Power grid mountain fire early warning method and system based on data monitoring

Abstract

The embodiment of the invention discloses a power grid mountain fire early warning method and system based on data monitoring, and the method comprises the steps: determining the position of a monitoring point and the sampling precision according to the structure and geographical conditions of a power grid; collecting monitoring data based on the monitoring point position and the sampling precision; encrypting the monitoring data to obtain encrypted data; transmitting the encrypted data to a data processing center; the data processing center decrypts the encrypted data to obtain decrypted data; the decrypted data are preprocessed, and preprocessed data are obtained; performing feature extraction on the preprocessed data to obtain feature information related to the forest fire; based on the feature information and a DTW dynamic time adjustment algorithm, constructing a fire early warning model; and inputting the decrypted data into the fire early warning model, and outputting a model prediction result. According to the method, the accuracy and efficiency of fire early warning can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of forest fire prevention of power grid transmission lines, and in particular to a power grid forest fire early warning method and system based on data monitoring. Background Art

[0002] In modern society, power grid fires have become a serious safety hazard, causing huge losses to people's lives and property. Therefore, early warning of power grid fires is particularly important.

[0003] At present, the traditional power grid fire early warning system is mainly used. The traditional power grid fire early warning system mainly relies on equipment such as smoke detectors. Its sensitivity and accuracy are limited by environmental factors and equipment performance, and it is difficult to meet the needs of power grid safety management. Therefore, there is an urgent need for a more intelligent and efficient power grid fire early warning system to improve the accuracy and timeliness of fire early warning. Summary of the invention

[0004] Based on this, it is necessary to propose a power grid wildfire early warning method and system based on data monitoring to address the above problems.

[0005] First aspect

[0006] A power grid wildfire early warning method based on data monitoring, the method comprising:

[0007] Determine the location of monitoring points and sampling accuracy based on the grid structure and geographical conditions;

[0008] Based on the monitoring point locations and sampling accuracy, collecting monitoring data;

[0009] Encrypting the monitoring data to obtain encrypted data;

[0010] Transmitting the encrypted data to a data processing center;

[0011] The data processing center decrypts the encrypted data to obtain decrypted data;

[0012] Preprocessing the decrypted data to obtain preprocessed data;

[0013] Performing feature extraction on the preprocessed data to obtain feature information related to wildfires;

[0014] Based on the characteristic information and the DTW dynamic time adjustment algorithm, a fire warning model is constructed;

[0015] The decrypted data is input into the fire warning model, and the model prediction result is output.

[0016] Preferably, the monitoring data includes at least one of current and voltage waveform data, temperature, humidity, wind speed and wind direction.

[0017] Preferably, the preprocessing of the decrypted data to obtain the preprocessed data specifically includes:

[0018] Perform outlier detection on the decrypted data to obtain data after excluding outliers;

[0019] Smoothing the data after excluding outliers to obtain denoised data;

[0020] The denoised data is normalized to obtain normalized data.

[0021] Preferably, the feature extraction of the preprocessed data to obtain feature information related to wildfires specifically includes:

[0022] The preprocessed data is subjected to time domain feature extraction and frequency domain feature extraction to obtain feature information related to wildfires.

[0023] Preferably, the fire warning model is constructed based on the feature information and the DTW dynamic time adjustment algorithm, specifically including:

[0024] Use feature information to construct the sampling sequence cost matrix;

[0025] Finding an optimal curved path in the sequence cost matrix composed of two sequences;

[0026] The Euclidean distances on the optimal curved path are summed up so that the cumulative distance between the two sequences is minimized.

[0027] Preferably, the method further comprises:

[0028] Combine real-time monitoring data and model prediction results to determine early warning information;

[0029] The warning information is sent and a warning is issued.

[0030] Preferably, the combining of real-time monitoring data and model prediction results to determine early warning information specifically includes:

[0031] Combine real-time monitoring data and model prediction results to determine whether the current fire is in a critical state and set the warning threshold;

[0032] Under the current fire critical state, the fire risk is assessed to obtain an assessment level;

[0033] Combine real-time monitoring data with model prediction results to determine the direction of fire spread;

[0034] The assessment level and fire spreading direction are used as early warning information.

[0035] Preferably, encrypting the monitoring data to obtain encrypted data specifically includes:

[0036] E(D)=AES(D,K)

[0037] Wherein, E(D) represents the result after encrypting data D, AES represents the AES encryption algorithm, and K represents the encryption key.

[0038] Preferably, the step of smoothing the data after excluding outliers to obtain denoised data specifically includes:

[0039]

[0040] Where D(t) represents the data after smoothing at time t, D(i) represents the original data collected at time i, and K represents the window size.

[0041] Second aspect

[0042] A power grid wildfire early warning system based on data monitoring, the system comprising:

[0043] Data acquisition module: used to determine the location of monitoring points and sampling accuracy according to the grid structure and geographical conditions; based on the location of monitoring points and sampling accuracy, collect monitoring data;

[0044] Data transmission module: used for encrypting the monitoring data to obtain encrypted data; transmitting the encrypted data to a data processing center;

[0045] Data processing module: used for the data processing center to decrypt the encrypted data to obtain decrypted data; pre-process the decrypted data to obtain pre-processed data;

[0046] Model building module: used to extract features from the preprocessed data to obtain feature information related to wildfires; and to build a fire warning model based on the feature information and the DTW dynamic time adjustment algorithm;

[0047] Fire warning module: used to input the decrypted data into the fire warning model and output the model prediction results.

[0048] The embodiments of the present invention have the following beneficial effects:

[0049] The present invention aims to utilize real-time monitoring and analysis of waveform data in the power grid, combined with the DTW dynamic time adjustment algorithm, to accurately identify signs of fire, and propose a new power grid fire warning system and method to overcome the limitations of traditional warning systems, improve the accuracy and efficiency of fire warnings, and thus better ensure the safety of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0051] in:

[0052] Figure 1 A flowchart of a power grid wildfire early warning method based on data monitoring in one embodiment;

[0053] Figure 2 A schematic diagram of a sampling sequence cost matrix in one embodiment;

[0054] Figure 3 The structure block diagram of a power grid wildfire early warning system based on data monitoring in one embodiment. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] The technical problem to be solved by the present invention is to provide a data monitoring power grid fire prevention system and method based on the DTW dynamic time adjustment algorithm, which realizes early monitoring and early warning of wildfires by deploying monitoring equipment on the power grid and collecting and analyzing waveform data in the power grid in real time.

[0057] Waveform data monitoring technology is a new type of power grid monitoring method. The current and voltage waveforms in the power grid reflect the operating status of the power grid. When a fault or abnormality occurs in the power grid, the waveform data will change accordingly. Therefore, by real-time monitoring and analysis of waveform data in the power grid, signs of power grid fire can be discovered in a timely manner.

[0058] Traditional waveform monitoring technology is mainly used for power grid fault diagnosis and operation status monitoring, but is less used in fire warning. In recent years, the combination of waveform monitoring technology and algorithms has become possible. The DTW dynamic time adjustment algorithm can match features from a large amount of waveform data and identify patterns and rules related to fire, thereby improving the accuracy and reliability of fire warning.

[0059] The present invention relates to a method and system for early warning of forest fires in power grids based on data monitoring, and more particularly to a system and method for early detection and early warning of forest fires using data monitoring technology. The system is composed of monitoring equipment, a data transmission system and a data processing center, a DTW dynamic time adjustment algorithm operation module, and an early warning system. By laying monitoring equipment on the transmission line, the waveform data of the power grid is monitored and analyzed in real time, and the waveform data is transmitted to the data processing center through a safe and reliable communication method. The data processing center establishes a stable data storage system, and uses noise removal, filtering, and data normalization methods to pre-process the waveform for subsequent algorithm analysis and processing, extracting features related to forest fires from the processed data, and establishing a prediction model based on the feature data and evaluating it. The system and method are based on data monitoring technology, can accurately identify features related to forest fires, and reduce the false alarm rate. The system and method of the present invention can be widely used in the field of safety monitoring and maintenance of transmission lines.

[0060] Monitoring system: The monitoring system is the basis of this system. Its main function is to collect current and voltage waveform data in the power grid in real time in order to monitor changes in the power grid status.

[0061] Data transmission and data processing system: This system is used to transmit and process data. The data transmission system collects data information in the monitoring system and encodes and compresses it, using encryption technology to ensure the security and accuracy of the data, and then transmits it to the data processing system through the network and other communication means. The data processing system receives data from the data transmission system, performs denoising and correction processing on the received data, and extracts feature information.

[0062] DTW dynamic time adjustment algorithm analysis module: This system is used to effectively learn and analyze data, thereby establishing a fire warning model and predicting the power grid status. The selected model is trained using an algorithm based on the characteristics of the data, and the model is established.

[0063] Early warning system: The early warning system determines whether to trigger an early warning based on pre-set fire warning conditions. Detailed early warning information is generated based on the results of monitoring data analysis, including fire risk level, expected spread direction, etc., and then the generated early warning information is converted into light, digital and other signals and sent to relevant personnel.

[0064] Figure 1FIG. 1 is a flow chart of a method for early warning of a forest fire in a power grid based on data monitoring in one embodiment. Figure 1 As shown, the method specifically comprises the following steps:

[0065] Step 101, determine the location of monitoring points and sampling accuracy according to the grid structure and geographical conditions.

[0066] Sensor deployment: Various sensors are deployed in the power grid, such as temperature sensors, humidity sensors, wind speed sensors, etc., to monitor changes in environmental parameters.

[0067] Sampling accuracy setting: For different types of monitoring parameters (such as temperature, humidity, wind speed, etc.), their change rates and fluctuation characteristics are different, and the appropriate sampling accuracy needs to be determined based on the specific parameter characteristics.

[0068] The data storage system has key technologies such as large-capacity storage, high-performance storage, data backup, data management, access control, real-time monitoring and scalability. It provides sufficient storage capacity to store monitoring data for a long time and supports high-speed reading, writing and access to large amounts of data. The stored data is backed up and managed regularly to ensure the security and reliability of the data. At the same time, the system has flexible access control and real-time monitoring functions to protect the security and privacy of data, and promptly discover and handle storage device failures and anomalies. Through real-time transmission, the data transmission system can transmit data to the data processing system in a timely manner to ensure the timeliness and accuracy of the data. Efficient and secure transmission protocols and encryption technologies are used to ensure the security and integrity of data during data transmission. At the same time, the system has flexible transmission configuration and settings, and can flexibly adjust the transmission frequency and data volume according to actual monitoring needs to improve the flexibility and adaptability of the system.

[0069] Data collection is a crucial step in the power grid wildfire early warning system, which directly affects the subsequent data processing and the establishment of early warning models. Data collection is carried out within the power grid by deploying a sensor network, including temperature sensors, humidity sensors, wind speed sensors, etc., which are distributed in various locations covered by the power grid to monitor environmental parameter data in real time. The collected data covers information on air temperature, relative humidity, wind speed and direction, etc., providing basic data support for subsequent fire early warnings. The data acquisition system has the characteristics of high precision and high sensitivity, and can accurately capture changes in environmental parameters, providing a reliable data foundation for the fire early warning system.

[0070] Step 102, collecting monitoring data based on the location of the monitoring point and the sampling accuracy.

[0071] Collect data. Determine the location of monitoring points based on the grid structure and geographical conditions, covering key nodes of the grid and potential fire risk areas. Determine the appropriate sampling accuracy based on the monitoring characteristics. The higher the sampling accuracy, the finer the data, but the complexity of data processing also increases, so the sampling accuracy needs to be set reasonably. The sensor equipment collects current and voltage waveform data in real time, converts the analog signal into a digital signal through an analog-to-digital converter (AD), and finally performs a data check to check whether the data has problems such as outliers and missing values. If so, perform data correction to correct the measurement error.

[0072] The monitoring data includes at least one of current and voltage waveform data, temperature, humidity, wind speed and wind direction.

[0073] Step 103, encrypt the monitoring data to obtain encrypted data.

[0074] According to monitoring needs, taking into account factors such as topography, vegetation, and population density, select appropriate deployment points to arrange temperature sensors, humidity sensors, and wind speed sensors. Set appropriate data collection frequencies for different types of monitoring parameters. For rapidly changing parameters, a higher sampling frequency is required. Configure a data transmission system to achieve real-time data transmission. Use encryption and compression technology in the data transmission system to ensure the security of data transmission. The encryption algorithm is:

[0075] Encrypt the monitoring data to obtain encrypted data, including:

[0076] E(D)=AES(D,K)

[0077] Wherein, E(D) represents the result after encrypting data D, AES represents the AES encryption algorithm, and K represents the encryption key.

[0078] Step 104, transmitting the encrypted data to a data processing center.

[0079] The raw data is processed to improve the quality and availability of the data. First, the collected raw data is denoised to filter out the noise signals caused by sensor errors or environmental interference and retain valid data information. Then data preprocessing is performed, including data cleaning, completion, calibration and other operations to ensure the integrity and accuracy of the data. After data processing, feature information related to wildfires is extracted from the data for subsequent model training and early warning analysis. Multiple environmental parameters such as temperature, humidity, and wind speed are selected as features to reflect environmental changes. In the feature extraction process, time series analysis and spectrum analysis are used to extract distinguishing features from the raw data. After feature extraction, the DTW dynamic time adjustment algorithm is used to establish a fire warning model. By compressing and stretching the sequence, one point in one sequence corresponds to multiple points in another sequence. The problem of asynchrony between the two sequences is solved by flexible matching, and the overall similarity of the two sequences is measured, which improves the generalization ability and prediction accuracy of the model. Once the early warning system detects signs of fire and meets the early warning conditions, detailed early warning information is generated. Combined with factors such as the direction of spread, wind direction, and terrain, the fire spread path and speed are predicted, and detailed early warning information including fire risk level, spread direction, and estimated spread time is generated. The early warning information is promptly communicated to relevant personnel so that countermeasures can be taken.

[0080] Step 105: The data processing center decrypts the encrypted data to obtain decrypted data.

[0081] Step 106, preprocessing the decrypted data to obtain preprocessed data.

[0082] The decrypted data is preprocessed to obtain preprocessed data, specifically including:

[0083] Perform outlier detection on the decrypted data to obtain data after excluding outliers;

[0084] Smoothing the data after excluding outliers to obtain denoised data;

[0085] Data quality control uses statistical methods to detect outliers in the collected data, exclude outliers, and smooth the data to eliminate noise and jitter in the data. The smoothing algorithm is:

[0086]

[0087] Where D(t) represents the data after smoothing at time t, D(i) represents the original data collected at time i, and K represents the window size.

[0088] The denoised data is normalized to obtain normalized data.

[0089] Data normalization maps the data to the same scale to facilitate model training. The formula is:

[0090]

[0091] Wherein, min(D) and max(D) represent the maximum and minimum values ​​of data D respectively.

[0092] Step 107 , extracting features from the preprocessed data to obtain feature information related to wildfires.

[0093] Feature extraction is performed on the preprocessed data to obtain feature information related to wildfires, including:

[0094] The preprocessed data is subjected to time domain feature extraction and frequency domain feature extraction to obtain characteristic information related to wildfires.

[0095] Feature extraction includes time domain feature extraction and frequency domain feature extraction, including statistical features such as variance, mean, and Fourier transform:

[0096]

[0097] x(t) represents time domain data, X(f) represents frequency domain data, f represents frequency, and N represents quantity.

[0098] Transmit and preprocess the data. Use a dedicated data transmission module (transmitted by mobile data) to convert the collected data into an input signal and transmit it to the data processing center. The preprocessing system uses interpolation and average filling methods to detect over-range and error values ​​on the received data. After eliminating bad values, use filters to denoise the data to improve data accuracy. Finally, scale the data to a uniform scale to extract wildfire characteristics from the frequency and time domains of the signal.

[0099] Step 108: construct a fire warning model based on the feature information and the DTW dynamic time adjustment algorithm.

[0100] Based on feature information and DTW dynamic time adjustment algorithm, a fire warning model is constructed, which includes:

[0101] Use feature information to construct the sampling sequence cost matrix;

[0102] Find the optimal curved path in the sequence cost matrix composed of two sequences;

[0103] The Euclidean distances on the optimal curved path are summed to minimize the cumulative distance between the two sequences.

[0104] The basic principle of the DTW dynamic time adjustment analysis module is to compress and stretch the sequence so that one point in one sequence corresponds to multiple points in another sequence. The problem of asynchrony between the two sequences is solved by flexible matching, and the overall similarity of the two sequences is measured. The specific method is as follows:

[0105] Construct the sampling sequence cost matrix: Figure 2 FIG. 1 is a schematic diagram of a sampling sequence cost matrix in an embodiment. Figure 2 As shown: The elements in D represent the Euclidean distance between the corresponding elements of the two sequences. Solving the dynamic time warping distance between two sequences is to find an optimal warping path P in the matrix D composed of the two sequences: {p1, p2, pi, ps}, where s is the optimal path length, max(m, n) < s < m + n, and sum the Euclidean distances on the optimal path to minimize the cumulative distance between the two sequences, that is

[0106]

[0107] Since the data information has a high similarity with the training database information when a wildfire occurs, the DTW distance is small; when no wildfire occurs, the similarity is low and the DTW distance is large. Therefore, a wildfire identification criterion can be constructed based on DTW.

[0108] When a wildfire occurs, the similarity with the information in the model database is small, and the distance map DTW value is small; conversely, the distance map DTW value is large.

[0109] The DTW dynamic time adjustment algorithm analysis module is used to establish a fire warning model. Its basic principle is to calculate the distance between two sequences, align the two time series according to the optimal path, find a "curved" path so that the sum of the distances between all points on this path is minimized, and identify wildfire characteristics through the measurement of flexible matching.

[0110] The early warning system determines the fire warning conditions and generates detailed warning information. Combined with the data analysis and prediction models in the wildfire monitoring system, the generated warning information includes the fire risk level, expected spread direction, and possible impact range. The generated warning information is then transmitted to relevant personnel.

[0111] Step 109, input the decrypted data into the fire warning model and output the model prediction result.

[0112] The method also includes:

[0113] Combine real-time monitoring data and model prediction results to determine early warning information;

[0114] Send warning information and issue warnings.

[0115] Generate detailed warning information based on factors such as warning conditions and spread direction. Analyze real-time monitored environmental data, including temperature, humidity, wind speed, wind direction, etc. Based on the classification of monitoring data, classify fire risks into different levels such as low risk, medium risk, and high risk, and predict the direction of fire spread based on topography and vegetation distribution. Finally, generate and transmit warning information to relevant personnel.

[0116] Determine the warning conditions. By analyzing historical data, determine the typical environmental conditions and changing trends for fires. Based on the model prediction results, judge whether the current environmental conditions are tending towards the critical state for fire occurrence. Finally, based on the output results of the prediction model, set certain warning conditions, such as temperature exceeding a certain threshold, humidity below a certain threshold, wind speed and direction consistent with historical data, etc., as conditions for triggering the warning.

[0117] Assess the risk level and establish a risk assessment formula based on quantified risk factors, including the weighted sum of various factors or a mathematical model to assess the risk level of fire. According to the assessment results, the fire risk is divided into different levels, such as low risk, medium risk, high risk, etc.

[0118] Prediction of fire spread direction: Based on historical fire data and environmental conditions, the laws and characteristics of fire spread are summarized, and the fire spread direction under current conditions is predicted based on the model. Combined with real-time monitoring data and model prediction results, the fire spread direction is determined, and early warning information is issued to relevant departments in a timely manner.

[0119] Combine real-time monitoring data and model prediction results to determine early warning information, including:

[0120] Combine real-time monitoring data and model prediction results to determine whether the current fire is in a critical state and set the warning threshold;

[0121] Under the current fire critical state, the fire risk is assessed to obtain an assessment level;

[0122] Combine real-time monitoring data with model prediction results to determine the direction of fire spread;

[0123] The assessment level and fire spread direction are used as early warning information.

[0124] The present invention, by utilizing the DTW dynamic time adjustment algorithm and geographic information system technology, combined with real-time monitored environmental data and historical data, can more accurately predict the possibility and spread direction of wildfires, improve the accuracy of wildfire warnings, help improve the safety of the power grid, and prevent the occurrence of power grid failures and accidents. By predicting the fire risk level and spread direction, it is possible to effectively optimize resource allocation, rationally arrange personnel and materials, improve resource utilization efficiency, and improve power grid operation efficiency. By generating detailed warning information, including fire risk level, expected spread direction, etc., and promptly delivering it to relevant personnel and institutions, it is possible to strengthen emergency response capabilities, improve the efficiency and timeliness of wildfire disaster response, help improve the reliability and stability of the power grid, and ensure the continuity and stability of power supply.

[0125] Figure 3 FIG. 1 is a structural block diagram of a power grid wildfire early warning system based on data monitoring in one embodiment. Figure 3 As shown, the system includes:

[0126] Data collection module 3100: used to determine the location of monitoring points and sampling accuracy according to the grid structure and geographical conditions; based on the location of monitoring points and sampling accuracy, collect monitoring data;

[0127] Data transmission module 3200: used to encrypt monitoring data to obtain encrypted data; and transmit the encrypted data to a data processing center;

[0128] Data processing module 3300: used for the data processing center to decrypt the encrypted data to obtain decrypted data; pre-process the decrypted data to obtain pre-processed data;

[0129] Model building module 3400: used to extract features from the preprocessed data to obtain feature information related to wildfires; and to build a fire warning model based on the feature information and the DTW dynamic time adjustment algorithm;

[0130] Fire warning module 3500: used to input the decrypted data into the fire warning model and output the model prediction results.

[0131] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0132] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0133] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A power grid wildfire early warning method based on data monitoring, characterized in that: The method comprises: Determine the location of monitoring points and sampling accuracy based on the grid structure and geographical conditions; Based on the monitoring point locations and sampling accuracy, collecting monitoring data; Encrypting the monitoring data to obtain encrypted data; Transmitting the encrypted data to a data processing center; The data processing center decrypts the encrypted data to obtain decrypted data; Preprocessing the decrypted data to obtain preprocessed data; Performing feature extraction on the preprocessed data to obtain feature information related to wildfires; Based on the characteristic information and the DTW dynamic time adjustment algorithm, a fire warning model is constructed; The decrypted data is input into the fire warning model, and the model prediction result is output.

2. The method for early warning of power grid wildfires based on data monitoring according to claim 1, characterized in that: The monitoring data includes at least one of current and voltage waveform data, temperature, humidity, wind speed and wind direction.

3. The method for early warning of power grid wildfires based on data monitoring according to claim 1, characterized in that: The preprocessing of the decrypted data to obtain the preprocessed data specifically includes: Perform outlier detection on the decrypted data to obtain data after excluding outliers; Smoothing the data after excluding outliers to obtain denoised data; The denoised data is normalized to obtain normalized data.

4. The method for early warning of power grid wildfires based on data monitoring according to claim 1, characterized in that: The feature extraction of the pre-processed data to obtain feature information related to wildfires specifically includes: The preprocessed data is subjected to time domain feature extraction and frequency domain feature extraction to obtain feature information related to wildfires.

5. The method for early warning of power grid wildfires based on data monitoring according to claim 1, characterized in that: The fire warning model is constructed based on the feature information and the DTW dynamic time adjustment algorithm, specifically including: Use feature information to construct the sampling sequence cost matrix; Finding an optimal curved path in the sequence cost matrix composed of two sequences; The Euclidean distances on the optimal curved path are summed up so that the cumulative distance between the two sequences is minimized.

6. The method for early warning of power grid wildfires based on data monitoring according to claim 1, characterized in that: The method further comprises: Combine real-time monitoring data and model prediction results to determine early warning information; The warning information is sent and a warning is issued.

7. The method for early warning of power grid wildfires based on data monitoring according to claim 6, characterized in that: The combination of real-time monitoring data and model prediction results to determine early warning information specifically includes: Combine real-time monitoring data and model prediction results to determine whether the current fire is in a critical state and set the warning threshold; Under the current fire critical state, the fire risk is assessed to obtain an assessment level; Combine real-time monitoring data with model prediction results to determine the direction of fire spread; The assessment level and fire spreading direction are used as early warning information.

8. The method for early warning of power grid wildfires based on data monitoring according to claim 1, characterized in that: The encrypting the monitoring data to obtain encrypted data specifically includes: E(D)=AES(D,K) Wherein, E(D) represents the result after encrypting data D, AES represents the AES encryption algorithm, and K represents the encryption key.

9. The method for early warning of power grid wildfires based on data monitoring according to claim 3 is characterized in that: The smoothing process of the data after excluding outliers to obtain denoised data specifically includes: Where D(t) represents the data after smoothing at time t, D(i) represents the original data collected at time i, and K represents the window size.

10. A power grid wildfire early warning system based on data monitoring, characterized in that: The system comprises: Data acquisition module: used to determine the location of monitoring points and sampling accuracy according to the grid structure and geographical conditions; based on the location of monitoring points and sampling accuracy, collect monitoring data; Data transmission module: used for encrypting the monitoring data to obtain encrypted data; transmitting the encrypted data to a data processing center; Data processing module: used for the data processing center to decrypt the encrypted data to obtain decrypted data; pre-process the decrypted data to obtain pre-processed data; Model building module: used to extract features from the preprocessed data to obtain feature information related to wildfires; and to build a fire warning model based on the feature information and the DTW dynamic time adjustment algorithm; Fire warning module: used to input the decrypted data into the fire warning model and output the model prediction results.

Images