trained abnormality classification model.

[0014] In a third aspect of the present invention, an electronic device is provided, comprising: at least one processor, a memory, and an input / output unit; wherein the memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute any one of the methods described in the first aspect.

[0015] In a fourth aspect of the present invention, a computer-readable storage medium is provided, comprising instructions, which, when executed on a computer, enable the computer to execute the method according to any one of the first aspects.

[0016] The above-described embodiments of the present invention have at least the following beneficial effects: The present invention can enhance the anomaly detection capabilities of power carrier communication systems. By jointly analyzing time-domain and frequency-domain characteristic parameters, combined with power carrier transmission environment parameters, it can more comprehensively capture signal anomaly characteristics, thereby improving detection accuracy and reliability. This method can automatically generate a set of anomaly feature intervals and achieve efficient classification using a pre-trained anomaly classification model, providing strong support for the stable operation of power carrier communication systems. This method optimizes the anomaly detection process. By calculating the difference sequence of signal features and the rate of change of features, it can quickly locate abnormal signal segments and conduct a comprehensive analysis based on environmental parameters to reduce false positives and missed negatives. Furthermore, this method enables a visual display of anomaly information, including anomaly levels and trajectory overlays, making it easier for operators to intuitively understand system status and improve fault diagnosis and operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily apparent by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which: Figure 1 A schematic diagram of a flow chart of a method for detecting abnormality in power line carrier data provided by one embodiment of the present invention; Figure 2 A schematic diagram of the structure of a power line carrier data anomaly detection device provided by one embodiment of the present invention; Figure 3 The figure schematically shows the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. Rather, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0019] Those skilled in the art will appreciate that embodiments of the present invention may be implemented as a device, apparatus, apparatus, method, or computer program product. Therefore, the present invention may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software.

[0020] It should be noted that any number of elements in the drawings is for illustration only and not for limitation, and any naming is only for distinction and does not have any limiting meaning.

[0021] Reference below Figure 1 , Figure 1 This is a flow chart of a method for detecting abnormality in power carrier data provided by one embodiment of the present invention. Figure 1 As shown, a method for detecting abnormality in power carrier data includes: S1 obtains the power carrier signal sequence and power carrier transmission environment parameters; S2 extracts time domain characteristic parameters and frequency domain characteristic parameters according to the power carrier signal sequence; S3: generating an abnormal characteristic sequence corresponding to the power carrier signal based on the time domain characteristic parameter and the frequency domain characteristic parameter; S4: generating an abnormal feature interval set according to the abnormal feature sequence and the power carrier transmission environment parameter; S5 generates a power line carrier data abnormality information sequence based on the abnormal feature interval set and the pre-trained abnormality classification model.

[0022] It should be noted that the core of the power carrier data anomaly detection method of the present invention lies in the comprehensive analysis of the time domain and frequency domain characteristics of the power carrier signal, and the combination of the power carrier transmission environment parameters to accurately detect anomalies. The power carrier signal sequence refers to the signal data sequence transmitted through the power line. These signals may be affected by various factors and become abnormal. The power carrier transmission environment parameters refer to environmental factors related to signal transmission, such as line length, load conditions, electromagnetic interference intensity, etc. These parameters will affect the transmission quality and characteristics of the signal. By extracting the time domain characteristic parameters and the frequency domain characteristic parameters, the characteristic changes of the signal can be fully reflected, and then an abnormal characteristic sequence can be generated to provide a basis for subsequent anomaly detection.

[0023] Specifically, time domain characteristic parameters refer to the characteristic parameters of the signal in the time domain, such as the amplitude, period, zero-crossing rate, etc. of the signal. Frequency domain characteristic parameters refer to the characteristic parameters of the signal in the frequency domain, such as the frequency distribution and power spectrum density of the signal. When extracting time domain characteristic parameters, the time domain characteristics of the signal can be characterized by performing time domain analysis on the signal sequence, such as calculating the statistical characteristics of the signal such as the peak value, mean, and variance. When extracting frequency domain characteristic parameters, the signal sequence can be subjected to Fourier transform or other frequency domain analysis methods to obtain the spectral characteristics of the signal. These characteristic parameters can reflect the characteristic changes of the signal in different domains and provide a basis for the subsequent generation of abnormal characteristics. At the same time, power carrier transmission environment parameters include line length, load conditions, electromagnetic interference intensity, etc. These parameters can be obtained through sensors or other monitoring equipment and used to generate the subsequent abnormal feature interval set.

[0024] Preferably, the process of generating an abnormal feature sequence can be further refined. First, based on the extracted time domain feature parameters and frequency domain feature parameters, the difference value between each two adjacent feature parameters, i.e., the feature difference value, is calculated. If the feature difference value is greater than a preset difference threshold, the signal segment corresponding to the difference value is considered to be abnormal and is marked as an abnormal feature difference value. The preset difference threshold can be set according to the actual application scenario and signal characteristics, for example, determined through historical data statistics or experiments. Then, based on the abnormal feature difference value and the corresponding signal time difference, the feature change rate is calculated to form a feature change rate set. The two feature parameters corresponding to each feature change rate constitute a feature point sequence, and the corresponding signal segment is then determined to be a feature signal sequence. By determining the acquisition time of the signal segment in the feature signal sequence as a time point and combining them into a feature change interval, a feature change interval set is finally formed. This process can effectively identify abnormal changes in the signal and provide detailed feature information for subsequent anomaly classification.

[0025] In an embodiment of the present invention, a pre-trained anomaly classification model is constructed using machine learning or deep learning techniques and is used to classify abnormal features of power carrier signals. The model training process is as follows: First, the model is trained based on a large amount of historical power carrier signal data, which includes samples of normal and abnormal signals. Abnormal signal samples are labeled with the anomaly type (e.g., mutation, drift, etc.) and the anomaly level (e.g., mild, moderate, severe, etc.). Furthermore, the training data also includes environmental parameters related to signal transmission (e.g., line length, load conditions, electromagnetic interference intensity, etc.) so that the model can learn the impact of the environment on signal anomalies.

[0026] During the data preprocessing phase, historical data is cleaned and normalized to ensure data quality. Next, time-domain and frequency-domain feature parameters (such as signal amplitude, period, and power spectral density) are extracted and combined with environmental parameters to form the model's input feature vector. Feature selection and extraction techniques are used to optimize the input feature vector to improve model performance.

[0027] Models can be selected from traditional machine learning models (such as support vector machines, decision trees, and random forests) or deep learning models (such as recurrent neural networks and convolutional neural networks). If the signal has temporal characteristics, time series models (such as LSTM) can be used. Models are trained using labeled data, and model parameters are adjusted to minimize prediction error. Cross-validation is used to evaluate model performance and ensure generalization. Model hyperparameters (such as learning rate and regularization strength) are optimized using grid search or random search.

[0028] The model's inputs are the characteristic parameters of the power carrier signal (time and frequency domain characteristics) and environmental parameters (such as line length, load conditions, and electromagnetic interference intensity). Its output is anomaly classification results, including anomaly type and severity. Through continuous learning and performance monitoring, the model can be regularly updated with new anomaly data to improve its adaptability and accuracy, and ensure its stability in different environments in real-world applications.

[0029] Through this training process, the pre-trained anomaly classification model can effectively classify abnormal features of power carrier signals, providing a reliable basis for anomaly detection. This model can be deployed in electronic devices and used in conjunction with the power carrier data anomaly detection method of the present invention to achieve efficient anomaly detection and classification.

[0030] In some embodiments, extracting time domain characteristic parameters and frequency domain characteristic parameters according to the power carrier signal sequence includes: Performing time domain analysis on each signal segment in the power carrier signal sequence to determine each time domain characteristic parameter characterizing the time domain characteristics of the signal; Determining the determined time domain characteristic parameters characterizing the time domain characteristics of the signal as time domain characteristic parameters; Frequency domain analysis is performed on each signal segment in the power carrier signal sequence to obtain frequency domain characteristic parameters, wherein characteristic values in the frequency domain characteristic parameters correspond to signal segments in the power carrier signal sequence.

[0031] It should be noted that the extraction of time domain and frequency domain characteristic parameters of the power carrier signal sequence in the present invention is to comprehensively analyze the characteristics of the signal, so as to more accurately identify anomalies. Time domain analysis processing refers to the analysis of the characteristics of the signal in the time domain, and characterizes the time domain characteristics of the signal by calculating the statistical characteristics of the signal such as the amplitude, period, and zero-crossing rate. Frequency domain analysis processing converts the signal from the time domain to the frequency domain, and obtains the frequency distribution and power spectrum density of the signal through methods such as Fourier transform. These characteristic parameters can reflect the characteristic changes of the signal from different angles, providing an important basis for subsequent anomaly detection.

[0032] Specifically, time domain characteristic parameters include the amplitude, period, mean, variance, and zero-crossing rate of the signal. Amplitude refers to the difference between the maximum and minimum values of the signal, reflecting the strength of the signal; period refers to the time interval between the recurrence of the signal, reflecting the frequency characteristics of the signal; mean and variance reflect the average strength and intensity fluctuation of the signal, respectively; zero-crossing rate refers to the number of times the signal crosses the zero point per unit time, reflecting the dynamic change characteristics of the signal. Frequency domain characteristic parameters include the frequency distribution and power spectrum density of the signal. Frequency distribution refers to the energy distribution of the signal at different frequencies, reflecting the spectral characteristics of the signal; power spectrum density is the power distribution of the signal per unit frequency, reflecting the energy distribution of the signal. When extracting time domain characteristic parameters, the values of the above characteristic parameters can be calculated by performing statistical analysis on the signal sequence; when extracting frequency domain characteristic parameters, the signal can be converted from the time domain to the frequency domain through Fourier transform to obtain the signal spectrum diagram, and then characteristic parameters such as frequency distribution and power spectrum density can be calculated.

[0033] Preferably, time-domain analysis and processing can be implemented through the following steps: first, sampling the power carrier signal sequence to obtain discrete signal data; then, calculating statistical characteristics of the signal data, such as amplitude, period, mean, variance, and zero-crossing rate. For example, the amplitude can be obtained by calculating the difference between the maximum and minimum values of the signal; the period can be obtained by calculating the time interval between adjacent peaks or valleys of the signal; the mean can be obtained by calculating the average value of the signal data; the variance can be obtained by calculating the average of the squared deviations of the signal data from the mean; and the zero-crossing rate can be obtained by counting the number of times the signal data crosses zero. Frequency-domain analysis and processing can be implemented through the following steps: first, performing a Fourier transform on the power carrier signal sequence to obtain a signal spectrum; then, extracting characteristic parameters such as frequency distribution and power spectral density from the spectrum. For example, frequency distribution can be obtained by statistically analyzing the energy distribution of the signal at different frequencies; and power spectral density can be obtained by calculating the power distribution of the signal per unit frequency. Through the above steps, the time-domain and frequency-domain characteristic parameters of the power carrier signal can be accurately extracted, providing reliable data support for subsequent anomaly detection.

[0034] In some embodiments, generating the abnormal characteristic sequence corresponding to the power carrier signal based on the time domain characteristic parameter and the frequency domain characteristic parameter includes: generating a signal feature difference sequence according to the time domain feature parameters and the frequency domain feature parameters; generating an abnormal feature sequence according to the signal feature difference sequence; According to the abnormal feature sequence, a set of abnormal feature intervals corresponding to the power carrier signal is generated.

[0035] It should be noted that the generation of signal feature difference sequences based on time domain feature parameters and frequency domain feature parameters mentioned in the present invention, and the further generation of abnormal feature sequences and abnormal feature interval sets, is to identify abnormal conditions in the signal by analyzing the changes in feature parameters. The signal feature difference sequence refers to a sequence generated by comparing the differences between adjacent feature parameters, and these difference values can reflect the changes in signal features. The abnormal feature sequence is a sequence with abnormal features screened out from the signal feature difference sequence, and is used to identify possible abnormal signal segments. The abnormal feature interval set refers to an interval set that combines the abnormal features in the abnormal feature sequence in chronological order, and is used to clarify the time range of the abnormal signal segment.

[0036] Specifically, time-domain and frequency-domain characteristic parameters describe signal characteristics in two different domains: time and frequency. Time-domain characteristic parameters include the signal's amplitude, period, mean, and variance, while frequency-domain characteristic parameters include the signal's frequency distribution and power spectral density. A signal feature difference sequence is generated by pairwise comparison of these characteristic parameters, calculating the difference between each pair of adjacent characteristic parameters. If this difference exceeds a preset difference threshold, the signal segment corresponding to that difference is considered abnormal and marked as an abnormal feature difference value. The preset difference threshold is a reference value set based on the normal range of signal variation and is used to distinguish normal from abnormal signal variations. The abnormal feature sequence is a sequence consisting of all abnormal feature difference values, reflecting possible anomalies in the signal. The abnormal feature interval set is a set of time intervals formed by combining the abnormal features in the abnormal feature sequence according to their corresponding signal acquisition times. It is used to identify the specific time range of the abnormal signal segment.

[0037] Preferably, the process of generating a signal feature difference sequence can be further refined. First, for each pair of adjacent feature parameters in the time domain and frequency domain, the difference between them is calculated. For example, if the feature parameter is the signal amplitude, the difference is the difference between the amplitudes of two adjacent signal segments. This difference is then compared with a preset difference threshold. The preset difference threshold can be set based on the normal fluctuation range of the signal, for example, by analyzing the variation range of the feature parameter in historical normal signal data to determine a reasonable threshold. If the difference exceeds the preset difference threshold, the difference is marked as an abnormal feature difference value. Next, based on the abnormal feature difference value and the acquisition time of the corresponding signal segment, the feature change rate is calculated. The feature change rate can be expressed as the ratio of the abnormal feature difference value to the acquisition time interval of the signal segment, which measures the speed of signal feature change. Finally, the two feature parameters corresponding to the feature change rate are determined as a feature point sequence, and the two signal segments corresponding to the feature point sequence are determined as a feature signal sequence. In this way, abnormal changes in the signal can be more accurately located and a set of abnormal feature intervals can be generated, providing detailed time and feature information for subsequent anomaly analysis and processing.

[0038] In some embodiments, generating a signal feature difference sequence according to the time domain feature parameter and the frequency domain feature parameter includes: For each two adjacent characteristic parameters of the time domain characteristic parameters and the frequency domain characteristic parameters, the following steps are performed: determining a difference between the two characteristic parameters as a characteristic difference value; Determining whether the feature difference value is greater than a preset difference threshold; In response to determining that the feature difference value is greater than the preset difference threshold, determining the feature difference value as an abnormal feature difference value; generating a characteristic change rate based on the abnormal characteristic difference value and the signal time difference corresponding to the power carrier signal sequence; Determine the generated individual feature change rates as a feature change rate set; For each feature change rate in the feature change rate set, perform the following steps: Determining two characteristic parameters corresponding to the characteristic change rate as a characteristic point sequence; determining two signal segments corresponding to the characteristic point sequence as a characteristic signal sequence; Determining the acquisition time corresponding to the first signal segment in the characteristic signal sequence as the first time point; Determining the acquisition time corresponding to the second signal segment in the characteristic signal sequence as the second time point; combining the first time point and the second time point into a feature change interval; The combined feature change intervals are determined as a feature change interval set.

[0039] It should be noted that the process of generating a signal feature difference sequence mentioned in the present invention is achieved by performing a difference analysis on the time domain feature parameters and the frequency domain feature parameters. This process aims to identify abnormal feature difference values by calculating the difference values between adjacent feature parameters and combining them with a preset difference threshold. The feature difference value refers to the amount of change between two adjacent feature parameters, which reflects the change in the signal feature over time. The preset difference threshold is a reference value used to determine whether the feature difference value belongs to an abnormal change, and is usually set according to the normal fluctuation range of the signal. The feature change rate refers to the ratio of the feature difference value to the signal acquisition time interval, which is used to measure the speed of the signal feature change. Through this process, abnormal changes in the signal can be effectively identified, and a basis can be provided for the subsequent generation of abnormal feature interval sets.

[0040] Specifically, time-domain and frequency-domain feature parameters describe signal characteristics in two different domains: time and frequency. Time-domain feature parameters include signal amplitude, period, mean, and variance, while frequency-domain feature parameters include signal frequency distribution and power spectral density. When calculating feature difference values, the difference between each pair of adjacent feature parameters is calculated. For example, if the feature parameter is signal amplitude, the difference value is the difference between the amplitudes of two adjacent signal segments. The preset difference threshold is set based on the normal signal fluctuation range and is used to distinguish normal from abnormal signal changes. If the feature difference value exceeds the preset difference threshold, the signal segment corresponding to the difference value is considered abnormal and marked as an abnormal feature difference value. The feature change rate is calculated by dividing the abnormal feature difference value by the acquisition time interval of the corresponding signal segment. It measures the speed of signal feature change. The feature change rate set is a set of all feature change rates and is used to further analyze changes in signal features. A feature point sequence refers to the two feature parameters corresponding to the feature change rate, while a feature signal sequence refers to the two signal segments corresponding to the feature point sequence. A feature change interval is the interval consisting of the acquisition times of two signal segments in a feature signal sequence, and is used to identify the time range of signal feature changes. A feature change interval set is a set of all feature change intervals, and is used to specify the time range of signal feature changes.

[0041] Preferably, the process of generating a signal feature difference sequence can be further refined. First, for each pair of adjacent feature parameters in the time domain and frequency domain, the difference between them is calculated. For example, if the feature parameter is the signal amplitude, the difference is the difference between the amplitudes of two adjacent signal segments. This difference is then compared with a preset difference threshold. The preset difference threshold can be set based on the normal fluctuation range of the signal, for example, by analyzing the variation range of the feature parameter in historical normal signal data to determine a reasonable threshold. If the difference exceeds the preset difference threshold, the difference is marked as an abnormal feature difference value. Next, based on the abnormal feature difference value and the acquisition time of the corresponding signal segment, the feature change rate is calculated. The feature change rate can be expressed as the ratio of the abnormal feature difference value to the acquisition time interval of the signal segment, which measures the speed of signal feature change. Finally, the two feature parameters corresponding to the feature change rate are determined as a feature point sequence, and the two signal segments corresponding to the feature point sequence are determined as a feature signal sequence. In this way, abnormal changes in the signal can be more accurately located and a set of feature change intervals can be generated, providing detailed time and feature information for subsequent anomaly analysis and processing.

[0042] In some embodiments, generating a set of abnormal feature intervals according to the abnormal feature sequence includes: Determining the abnormality type corresponding to each abnormal feature in the abnormal feature sequence; Determining the first two abnormal features in the abnormal feature sequence as an initial abnormal feature group; Based on the initial set of anomaly features, the following loop steps are performed: Determine the initial abnormal feature of the corresponding abnormal type characterizing the mutation in the initial abnormal feature group as the mutation feature; Determine the initial abnormal feature of the drift corresponding to the abnormal type in the initial abnormal feature group as the drift feature; Determining the signal acquisition time corresponding to the mutation feature as the third time point; Determining the signal acquisition time corresponding to the drift feature as a fourth time point; combining the third time point and the fourth time point into an abnormal time interval; In response to determining that the initial abnormal feature group is not the last two abnormal features in the abnormal feature sequence, determining two abnormal features arranged after the initial abnormal feature group in the abnormal feature sequence as the initial abnormal feature group to update the initial abnormal feature group, and executing the loop step again based on the updated initial abnormal feature group; The combined abnormal time intervals are determined as an abnormal feature interval set.

[0043] It should be noted that the process of generating an abnormal feature interval set mentioned in the present invention is achieved by classifying the abnormal features in the abnormal feature sequence and combining them in time intervals. This process aims to clarify the time range of the abnormal signal segment by identifying the type of abnormal feature (such as mutation or drift) and combining it with the corresponding signal acquisition time. The abnormal feature sequence is a sequence with abnormal features screened out in the previous step, where each abnormal feature corresponds to a specific signal segment. The abnormal type refers to the specific manifestation of the abnormal feature, such as mutation (rapid change of signal feature) or drift (slow change of signal feature). By combining these abnormal features into an interval set in chronological order, the time distribution of the abnormal signal segment can be more intuitively displayed.

[0044] Specifically, each abnormal feature in the abnormal feature sequence has its corresponding abnormal type, which can be identified by pre-set rules or models. For example, mutation features are usually manifested as rapid changes in signal features, while drift features are manifested as slow changes in signal features. The initial abnormal feature group refers to the combination of the first two abnormal features in the abnormal feature sequence, which is used to start the calculation of the abnormal time interval. In the loop step, the mutation features and drift features in the initial abnormal feature group are determined separately, and their corresponding signal acquisition times are used as time points to combine into abnormal time intervals. If the initial abnormal feature group is not the last two abnormal features in the abnormal feature sequence, the subsequent two abnormal features are updated to a new initial abnormal feature group, and the above loop steps are repeated until all abnormal features have been processed. Finally, all combined abnormal time intervals are determined as an abnormal feature interval set, which is used to clarify the time range of the abnormal signal segment.

[0045] Preferably, the process of generating a set of abnormal feature intervals can be further refined. First, each abnormal feature in the abnormal feature sequence is classified to determine whether it is a mutation feature or a drift feature. This can be achieved by setting a threshold or using a machine learning model. For example, if the feature change rate exceeds a certain threshold, it is considered a mutation feature; if the feature change rate is below a certain threshold, it is considered a drift feature. Then, the first two abnormal features in the abnormal feature sequence are determined as the initial abnormal feature group, and a loop process begins. In each loop, the signal acquisition time corresponding to the mutation feature in the initial abnormal feature group is determined as the third time point, and the signal acquisition time corresponding to the drift feature is determined as the fourth time point. These two time points are combined to form an abnormal time interval. If the initial abnormal feature group is not the last two abnormal features in the abnormal feature sequence, the next two abnormal features are updated to form a new initial abnormal feature group, and the above steps are repeated. Finally, all generated abnormal time intervals are combined in chronological order to form an abnormal feature interval set. This process can be automated through programming, improving the efficiency and accuracy of anomaly detection.

[0046] In some embodiments, generating the abnormal feature interval set according to the abnormal feature interval set and the power carrier transmission environment parameter includes: Dividing the power carrier signal according to the abnormal feature interval set to obtain an abnormal signal segment set and a normal signal segment set; For each abnormal signal segment in the abnormal signal segment set, performing feature analysis in combination with the power carrier transmission environment parameter to obtain an abnormal feature analysis result; The obtained abnormal feature analysis results and the normal signal segment set are spliced in chronological order to obtain an abnormality detection result.

[0047] It should be noted that the process of generating an abnormal feature interval set based on the abnormal feature interval set and the power carrier transmission environment parameters mentioned in the present invention is to further combine the impact of the transmission environment on the signal, so as to more accurately divide the abnormal signal segments and normal signal segments, and obtain the final abnormality detection results. The power carrier transmission environment parameters refer to environmental factors related to signal transmission, such as line length, load conditions, electromagnetic interference intensity, etc. These parameters will affect the transmission quality and characteristics of the signal. The abnormal signal segment set refers to the set of signal segments determined to be abnormal after analysis, and the normal signal segment set refers to the set of signal segments in which no abnormalities are detected. By combining the power carrier transmission environment parameters for feature analysis, the abnormality of the signal can be evaluated more comprehensively.

[0048] Specifically, the abnormal feature interval set is generated through the previous steps and identifies time intervals in the signal where anomalies may exist. Power carrier transmission environment parameters include line length, load conditions, electromagnetic interference intensity, etc. These parameters can be obtained through sensors or other monitoring equipment. When dividing the abnormal signal segment set and the normal signal segment set, the power carrier signal sequence can be divided into multiple segments based on the time intervals in the abnormal feature interval set. For each abnormal signal segment, feature analysis is performed in combination with the power carrier transmission environment parameters, such as analyzing whether the electromagnetic interference intensity is too high or whether the line load is abnormal, thereby obtaining an abnormal feature analysis result. The abnormal feature analysis result can include information such as the type and severity of the anomaly. Finally, all abnormal feature analysis results and the normal signal segment set are spliced in chronological order to obtain a complete anomaly detection result for subsequent processing and analysis.

[0049] Preferably, the process of generating a set of abnormal feature intervals can be further refined. First, the power carrier signal sequence is divided into multiple segments based on the time intervals in the abnormal feature interval set. For each abnormal signal segment, a detailed feature analysis is performed in combination with the power carrier transmission environment parameters. For example, if the electromagnetic interference intensity exceeds a preset threshold, it can be determined that the abnormal signal segment is caused by electromagnetic interference. At the same time, a similar analysis can be performed on normal signal segments to confirm their normality. During the feature analysis process, a machine learning model can be used to identify the type and severity of the anomaly. The machine learning model can be trained using historical data, with input parameters including the feature parameters of the abnormal signal segment and the power carrier transmission environment parameters, and output as the anomaly type and severity. Finally, all abnormal feature analysis results and the set of normal signal segments are spliced in chronological order to form a complete anomaly detection result. This process can be implemented using an automated algorithm to improve the efficiency and accuracy of anomaly detection.

[0050] In some embodiments, the method further comprises: Performing visualization processing on each abnormal information in the power carrier data abnormal information sequence according to each abnormal level included in the power carrier data abnormal information sequence; Based on the abnormal feature interval set, the abnormal trajectory is superimposed and displayed in the target monitoring space, wherein the target monitoring space corresponds to the transmission scenario when the power carrier signal is acquired; According to the abnormal feature interval set and the abnormal information sequence, the abnormal information sequence is visualized in the target monitoring space.

[0051] It should be noted that the process of visualizing the abnormal information sequence of power carrier data mentioned in the present invention is intended to display the abnormal information in an intuitive way so that monitoring personnel can quickly understand the abnormal situation and take corresponding measures. The abnormal information sequence refers to the data sequence containing abnormal information obtained after detection and analysis, in which the abnormal information may include the abnormal type, abnormal level, time and location of the abnormality, etc. Visualization processing refers to converting these abstract data information into graphics, charts or other visual forms to facilitate users to understand more intuitively. The target monitoring space refers to the actual physical space or virtual space related to the power carrier signal transmission, such as the geographical distribution map of the power line or the system topology map, which is used to display the geographical location and trajectory of the abnormal information.

[0052] Specifically, the abnormality level in the abnormal information sequence refers to the classification of the severity of the abnormality, for example, it can be divided into different levels such as mild, moderate, and severe. Visualization processing can be achieved in a variety of ways, such as using elements such as colors, icons, and lines to distinguish different abnormality types and levels. The target monitoring space can be a geographic information system (GIS) platform, which is used to display the geographical location of the power line and the spatial distribution of abnormal information. The abnormal trajectory refers to the propagation path or position change of the abnormal signal in the target monitoring space, which can be displayed by drawing a trajectory line on the GIS platform. Superimposing the abnormal trajectory in the target monitoring space can intuitively display the propagation path and impact range of the abnormal signal. At the same time, combining the abnormal feature interval set and the abnormal information sequence for visualization processing can more comprehensively display the temporal and spatial characteristics of the abnormal information.

[0053] Preferably, the visualization process can be further refined. First, according to the abnormality level in the abnormal information sequence, different colors or icons are selected to represent abnormalities of different levels. For example, minor abnormalities can be represented by green, moderate abnormalities by yellow, and severe abnormalities by red. Then, in the target monitoring space, the abnormal trajectory is drawn according to the time information in the abnormal feature interval set. For example, the propagation process of the abnormal signal can be displayed by drawing a trajectory line with a timestamp on the GIS platform. In addition, interactive functions can be added to the target monitoring space. For example, clicking on a certain abnormal point can display detailed abnormality information, including the abnormality type, occurrence time, duration, etc. If a model is involved, such as a model for abnormality level classification, it can be trained based on historical data. The input parameters include abnormal feature parameters and power carrier transmission environment parameters, and the output is the abnormality level. In this way, the abnormal information of the power carrier data can be displayed more intuitively, helping monitoring personnel to quickly locate and handle abnormalities.

[0054] The above-described embodiments of the present invention have the following beneficial effects: The present invention can comprehensively cover all aspects of power carrier data anomaly detection, forming a complete technical closed loop from signal feature extraction to anomaly classification. Through the collaborative analysis of time-domain and frequency-domain characteristic parameters, signal anomaly characteristics can be accurately captured; combining environmental parameters to generate a set of anomaly feature intervals can improve the interpretability of detection results; and intelligent classification based on pre-trained models can automatically output anomaly level information, providing a reliable basis for subsequent operation and maintenance decisions. This method further refines the granularity of anomaly detection. By calculating feature difference sequences and change rates, it can dynamically identify anomalous signal segments. By distinguishing between mutation and drift features, it can accurately locate the anomaly type. Furthermore, by intelligently splicing anomalous and normal signal segments, it can generate structured detection reports. Combined with visualization technology, it can integrate anomaly trajectories with monitoring scenarios, significantly enhancing operations personnel's perception of system status.

[0055] like Figure 2 As shown, some embodiments provide a power line carrier data anomaly detection device, the device comprising: An acquisition unit 201 is configured to acquire a power carrier signal sequence and power carrier transmission environment parameters; The feature extraction unit 202 is configured to extract time domain feature parameters and frequency domain feature parameters according to the power carrier signal sequence; A feature generating unit 203 is configured to generate an abnormal feature sequence corresponding to the power carrier signal based on the time domain feature parameters and the frequency domain feature parameters; The detection unit 204 is configured to generate an abnormal feature interval set according to the abnormal feature sequence and the power carrier transmission environment parameter; The information generating unit 205 is configured to generate a power carrier data abnormality information sequence based on the abnormal feature interval set and a pre-trained abnormality classification model.

[0056] It is understandable that the modules recorded in the power carrier data anomaly detection device are similar to those in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the power carrier data anomaly detection method are also applicable to the power carrier data anomaly detection device and the modules contained therein, and will not be repeated here.

[0057] Reference below Figure 3 , which shows a schematic structural diagram of an electronic device 300 suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 3 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0058] like Figure 3As shown, electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 302 or programs loaded from a storage device 308 into a random access memory (RAM) 303. RAM 303 also stores various programs and data required for the operation of electronic device 300. Processing device 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to bus 304.

[0059] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as needed.

[0060] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or a terminal device such as a computer, server, mobile phone, or tablet.

[0061] The above descriptions merely illustrate some preferred embodiments of the present invention and the underlying technical principles. Those skilled in the art should understand that the scope of the invention encompassed by the embodiments of the present invention is not limited to technical solutions formed by specific combinations of the aforementioned technical features. It also encompasses other technical solutions formed by any combination of the aforementioned technical features or their equivalents, without departing from the aforementioned inventive concept. For example, a technical solution formed by replacing the aforementioned features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A method for detecting abnormality in power line carrier data, characterized in that: include: Obtaining power carrier signal sequence and power carrier transmission environment parameters; Extracting time domain characteristic parameters and frequency domain characteristic parameters according to the power carrier signal sequence; generating an abnormal characteristic sequence corresponding to the power carrier signal based on the time domain characteristic parameters and the frequency domain characteristic parameters; generating an abnormal feature interval set according to the abnormal feature sequence and the power carrier transmission environment parameter; Based on the abnormal feature interval set and the pre-trained abnormal classification model, a power line carrier data abnormal information sequence is generated.

2. The method according to claim 1, characterized in that The extracting time domain characteristic parameters and frequency domain characteristic parameters according to the power carrier signal sequence includes: Performing time domain analysis on each signal segment in the power carrier signal sequence to determine each time domain characteristic parameter characterizing the time domain characteristics of the signal; Determining the determined time domain characteristic parameters characterizing the time domain characteristics of the signal as time domain characteristic parameters; Frequency domain analysis is performed on each signal segment in the power carrier signal sequence to obtain frequency domain characteristic parameters, wherein characteristic values in the frequency domain characteristic parameters correspond to signal segments in the power carrier signal sequence.

3. The method according to claim 1, characterized in that The generating, based on the time domain characteristic parameters and the frequency domain characteristic parameters, an abnormal characteristic sequence corresponding to the power carrier signal includes: generating a signal feature difference sequence according to the time domain feature parameters and the frequency domain feature parameters; generating an abnormal feature sequence according to the signal feature difference sequence; According to the abnormal feature sequence, a set of abnormal feature intervals corresponding to the power carrier signal is generated.

4. The method according to claim 3, characterized in that Generating a signal feature difference sequence according to the time domain feature parameters and the frequency domain feature parameters includes: For each two adjacent characteristic parameters of the time domain characteristic parameters and the frequency domain characteristic parameters, the following steps are performed: determining a difference between the two characteristic parameters as a characteristic difference value; Determining whether the feature difference value is greater than a preset difference threshold; In response to determining that the feature difference value is greater than the preset difference threshold, determining the feature difference value as an abnormal feature difference value; generating a characteristic change rate based on the abnormal characteristic difference value and the signal time difference corresponding to the power carrier signal sequence; Determine the generated individual feature change rates as a feature change rate set; For each feature change rate in the feature change rate set, perform the following steps: Determining two characteristic parameters corresponding to the characteristic change rate as a characteristic point sequence; determining two signal segments corresponding to the characteristic point sequence as a characteristic signal sequence; Determining the acquisition time corresponding to the first signal segment in the characteristic signal sequence as the first time point; Determining the acquisition time corresponding to the second signal segment in the characteristic signal sequence as the second time point; combining the first time point and the second time point into a feature change interval; The combined feature change intervals are determined as a feature change interval set.

5. The method according to claim 3, characterized in that Generating an abnormal feature interval set according to the abnormal feature sequence includes: Determining the abnormality type corresponding to each abnormal feature in the abnormal feature sequence; Determining the first two abnormal features in the abnormal feature sequence as an initial abnormal feature group; Based on the initial set of anomaly features, the following loop steps are performed: Determine the initial abnormal feature of the corresponding abnormal type characterizing the mutation in the initial abnormal feature group as the mutation feature; Determine the initial abnormal feature of the drift corresponding to the abnormal type in the initial abnormal feature group as the drift feature; Determining the signal acquisition time corresponding to the mutation feature as the third time point; Determining the signal acquisition time corresponding to the drift feature as a fourth time point; combining the third time point and the fourth time point into an abnormal time interval; In response to determining that the initial abnormal feature group is not the last two abnormal features in the abnormal feature sequence, determining two abnormal features in the abnormal feature sequence that are arranged after the initial abnormal feature group as the initial abnormal feature group to update the initial abnormal feature group, and executing the loop step again based on the updated initial abnormal feature group; The combined abnormal time intervals are determined as an abnormal feature interval set.

6. The method according to claim 1, characterized in that The generating of the abnormal feature interval set according to the abnormal feature interval set and the power carrier transmission environment parameter includes: Dividing the power carrier signal according to the abnormal feature interval set to obtain an abnormal signal segment set and a normal signal segment set; For each abnormal signal segment in the abnormal signal segment set, performing feature analysis in combination with the power carrier transmission environment parameter to obtain an abnormal feature analysis result; The obtained abnormal feature analysis results and the normal signal segment set are spliced in chronological order to obtain an abnormality detection result.

7. The method according to claim 1, characterized in that The method further comprises: Performing visualization processing on each abnormal information in the power carrier data abnormal information sequence according to each abnormal level included in the power carrier data abnormal information sequence; Based on the abnormal feature interval set, the abnormal trajectory is superimposed and displayed in the target monitoring space, wherein the target monitoring space corresponds to the transmission scenario when the power carrier signal is acquired; According to the abnormal feature interval set and the abnormal information sequence, the abnormal information sequence is visualized in the target monitoring space.

8. A power line carrier data anomaly detection device, characterized in that: include: an acquisition unit configured to acquire a power carrier signal sequence and a power carrier transmission environment parameter; a feature extraction unit configured to extract time domain feature parameters and frequency domain feature parameters according to the power carrier signal sequence; a feature generating unit configured to generate an abnormal feature sequence corresponding to the power carrier signal based on the time domain feature parameter and the frequency domain feature parameter; a detection unit configured to generate an abnormal feature interval set according to the abnormal feature sequence and the power carrier transmission environment parameter; The information generating unit is configured to generate a power carrier data abnormality information sequence based on the abnormal feature interval set and a pre-trained abnormality classification model.

9. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer-readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images