SVM-based power communication equipment anomaly detection method and system, and medium

Abstract

The invention relates to an SVM-based power communication equipment anomaly detection method and system and a medium, and the method comprises the steps: collecting the data of various sensors in power communication equipment, including a current sensor, a voltage sensor, a temperature sensor, a humidity sensor, a vibration sensor and the like; all the data are preprocessed, and the data are marked. Feature samples in the data are extracted, and a training set is constructed; and carrying out training by using the optimized SVM algorithm, and generating an anomaly detection model. And extracting data of the to-be-detected electric power communication equipment, inputting the data into the anomaly detection model, and judging an abnormal state of the electric power communication equipment according to a model output result. Compared with a conventional electric power communication equipment anomaly detection method, the electric power communication equipment anomaly detection method adopts a supervised learning method and quantifies the data characteristic indexes, so that the abnormal state of the electric power communication equipment can be accurately and efficiently detected in multiple dimensions.

Description

Technical Field

[0001] The present application relates to the field of power system security, and in particular to a method, system and medium for detecting abnormality of power communication equipment based on SVM. Background Art

[0002] With the growing demand for power supply in modern society, power communication equipment plays a vital role in the power system. These devices are used to monitor, control and protect the power system to ensure the reliability and stability of power supply. However, power communication equipment itself is also subject to a variety of faults and abnormal conditions, which may have a serious impact on the operation of the power system. Therefore, the development of power communication equipment anomaly detection technology has become one of the important tasks in the power field.

[0003] Due to the complexity and diversity of power communication equipment, how to accurately detect anomalies in equipment is crucial to maintaining the stability of the power system. According to existing research, there are several main problems with anomaly detection of power communication equipment. First, data quality issues. The data format and accuracy of different equipment manufacturers and models are inconsistent, which may lead to difficulties in data integration and analysis. Second, the generalization ability and stability of the model need to be further improved to cope with the ever-changing equipment and working environment. Third, privacy and security issues, because anomaly detection of power communication equipment usually involves sensitive power system data. Fourth, due to the complexity and diversity of power communication equipment, it is difficult to accurately extract features that can detect the status of equipment in multiple dimensions. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, system and medium for detecting abnormalities in power communication equipment based on SVM, by collecting data from various sensors in the power communication equipment, mainly including current sensors, voltage sensors, temperature sensors, humidity sensors and vibration sensors. And pre-process the collected various sensor data and mark the data. At the same time, extract feature samples from the data and construct a training set. Use the optimized SVM algorithm for training, and generate an abnormality detection model and decision threshold. Finally, extract the data of the power communication equipment to be tested and input it into the abnormality detection model, and judge the abnormal state of the power communication equipment according to the output result of the model.

[0005] To achieve the above objectives, this application provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present application provides a method for detecting abnormality of a power communication device based on SVM, comprising the following steps:

[0007] 1) Collect data from various sensors in power communication equipment, including current sensors, voltage sensors, temperature sensors, humidity sensors, and vibration sensors.

[0008] 2) Preprocess the various sensor data collected in step 1) and mark the data.

[0009] 3) Extract feature samples from the data and construct a training set.

[0010] 4) Use the optimized SVM algorithm for training and generate anomaly detection model.

[0011] 5) Extract the data of the power communication equipment to be tested and input it into the anomaly detection model, and judge the abnormal state of the power communication equipment according to the output result of the model.

[0012] In step 1), various sensor data in the power communication equipment are collected, mainly data of current sensors, voltage sensors, temperature sensors, humidity sensors and vibration sensors.

[0013] Data from current sensors, voltage sensors, temperature sensors, humidity sensors, and vibration sensors are collected. The current sensor collects current values ​​to monitor the current load and current waveform of the power equipment in order to detect problems such as overload and short circuit; the voltage sensor collects voltage values ​​to monitor the voltage quality and voltage stability of the power system; the temperature sensor collects temperature data of the equipment or environment to monitor the temperature change and heat distribution of the equipment to prevent overheating or overcooling. The humidity sensor collects humidity data to monitor the ambient humidity to ensure that the equipment is not affected by moisture or excessive dryness; the vibration sensor collects vibration data to monitor the vibration and vibration of the equipment to detect whether the equipment is impacted or damaged.

[0014] In step 2), the various sensor data collected in step 1) need to be preprocessed and labeled. The main approach is to distinguish the normal and abnormal data samples from the collected initial data, and label the normal and abnormal data samples separately. At the same time, in order to ensure the data quality, the following method is used for processing:

[0015] 1) Handle missing values ​​in the data, including deleting missing values ​​and filling missing values ​​with the mean or median.

[0016] 2) Normalize different feature data to avoid certain features from having too much influence on the model.

[0017] In step 3), extract feature samples from the data and construct a training set. First, define F as a feature matrix, f as a specific feature, and the extracted features include: current value, voltage value, temperature, humidity, and vibration frequency. Since the data at a certain moment may be disturbed and the state of the power communication equipment cannot be clearly determined, the data features of five moments are extracted to form a 5*5 feature vector matrix, and each feature vector matrix is ​​labeled to indicate whether the data sample is abnormal:

[0018]

[0019] In step 4), based on the data set constructed in step 3), the optimized SVM algorithm is used for training and an anomaly detection model is generated. The anomaly detection model is trained through the following process:

[0020] 1) Data preparation: Normal samples construct a training data set (X, y), where X is the feature matrix, y is the label vector, each row of X is a sample, and each element of y corresponds to the category label of the sample.

[0021] 2) The optimized SVM objective is to minimize the following objective function, where ω is the weight vector of the decision boundary, b is the bias term, C is the regularization parameter, and ξ i is the slack variable and n is the number of samples:

[0022]

[0023] The slack variable ξ i Some samples are allowed not to meet the constraints, and their number is controlled by the regularization parameter C. When C is large, the penalty for slack variables is large, so it is hoped that the samples will meet the constraints as much as possible.

[0024] 3) The constraint of SVM is to ensure that each sample is correctly classified. is the data point x i The feature map is:

[0025]

[0026] 4) Based on the validation dataset, set an appropriate decision threshold for the trained anomaly detection model to determine when to mark a new data point as an anomaly.

[0027] In step 5), the data of the power communication device to be tested is extracted and input into the anomaly detection model, the threshold is compared according to the model output result, the classification result is output, and the abnormal state of the power communication device is further determined according to the classification result.

[0028] In a second aspect, an embodiment of the present application provides a power communication equipment abnormality detection system based on SVM, which is used to implement the sharing method as described above, and is characterized by comprising a data acquisition unit, a data processing unit and a data monitoring unit;

[0029] The data collection unit first distinguishes the electricity consumption types of users in different areas, and uses the sensor group to collect the original electricity data in different areas; the data processing unit calculates the normal distribution density probability of the encrypted data in the data set, establishes a data encryption index, and analyzes the data encryption data index to generate key monitoring signals and general monitoring signals respectively; the data monitoring unit monitors each data collection node during the data collection process. After receiving the key monitoring signal during the data transmission process, the correlation coefficient between the data collection nodes is established to compare the correlation between the data of the two data nodes, and an early warning is issued for the data transmission nodes that are abnormal or missing during the data transmission process.

[0030] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a program code, and when the program code is executed by a processor, the steps of the SVM-based power communication equipment abnormality detection method as described above are implemented.

[0031] Compared with the prior art, the beneficial effect of the present application is: compared with the existing method for detecting abnormalities in power communication equipment, the present invention extracts sensor data with representative features as a data set, and adopts a supervised learning method to quantify data feature indicators, thereby being able to accurately and efficiently detect abnormal states of power communication equipment in multiple dimensions. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 This is a schematic diagram of the model based on SVM in the present invention;

[0034] Figure 2 This is a flow chart of abnormality detection of power communication equipment based on SVM in the present invention. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0036] The terms "comprises," "comprising," or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0037] The terms "first", "second", etc. are only used to distinguish one entity or operation from another entity or operation, and should not be understood as indicating or implying relative importance, nor should they be understood as requiring or implying any such actual relationship or order between these entities or operations.

[0038] like Figure 1-2 As shown, the present invention provides a method and system for detecting abnormality of power communication equipment based on SVM, comprising the following steps:

[0039] Step 1: Collect data from various sensors in the power communication equipment, mainly current sensors, voltage sensors, temperature sensors, humidity sensors and vibration sensors. The current sensor collects current values ​​to monitor the current load and current waveform of the power equipment in order to detect problems such as overload and short circuit; the voltage sensor collects voltage values ​​to monitor the voltage quality and voltage stability of the power system; the temperature sensor collects temperature data of the equipment or environment to monitor the temperature change and heat distribution of the equipment to prevent overheating or overcooling. The humidity sensor collects humidity data to monitor the ambient humidity to ensure that the equipment is not affected by moisture or excessive dryness; the vibration sensor collects vibration data to monitor the vibration and vibration of the equipment to detect whether the equipment is impacted or damaged.

[0040] Step 2: Preprocess the various sensor data collected in step 1 and label the data. The main approach is to distinguish the normal and abnormal data samples from the collected initial data, and label the normal and abnormal data samples separately. At the same time, in order to ensure the quality of the data, the following methods are used for processing:

[0041] 1) Handle missing values ​​in the data, including deleting missing values ​​and filling missing values ​​with the mean or median.

[0042] 2) Normalize different feature data to avoid certain features from having too much influence on the model.

[0043] Step 3: Extract feature samples from the data and construct a training set. First, define F as a feature matrix, f as a specific feature, and the extracted features include: current value, voltage value, temperature, humidity, and vibration frequency. Since the data at a certain moment may be disturbed and the state of the power communication equipment cannot be clearly determined, the data features of the five moments are extracted to form a 5*5 feature vector matrix, and each feature vector matrix is ​​labeled to indicate whether the data sample is abnormal:

[0044]

[0045] Step 4: Based on the data set constructed in step 3, use the optimized SVM algorithm for training and generate an anomaly detection model. The anomaly detection model is trained through the following process:

[0046] 1) Data preparation: Normal samples construct a training data set (X, y), where X is the feature matrix, y is the label vector, each row of X is a sample, and each element of y corresponds to the category label of the sample.

[0047] 2) The optimized SVM objective is to minimize the following objective function, where ω is the weight vector of the decision boundary, b is the bias term, C is the regularization parameter, and ξ i is the slack variable and n is the number of samples:

[0048]

[0049] The slack variable ξ i Some samples are allowed not to meet the constraints, and their number is controlled by the regularization parameter C. When C is large, the penalty for slack variables is large, so it is hoped that the samples will meet the constraints as much as possible.

[0050] 3) The constraint of SVM is to ensure that each sample is correctly classified. is the data point x i The feature map is:

[0051]

[0052] 4) Based on the validation dataset, set an appropriate decision threshold for the trained anomaly detection model to determine when to mark a new data point as an anomaly.

[0053] Step 5: Extract the data of the power communication equipment to be tested and input it into the anomaly detection model, compare the threshold value according to the model output result, output the classification result, and further judge the abnormal state of the power communication equipment according to the classification result.

[0054] The present invention discloses an abnormality detection method for power communication equipment based on SVM. The method collects data from various sensors in the power communication equipment, mainly including current sensors, voltage sensors, temperature sensors, humidity sensors, vibration sensors, etc. The collected data from various sensors are preprocessed and the data is marked. At the same time, feature samples in the data are extracted and a training set is constructed. The optimized SVM algorithm is used for training, and an abnormality detection model and a decision threshold are generated. Finally, the data of the power communication equipment to be tested is extracted and input into the abnormality detection model, and the abnormal state of the power communication equipment is judged according to the output result of the model.

[0055] The embodiment of the present application provides a power communication equipment abnormality detection system based on SVM, which is used to implement the sharing method as described above, and is characterized by comprising a data acquisition unit, a data processing unit and a data monitoring unit;

[0056] The data collection unit first distinguishes the electricity consumption types of users in different areas, and uses the sensor group to collect the original electricity data in different areas; the data processing unit calculates the normal distribution density probability of the encrypted data in the data set, establishes a data encryption index, and analyzes the data encryption data index to generate key monitoring signals and general monitoring signals respectively; the data monitoring unit monitors each data collection node during the data collection process. After receiving the key monitoring signal during the data transmission process, the correlation coefficient between the data collection nodes is established to compare the correlation between the data of the two data nodes, and an early warning is issued for the data transmission nodes that are abnormal or missing during the data transmission process.

[0057] An embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores program code, and when the program code is executed by a processor, the steps of the above-mentioned method for detecting abnormality of power communication equipment based on SVM are implemented.

[0058] An embodiment of the present application provides a computer-readable storage medium, which stores program code. When the program code is executed by a processor, the steps of the method for securely sharing power data as described above are implemented.

[0059] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0060] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0061] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0062] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0063] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0064] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0065] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0066] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting abnormality of power communication equipment based on SVM, characterized in that: The following steps are involved: 1) Collect data from various sensors in power communication equipment; 2) Preprocessing the various sensor data collected in step 1) and marking the data; 3) Extract feature samples from the data and construct a training set; 4) Use the optimized SVM algorithm for training and generate anomaly detection model; 5) Extract the data of the power communication equipment to be tested and input it into the anomaly detection model, and judge the abnormal state of the power communication equipment according to the output result of the model.

2. The method for detecting abnormality of power communication equipment based on SVM according to claim 1, characterized in that: In step 1), various sensor data in the power communication equipment are collected, mainly data of current sensors, voltage sensors, temperature sensors, humidity sensors and vibration sensors.

3. The method for detecting abnormality of power communication equipment based on SVM according to claim 2, characterized in that: The current sensor collects current values ​​for monitoring the current load and current waveform of the power equipment in order to detect overload and short circuit problems; the voltage sensor collects voltage values ​​for monitoring the voltage quality and voltage stability of the power system; The temperature sensor collects temperature data of the device or environment to monitor the temperature changes and heat distribution of the device to prevent overheating or overcooling. The humidity sensor collects humidity data to monitor the ambient humidity to ensure that the device is not affected by moisture or excessive dryness. The vibration sensor collects vibration data to monitor the vibration and vibration of the device to detect whether the device is impacted or damaged.

4. The method for detecting abnormality of power communication equipment based on SVM according to claim 1, characterized in that: In step 2), the various sensor data collected in step 1) need to be preprocessed and the data need to be labeled. Specifically, the collected initial data is distinguished into normal and abnormal data samples, and the normal and abnormal data samples are labeled with different categories. In order to ensure the data quality, the following method is used for processing: 1) Handle missing values ​​in the data, including deleting missing values ​​and filling them with mean or median values; 2) Normalize different feature data to avoid certain features from having too much influence on the model.

5. The method for detecting abnormality of power communication equipment based on SVM according to claim 1, characterized in that: In step 3), feature samples in the data are extracted and a training set is constructed. Specifically, F is first defined as a feature matrix, f is a specific feature, and the extracted features include: current value, voltage value, temperature, humidity and vibration frequency. Since the data at a certain moment may be disturbed and the state of the power communication equipment cannot be clearly determined, the data features at five moments are extracted to form a 5*5 feature vector matrix, and each feature vector matrix is ​​labeled to indicate whether the data sample is abnormal:

6. The method for detecting abnormality of power communication equipment based on SVM according to claim 1, characterized in that: In step 4), based on the data set constructed in step 3), the optimized SVM algorithm is used for training and an anomaly detection model is generated, wherein the anomaly detection model is trained through the following process: 1) Data preparation: Normal samples are used to construct a training data set (X, y), where X is the feature matrix, y is the label vector, each row of X is a sample, and each element of y corresponds to the category label of the sample; 2) The optimized SVM objective is to minimize the following objective function, where ω is the weight vector of the decision boundary, b is the bias term, C is the regularization parameter, and ξ i is the slack variable and n is the number of samples: The slack variable ξ i Some samples are allowed not to meet the constraints, and their number is controlled by the regularization parameter C. When C is large, the penalty for slack variables is large, so it is hoped that the samples meet the constraints as much as possible. 3) The constraint of SVM is to ensure that each sample is correctly classified. is the data point x i The feature map is: 4) Based on the validation dataset, set an appropriate decision threshold for the trained anomaly detection model to determine when to mark a new data point as an anomaly.

7. The method for detecting abnormality of power communication equipment based on SVM according to claim 1, characterized in that: In step 5), the data of the power communication device to be tested is extracted and input into the anomaly detection model, the threshold is compared according to the model output result, the classification result is output, and the abnormal state of the power communication device is further determined according to the classification result.

8. A power communication equipment anomaly detection system based on SVM, used to implement the method according to any one of claims 1 to 7, characterized in that: It includes a data acquisition unit, a data processing unit and a data monitoring unit; The data collection unit first distinguishes the electricity consumption types of users in different areas, and uses the sensor group to collect the original electricity data in different areas; the data processing unit calculates the normal distribution density probability of the encrypted data in the data set, establishes a data encryption index, and analyzes the data encryption data index to generate key monitoring signals and general monitoring signals respectively; the data monitoring unit monitors each data collection node during the data collection process, and after receiving the key monitoring signal during the data transmission process, it establishes a correlation coefficient between the data collection nodes to compare the correlation between the data of the two data nodes, and issues early warnings for data transmission nodes that are abnormal or missing during the data transmission process.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, and when the program codes are executed by a processor, the steps of the SVM-based power communication equipment abnormality detection method according to any one of claims 1 to 7 are implemented.

Images