Intelligent power grid fault monitoring method and system applying artificial intelligence

Abstract

The invention discloses an intelligent power grid fault monitoring method and system applying artificial intelligence, and relates to the technical field of power system fault monitoring, and the method comprises the steps: obtaining power internet-of-things monitoring data through a sensor, carrying out the preprocessing of the data, extracting fault-related features based on the preprocessed data through a machine learning algorithm, and carrying out the fault-related features; based on historical data of the power grid, a preliminary fault monitoring threshold range is determined through statistical analysis; and comparing the fault feature vector with real-time monitoring power grid operation state data, dynamically adjusting a fault monitoring threshold value by using an adaptive algorithm based on a comparison result of historical data and real-time state data of the power grid, giving an alarm after an abnormal condition is monitored, and optimizing a fault monitoring mechanism according to feedback of a fault processing result. According to the invention, through data preprocessing, noise removal is realized, the monitoring precision is improved, and the effect of improving the fault detection accuracy is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system fault monitoring, and in particular to a smart power grid fault monitoring method and system using artificial intelligence. Background Art

[0002] Traditional power grid fault detection methods mostly rely on manual inspections and simple automated equipment. This method is not only inefficient, but also difficult to detect hidden faults in complex power grid structures in a timely manner. In recent years, with the development of Internet of Things technology, a large number of sensors have been deployed in power systems to achieve real-time monitoring of the power grid operation status. This not only provides a rich data source for fault detection, but also lays the foundation for the realization of intelligent fault diagnosis.

[0003] Traditional fault detection methods often lack effective data preprocessing steps, resulting in noise data and irrelevant data affecting the accuracy of fault detection. Secondly, in the feature extraction process, many methods do not fully utilize the multi-dimensional information in the power grid operation data, resulting in incomplete feature expression, which in turn affects the effect of fault identification. After the current fault monitoring system detects an anomaly, it lacks an effective feedback mechanism to continuously optimize the monitoring strategy, which limits the system's adaptive ability. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a smart grid fault monitoring method and system using artificial intelligence to solve the problems of insufficient data preprocessing, incomplete feature extraction, poor adaptability of fixed thresholds, and lack of effective feedback optimization mechanism in existing smart grid fault monitoring technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a smart grid fault monitoring method using artificial intelligence, which includes:

[0008] Obtain power IoT monitoring data through sensors and pre-process it;

[0009] Based on the preprocessed data, machine learning algorithms are used to extract fault-related features and fuse them into a unified fault feature vector;

[0010] Based on the historical data of the power grid, the preliminary fault monitoring threshold range is determined through statistical analysis; and the fault feature vector is compared with the real-time monitoring power grid operation status data;

[0011] Based on the comparison results of historical data and real-time status data of the power grid, the fault monitoring threshold is dynamically adjusted using an adaptive algorithm;

[0012] An alarm is issued after an abnormal situation is detected, and the fault monitoring mechanism is optimized based on the feedback of the fault handling results.

[0013] As a preferred solution of the smart grid fault monitoring method using artificial intelligence described in the present invention, wherein: the sensor includes a temperature sensor, a current sensor, and a voltage sensor;

[0014] The preprocessing includes data cleaning and data denoising;

[0015] The real-time status data includes grid load and temperature changes.

[0016] As a preferred solution of the smart grid fault monitoring method using artificial intelligence described in the present invention, wherein: the power Internet of Things monitoring data is obtained through sensors and preprocessed, the specific steps are as follows:

[0017] Deploy temperature sensors, current sensors, and voltage sensors at power grid substations and transmission line nodes to monitor the power grid in real time and send data to the central data processing center in real time;

[0018] The central data processing center receives the raw data, cleans it, denoises it, removes invalid data, and uses a low-pass filter to smooth the current and voltage signals and remove noise.

[0019] As a preferred solution of the smart grid fault monitoring method using artificial intelligence described in the present invention, wherein: based on the preprocessed data, the machine learning algorithm is used to extract fault-related features and fuse them into a unified fault feature vector. The specific steps are as follows:

[0020] A method combining principal component analysis and convolutional neural network is used to extract fault-related features from the preprocessed data. The expression is as follows:

[0021]

[0022] Among them, F is the extracted fault feature vector, D is the preprocessed data set, PCA (D i ) is through D i The main components obtained by principal component analysis, D i is the data of the i-th dimension in the dataset D, i is the index variable, CNN(D i ) is obtained from D through convolutional neural network i The features extracted from Var(D i ) is D i The variance of

[0023] Feature fusion is performed based on the mutual information between features. The expression is as follows:

[0024]

[0025] Among them, V final is the fused fault feature vector, F i ,F j are the feature vectors extracted from two different dimensions, MI(F i ,F j ) is F i ,F j The mutual information between them.

[0026] As a preferred solution of the smart grid fault monitoring method using artificial intelligence described in the present invention, wherein: based on the historical data of the power grid, the preliminary fault monitoring threshold range is determined by statistical analysis, and the fault feature vector is compared with the real-time monitoring power grid operation status data. The specific steps are as follows:

[0027] Summarize the feature vectors of normal state and fault state in the historical data of the power grid. The distribution of feature vectors in normal operation state follows normal distribution, while the distribution of feature vectors in fault state will be significantly different.

[0028] Calculate the mean and standard deviation of the fault feature vector under normal conditions, and set the initial fault monitoring threshold range based on the principle of three times the standard deviation;

[0029] The grid operation status data obtained by real-time monitoring is converted into a fault feature vector. If the fault feature vector obtained by real-time monitoring data is within the fault threshold range, it is considered that the grid operation status is good, otherwise it is considered that there is a fault in the grid operation.

[0030] As a preferred solution of the smart grid fault monitoring method using artificial intelligence described in the present invention, wherein: the fault monitoring threshold is dynamically adjusted using an adaptive algorithm based on the comparison result of the historical data and the real-time status data of the power grid. The specific steps are as follows:

[0031] Based on the trend change of real-time data and the statistical characteristics of historical data, an adaptive adjustment factor α is defined to reflect the degree of deviation of real-time status data relative to historical data. The expression is as follows:

[0032]

[0033] Among them, α(t) is the adaptive adjustment factor at time point t, V final (t) is the fault feature vector after fusion at time point t, μ n and σ nare the mean and standard deviation of the fault feature vector under normal conditions in the historical data, t0 and t are the time points for starting and ending the real-time data collection, and dt refers to the integration of the absolute value of the feature vector’s deviation from the normal mean in the time interval [t0, t];

[0034] The adaptive adjustment factor α(t) is used to adjust the fault detection threshold. The expression is as follows:

[0035] T(t)=μ n ±3σ n +α(t);

[0036] Where T(t) is the dynamic fault monitoring threshold at time point t.

[0037] As a preferred solution of the smart grid fault monitoring method using artificial intelligence described in the present invention, wherein: an alarm is issued after an abnormal situation is detected, and the fault monitoring mechanism is optimized according to the feedback of the fault handling result. The specific steps are as follows:

[0038] When the fault feature vector is detected to exceed the dynamic fault monitoring threshold, the early warning mechanism is triggered, the redundant monitoring program is started, and the extraction and comparison process of the fault feature vector is repeated for secondary verification;

[0039] When a fault is confirmed, the log records the time, location, type and fault feature vector of the fault and sends it to the control center to activate the emergency response and redistribute the grid load;

[0040] During the fault handling process, the impact data of the handling actions and power grid status are continuously collected and compared with the pre-set monitoring process, the handling effect is evaluated, and the monitoring process is adjusted according to the actual situation.

[0041] In a second aspect, the present invention provides a smart grid fault monitoring system using artificial intelligence, including a data acquisition and preprocessing module, a feature extraction and fusion module, a fault monitoring threshold setting and comparison module, an adaptive threshold adjustment module and an alarm and feedback optimization module;

[0042] The data acquisition and preprocessing module is used to obtain power IoT monitoring data through sensors and preprocess it;

[0043] The feature extraction and fusion module is used to extract fault-related features based on the preprocessed data using a machine learning algorithm and fuse them into a unified fault feature vector;

[0044] The fault monitoring threshold setting and comparison module is used to determine the preliminary fault monitoring threshold range through statistical analysis based on the historical data of the power grid, and compare it with the real-time monitoring power grid operation status data according to the fault feature vector;

[0045] The adaptive threshold adjustment module is used to dynamically adjust the fault monitoring threshold using an adaptive algorithm based on the comparison result between the historical data and the real-time status data of the power grid;

[0046] The alarm and feedback optimization module is used to issue an alarm after detecting an abnormal situation and optimize the fault monitoring mechanism according to the feedback of the fault handling result.

[0047] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the smart grid fault monitoring method using artificial intelligence as described in the first aspect of the present invention is implemented.

[0048] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the smart grid fault monitoring method using artificial intelligence as described in the first aspect of the present invention.

[0049] The beneficial effects of the present invention are as follows: by deploying a variety of sensors and performing data preprocessing, the present invention realizes comprehensive monitoring of the operation status of the power grid and improvement of data quality, reduces false alarms, and improves monitoring accuracy. By utilizing a machine learning algorithm to extract fault features from preprocessed data and fuse them into a unified fault feature vector, a high degree of generalization and integration of fault features is achieved, the accuracy and classification capabilities of fault identification are enhanced, and the sensitivity of fault detection is improved. By statistically analyzing historical data to determine a preliminary fault monitoring threshold and comparing it with real-time data, real-time monitoring of the operation status of the power grid and rapid response to faults are achieved, the impact of faults is reduced, and the fault monitoring threshold is dynamically adjusted through an adaptive algorithm, self-optimization and enhanced adaptability of the monitoring system are achieved, and system stability and adaptability to complex environments are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0051] Figure 1 This is a flow chart of the smart grid fault monitoring method using artificial intelligence in Example 1.

[0052] Figure 2 This is a module diagram of the smart grid fault monitoring system using artificial intelligence in Example 1. DETAILED DESCRIPTION

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0055] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0056] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a smart grid fault monitoring method using artificial intelligence, comprising the following steps:

[0057] S1. Obtain power IoT monitoring data through sensors and pre-process it.

[0058] Furthermore, temperature sensors, current sensors, and voltage sensors are deployed at power grid substations and transmission line nodes to monitor the power grid in real time and send the data to the central data processing center in real time;

[0059] The central data processing center receives the raw data and performs cleaning and denoising to remove invalid data, using a low-pass filter to smooth the current and voltage signals and remove noise;

[0060] S2. Based on the preprocessed data, the machine learning algorithm is used to extract fault-related features and fuse them into a unified fault feature vector.

[0061] Furthermore, the cleaned and denoised data are standardized to ensure that different types of sensor data can be compared in the same dimension;

[0062] Preferably, a method combining principal component analysis (PCA) and convolutional neural network (CNN) is used, PCA is used to reduce the dimension of data and reduce redundant information; while CNN is good at capturing local patterns in data, which is particularly useful for detecting anomalies in power systems;

[0063] Extract fault-related features from the preprocessed data. The expression is as follows:

[0064]

[0065] Among them, F is the extracted fault feature vector, D is the preprocessed data set, PCA (D i ) is through D i The main components obtained by principal component analysis, D i is the data of the i-th dimension in the dataset D, i is the index variable, CNN(D i ) is obtained from D through convolutional neural network i The features extracted from Var(D i ) is D i The variance of is used to measure the importance of features;

[0066] The reciprocal square root of the variance is used as the criterion for feature selection. The larger the variance, the more stable the feature, and therefore the smaller its contribution to the final feature vector. This adaptive mechanism makes those features with larger fluctuations (i.e. more sensitive) contribute more to the final feature vector.

[0067] Furthermore, the relationship between features is measured based on the mutual information (MI) between features, and feature fusion is performed. The expression is as follows:

[0068]

[0069] Among them, V final is the fused fault feature vector, F i ,F j are the feature vectors extracted from two different dimensions, MI(F i ,F j ) is F i ,F j The mutual information between them measures the strength of the dependency between two features;

[0070] V final The value range of is (0,1). The closer the value is to 1, the more important the feature is for fault identification, while the value close to 0 means that the feature has a weaker effect in fault monitoring.

[0071] It should be noted that through the effective extraction and fusion of fault features, a feature vector is also constructed that can accurately reflect the health status of the power system, effectively improving the accuracy and efficiency of fault detection, thereby realizing real-time and precise monitoring of the health status of the power system.

[0072] S3. Based on the historical data of the power grid, determine the preliminary fault monitoring threshold range through statistical analysis, and compare it with the real-time monitoring power grid operation status data based on the fault feature vector.

[0073] Furthermore, the feature vectors of normal state and fault state in the historical data of the power grid are summarized. The distribution of feature vectors in the normal operation state follows the normal distribution, while the distribution of feature vectors in the fault state has significant differences.

[0074] Calculate the mean and standard deviation of the fault feature vector under normal conditions. The expressions are as follows:

[0075]

[0076]

[0077] Where μ is the mean of the fault feature vector under normal operating conditions, N is the number of samples under normal operating conditions, and σ is the standard deviation of the fault feature vector under normal operating conditions. is the fault feature vector of the i-th sample;

[0078] According to the principle of three times standard deviation, the initial fault monitoring threshold range is set, [μ-3σ,μ+3σ]. The feature vectors beyond this range are considered to be in a possible fault state.

[0079] Furthermore, the grid operation status data obtained by real-time monitoring is converted into a fault feature vector. If the fault feature vector obtained by the real-time monitoring data is within the fault threshold range, it is considered that the grid operation status is good, otherwise it is considered that there is a fault in the grid operation;

[0080] It should be noted that through real-time monitoring and comparison with the fault detection threshold range derived from historical data analysis, potential faults in the power system can be identified in a timely and accurate manner, effectively improving the safety and stability of power grid operation.

[0081] S4. Based on the comparison results between the historical data and the real-time status data of the power grid, the fault monitoring threshold is dynamically adjusted using an adaptive algorithm.

[0082] Furthermore, based on the trend change of real-time data and the statistical characteristics of historical data, an adaptive adjustment factor α is defined to reflect the degree of deviation of real-time status data relative to historical data. The expression is as follows:

[0083]

[0084] Among them, α(t) is the adaptive adjustment factor at time point t, V final (t) is the fault feature vector after fusion at time point t, μ n and σ nare the mean and standard deviation of the fault feature vector under normal conditions in the historical data, t0 and t are the time points for starting and ending the real-time data collection, and dt refers to the integration of the absolute value of the feature vector’s deviation from the normal mean in the time interval [t0, t];

[0085] Furthermore, the adaptive adjustment factor α(t) is used to adjust the fault detection threshold, which is expressed as follows:

[0086] T(t)=μ n ±3σ n +α(t);

[0087] Where T(t) is the dynamic fault monitoring threshold at time point t.

[0088] The increase of α(t) will lead to the increase of T(t), which means that the real-time data shows an abnormal trend, and the system will automatically relax the fault monitoring threshold to avoid false alarms; vice versa, if the data tends to be stable, the fault monitoring threshold will be stricter, thereby improving the sensitivity of fault monitoring;

[0089] It should be noted that the enhanced adaptive capability of the system makes fault monitoring more accurate, avoids the risk of misjudgment due to fixed thresholds, flexibly responds to dynamic changes in the operating status of the power grid, and ensures the effectiveness and timeliness of fault monitoring.

[0090] S5. After detecting an abnormal situation, an alarm is issued, and the fault monitoring mechanism is optimized based on the feedback of the fault handling results.

[0091] Furthermore, if the fault feature vector is detected to exceed the dynamic fault monitoring threshold, the early warning mechanism is triggered, the redundant monitoring program is started, and the extraction and comparison process of the fault feature vector is repeated for secondary verification. If the detection results are consistent, it is confirmed to be a real fault;

[0092] When a fault is confirmed, the log records the time, location, type and fault feature vector of the fault and sends it to the control center to activate the emergency response, redistribute the grid load, reduce the impact of the fault area, and ensure the power supply stability of the non-fault area;

[0093] During the fault handling process, continuously collect data on the impact of the handling actions and the grid status, including the speed of restoring power supply, the number of affected users, and the time required for fault repair. After the handling is completed, compare and analyze with the pre-set monitoring process, evaluate the handling effect, and adjust the monitoring process according to the actual situation;

[0094] It should be noted that the entire fault monitoring system can not only detect and handle abnormal situations in the power grid in a timely manner, but also self-learn and optimize after each fault handling, gradually improve its own fault detection capabilities and response strategies, and ultimately achieve a higher level of automation and intelligence.

[0095] The present embodiment also provides a smart grid fault monitoring system using artificial intelligence, including: a data acquisition and preprocessing module, a feature extraction and fusion module, a fault monitoring threshold setting and comparison module, an adaptive threshold adjustment module and an alarm and feedback optimization module; the data acquisition and preprocessing module is used to obtain power Internet of Things monitoring data through sensors and preprocess it; the feature extraction and fusion module is used to extract fault-related features based on the preprocessed data using a machine learning algorithm and fuse them into a unified fault feature vector; the fault monitoring threshold setting and comparison module is used to determine a preliminary fault monitoring threshold range through statistical analysis based on historical grid data, and compare the fault feature vector with real-time monitoring grid operation status data; the adaptive threshold adjustment module is used to dynamically adjust the fault monitoring threshold using an adaptive algorithm based on the comparison results of the historical data of the grid and the real-time status data; the alarm and feedback optimization module is used to issue an alarm after detecting an abnormal situation, and optimize the fault monitoring mechanism according to the feedback of the fault handling results.

[0096] This embodiment also provides a computer device, which is suitable for the case of a smart grid fault monitoring method using artificial intelligence, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the smart grid fault monitoring method using artificial intelligence proposed in the above embodiment.

[0097] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0098] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing smart grid fault monitoring using artificial intelligence as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0099] In summary, the present invention achieves comprehensive monitoring of the power grid operation status and improvement of data quality, reduces false alarms, and improves monitoring accuracy by: deploying multiple sensors and performing data preprocessing; extracting fault features from preprocessed data using a machine learning algorithm and fusing them into a unified fault feature vector; achieving a high degree of generalization and integration of fault features, enhancing the accuracy and classification capabilities of fault identification, and improving the sensitivity of fault detection; determining a preliminary fault monitoring threshold through statistical analysis of historical data and comparing it with real-time data; achieving real-time monitoring of the power grid operation status and rapid response to faults, reducing the impact of faults; dynamically adjusting the fault monitoring threshold through an adaptive algorithm; achieving self-optimization and enhanced adaptability of the monitoring system; and improving system stability and adaptability to complex environments.

[0100] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a smart grid fault monitoring method using artificial intelligence are provided.

[0101] In order to verify the effectiveness and superiority of the proposed smart grid fault monitoring method using artificial intelligence, the traditional artificial intelligence algorithm was used for fault detection as a control group in the experiment, and the method of the present invention was used as an experimental group for comparison.

[0102] First, temperature sensors, current sensors, and voltage sensors were deployed at key nodes in a regional power grid to monitor the power grid status in real time. These sensors were connected to a central data processing center, which was responsible for receiving and processing data from the sensors.

[0103] After receiving the raw data, the central data processing center immediately cleans and removes the noise. For the current and voltage signals, a low-pass filter is used for smoothing to reduce noise interference. The pre-processed data is used for subsequent fault feature extraction.

[0104] Then a method combining principal component analysis (PCA) and convolutional neural network (CNN) is used to extract fault features and fuse them into a unified fault feature vector.

[0105] Then, by statistically analyzing the characteristic vectors of normal and fault states in the historical data of the power grid in the past year, the mean and standard deviation of the fault characteristic vector under normal operating conditions were calculated, and the preliminary fault monitoring threshold range was set according to the principle of three times the standard deviation.

[0106] Finally, the grid operation status data obtained from real-time monitoring is converted into a fault feature vector and compared with the set threshold range. In addition, an adaptive adjustment factor is defined based on the trend changes of real-time data and the statistical characteristics of historical data to dynamically adjust the fault monitoring threshold.

[0107] The details are shown in Table 1 below:

[0108] Table 1 Performance comparison of smart grid fault monitoring methods

[0109]

[0110] As can be seen from the above table, the method of the present invention has smaller mean and standard deviation of fault feature vectors than the traditional method, which means that the fault features extracted by the method of the present invention are more concentrated, reducing the possibility of false alarms. In terms of fault monitoring threshold, the method of the present invention is also lower, which shows that under the same fault conditions, the method of the present invention can detect potential faults earlier, thereby improving the safety of the system.

[0111] More importantly, in terms of the key indicator of fault detection accuracy, the method of the present invention reached 93%, while the traditional method was only 85%. This shows that by introducing an adaptive algorithm to dynamically adjust the fault monitoring threshold and combining a redundant monitoring program for secondary verification, the method of the present invention can significantly improve the accuracy and reliability of fault detection.

[0112] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A smart grid fault monitoring method using artificial intelligence, characterized in that: include, Obtain power IoT monitoring data through sensors and pre-process it; Based on the preprocessed data, machine learning algorithms are used to extract fault-related features and fuse them into a unified fault feature vector; Based on the historical data of the power grid, the preliminary fault monitoring threshold range is determined through statistical analysis; And compare the fault feature vector with the real-time monitoring grid operation status data; Based on the comparison results of historical data and real-time status data of the power grid, the fault monitoring threshold is dynamically adjusted using an adaptive algorithm; An alarm is issued after an abnormal situation is detected, and the fault monitoring mechanism is optimized based on the feedback of the fault handling results.

2. The method for monitoring faults in a smart grid using artificial intelligence as claimed in claim 1, characterized in that: The sensors include a temperature sensor, a current sensor, and a voltage sensor; The preprocessing includes data cleaning and data denoising; The real-time status data includes grid load and temperature changes.

3. The method for monitoring faults in a smart grid using artificial intelligence as claimed in claim 2, characterized in that: The specific steps of obtaining the power IoT monitoring data through sensors and preprocessing it are as follows: Deploy temperature sensors, current sensors, and voltage sensors at power grid substations and transmission line nodes to monitor the power grid in real time and send data to the central data processing center in real time; The central data processing center receives the raw data, cleans it, denoises it, removes invalid data, and uses a low-pass filter to smooth the current and voltage signals and remove noise.

4. The method for monitoring faults in a smart power grid using artificial intelligence as claimed in claim 3, characterized in that: Based on the preprocessed data, the machine learning algorithm is used to extract fault-related features and fuse them into a unified fault feature vector. The specific steps are as follows: A method combining principal component analysis and convolutional neural network is used to extract fault-related features from the preprocessed data. The expression is as follows: Among them, F is the extracted fault feature vector, D is the preprocessed data set, PCA (D i ) is through D i The main components obtained by principal component analysis, D i is the data of the i-th dimension in the dataset D, i is the index variable, CNN(D i ) is obtained from D through convolutional neural network i The features extracted from Var(D i ) is D i The variance of Feature fusion is performed based on the mutual information between features. The expression is as follows: Among them, V final is the fused fault feature vector, F i ,F j are the feature vectors extracted from two different dimensions, MIF i ,F j F i ,F j The mutual information between them.

5. The method for monitoring faults in a smart grid using artificial intelligence as claimed in claim 4, characterized in that: Based on the historical data of the power grid, the preliminary fault monitoring threshold range is determined through statistical analysis, and the fault feature vector is compared with the real-time monitoring power grid operation status data. The specific steps are as follows: Summarize the feature vectors of normal state and fault state in the historical data of the power grid. The distribution of feature vectors in normal operation state follows normal distribution, while the distribution of feature vectors in fault state will be significantly different. Calculate the mean and standard deviation of the fault feature vector under normal conditions, and set the initial fault monitoring threshold range based on the principle of three times the standard deviation; The grid operation status data obtained by real-time monitoring is converted into a fault feature vector. If the fault feature vector obtained by real-time monitoring data is within the fault threshold range, it is considered that the grid operation status is good, otherwise it is considered that there is a fault in the grid operation.

6. The method for monitoring faults in a smart power grid using artificial intelligence as claimed in claim 5, characterized in that: The comparison result based on the historical data of the power grid and the real-time status data is used to dynamically adjust the fault monitoring threshold using an adaptive algorithm. The specific steps are as follows: Based on the trend change of real-time data and the statistical characteristics of historical data, an adaptive adjustment factor α is defined to reflect the degree of deviation of real-time status data relative to historical data. The expression is as follows: Among them, α(t) is the adaptive adjustment factor at time point t, V final (t) is the fault feature vector after fusion at time point t, μ n and σ n are the mean and standard deviation of the fault feature vector under normal conditions in the historical data, t0 and t are the time points for starting and ending the real-time data collection, and dt refers to the integration of the absolute value of the feature vector’s deviation from the normal mean in the time interval [t0, t]; The adaptive adjustment factor α(t) is used to adjust the fault detection threshold. The expression is as follows: T(t)=μ n ±3σ n +α(t); Where T(t) is the dynamic fault monitoring threshold at time point t.

7. The method for monitoring faults in a smart power grid using artificial intelligence according to claim 6, characterized in that: After the abnormal situation is detected, an alarm is issued, and the fault monitoring mechanism is optimized according to the feedback of the fault handling result. The specific steps are as follows: When the fault feature vector is detected to exceed the dynamic fault monitoring threshold, the early warning mechanism is triggered, the redundant monitoring program is started, and the extraction and comparison process of the fault feature vector is repeated for secondary verification; When a fault is confirmed, the log records the time, location, type and fault feature vector of the fault and sends it to the control center to activate the emergency response and redistribute the grid load; During the fault handling process, the impact data of the handling actions and power grid status are continuously collected and compared with the pre-set monitoring process, the handling effect is evaluated, and the monitoring process is adjusted according to the actual situation.

8. A smart grid fault monitoring system using artificial intelligence, based on the smart grid fault monitoring method using artificial intelligence according to any one of claims 1 to 7, characterized in that: Including data acquisition and preprocessing module, feature extraction and fusion module, fault monitoring threshold setting and comparison module, adaptive threshold adjustment module and alarm and feedback optimization module; The data acquisition and preprocessing module is used to obtain power IoT monitoring data through sensors and preprocess it; The feature extraction and fusion module is used to extract fault-related features based on the preprocessed data using a machine learning algorithm and fuse them into a unified fault feature vector; The fault monitoring threshold setting and comparison module is used to determine the preliminary fault monitoring threshold range through statistical analysis based on the historical data of the power grid, and compare it with the real-time monitoring power grid operation status data according to the fault feature vector; The adaptive threshold adjustment module is used to dynamically adjust the fault monitoring threshold using an adaptive algorithm based on the comparison result between the historical data and the real-time status data of the power grid; The alarm and feedback optimization module is used to issue an alarm after detecting an abnormal situation and optimize the fault monitoring mechanism according to the feedback of the fault handling result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the smart grid fault monitoring method using artificial intelligence described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the smart grid fault monitoring method using artificial intelligence described in any one of claims 1 to 7 are implemented.

Images