Power distribution network mountain fire disaster state monitoring and early warning system

Abstract

The invention is suitable for the technical field of mountain fire disaster monitoring and early warning, and provides a power distribution network mountain fire disaster state monitoring and early warning system, which comprises a multi-source data processing module, a space-time coupling monitoring module, an intelligent response module and a remote management platform, the multi-source data processing module is used for integrating and processing satellite remote sensing data, meteorological data, topographic data and power distribution network topology data; the space-time coupling monitoring module is connected with the multi-source data processing module and is used for dynamically analyzing the space-time characteristics of the fire behavior; the intelligent response module is connected with the space-time coupling monitoring module and is used for executing hierarchical prevention and control operation according to a fire behavior space-time characteristic analysis result; the remote management platform is connected with the space-time coupling monitoring module and the intelligent response module and used for outputting a forest fire disaster early warning scheme according to the fire behavior space-time characteristic analysis result and the graded prevention and control operation process, the emergency evacuation efficiency can be improved, and power grid facility damage and economic loss caused by fire behavior diffusion are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of mountain fire disaster monitoring and early warning, and particularly relates to a mountain fire disaster status monitoring and early warning system for a distribution network. Background Art

[0002] Currently, in many places, due to the lack of multi-dimensional and multi-scale spatio-temporal collaborative analysis of mountain fire disasters, various problems have occurred. For example, although satellite remote sensing can detect heat sources over a large area, affected by factors such as cloud cover, resolution limitations, and long revisit periods, it is difficult to timely capture small-scale hotspots in the initial stage of a fire, and it is easily interfered by industrial heat sources and surface high temperatures, resulting in frequent false alarms. The ground sensor network can achieve local real-time monitoring, but the deployment cost is high in the mountainous distribution network, and there are many coverage blind spots due to terrain occlusion, making it difficult to achieve full-domain connection monitoring. In addition, most existing early warning models use static thresholds or simple linear regression, and do not fully consider the spatio-temporal heterogeneity of fire spread, resulting in relatively large prediction deviations. Especially in some special areas such as karst landforms, it cannot meet the minute-level response requirements of the distribution network, leading to low emergency evacuation efficiency and even serious economic losses.

[0003] In view of this, a mountain fire disaster status monitoring and early warning system for a distribution network is needed to solve the above problems. Summary of the Invention

[0004] An embodiment of the present application provides a mountain fire disaster status monitoring and early warning system for a distribution network, which is used to solve the problems of low emergency evacuation efficiency and even serious economic losses.

[0005] An embodiment of the present application provides a mountain fire disaster status monitoring and early warning system for a distribution network, including a multi-source data processing module, a spatio-temporal coupling monitoring module, an intelligent response module, and a remote management platform. The multi-source data processing module is used to integrate and process satellite remote sensing data, meteorological data, terrain data, and distribution network topology data; the spatio-temporal coupling monitoring module is connected to the multi-source data processing module and is used to dynamically analyze the spatio-temporal characteristics of the fire; the intelligent response module is connected to the spatio-temporal coupling monitoring module and is used to perform hierarchical prevention and control operations according to the analysis results of the spatio-temporal characteristics of the fire; the remote management platform is respectively connected to the spatio-temporal coupling monitoring module and the intelligent response module and is used to output a mountain fire disaster early warning plan according to the analysis results of the spatio-temporal characteristics of the fire and the process of hierarchical prevention and control operations.

[0006] Furthermore, the multi-source data processing module is used to integrate and process satellite remote sensing data, meteorological data, terrain data, and distribution network topology data, including:

[0007] Performing format unification and standardization processing on the obtained satellite remote sensing data, meteorological data, terrain data, and distribution network topology data;

[0008] Extract feature vectors based on the processed satellite remote sensing data, meteorological data, terrain data, and distribution network topology data;

[0009] Determine a standardized data set according to the extracted feature vectors, where the standardized data set includes a satellite brightness temperature sequence, meteorological grid data, terrain slope map, and power grid facility distribution map that are spatio-temporally aligned.

[0010] Furthermore, the determination of the standardized data set according to the extracted feature vectors, where the standardized data set includes a brightness temperature sequence, meteorological grid data, terrain slope map, and power grid facility distribution map that are spatio-temporally aligned, includes:

[0011] Expression of the satellite brightness temperature sequence:

[0012]

[0013] Where: T 3.9 is the brightness temperature in the mid-infrared band, T 11 is the brightness temperature in the far-infrared band, C 1 and C 2 are the first radiation constant and the second radiation constant, λ is the center wavelength of the band, λ 5 is the wavelength of band 5, L 3.9 is the radiance in the mid-infrared band, L 11 is the radiance in the far-infrared band;

[0014] Expression of the meteorological grid data:

[0015] FWI = I T (T) + I r (r) + I v (v) + I u (u)

[0016] Where: FWI is the fire weather index, I T (T) is the temperature index, based on the daily maximum temperature, I r (r) is the precipitation index, based on the daily precipitation, I v (v) is the wind speed index, based on the daily maximum wind speed, I u (u) is the humidity index, based on the daily minimum relative humidity;

[0017] Expression of the terrain slope map:

[0018]

[0019] Where: S is the slope, indicating the steepness of the surface, f x is the elevation change rate in the x direction, f y is the elevation change rate in the y direction;

[0020] Expression of the power grid facility distribution map:

[0021]

[0022] Where: D fire is the straight-line distance between the fire site and the power grid facilities, and x fire , y fire are the longitude and latitude coordinates of the fire site, and y fire , y grid are the longitude and latitude coordinates of the power grid facilities.

[0023] Furthermore, the spatio-temporal coupling monitoring module is connected to the multi-source data processing module for dynamically analyzing the spatio-temporal characteristics of the fire situation, including:

[0024] Performing data preprocessing on the standardized data set, and determining whether there are fire points in each region by combining the time series method and the spatial threshold method;

[0025] When there are fire points, predicting the fire arrival time, the fire spread path, the fire spread speed in the area with fire points, and evaluating the threat level of the fire situation in the area with fire points to the power grid facilities.

[0026] Furthermore, the performing data preprocessing on the standardized data set and determining whether there are fire points in each region by combining the time series method and the spatial threshold method includes:

[0027] Calculating the brightness temperature change rate of each region in the time series based on the satellite brightness temperature sequence, and the calculation formula:

[0028]

[0029] Where: ΔΛ is the brightness temperature change rate, reflecting the dynamic fire situation, and are the mid-infrared brightness temperatures at t+1 and t moments respectively, and Δu is the time interval;

[0030] Constructing an ideal curve of the brightness temperature changing with time under clear sky conditions, setting a dynamic threshold. When the brightness temperature change rate is greater than the first threshold, there are fire points in the current region. When the brightness temperature change rate is less than the second threshold, the natural cooling interference in the current region is excluded.

[0031] Furthermore, the predicting the fire arrival time, the fire spread path, the fire spread speed in the area with fire points, and evaluating the threat level of the fire situation in the area with fire points to the power grid facilities when there are fire points includes:

[0032] Predicting the fire arrival time in the area with fire points based on the LSTM model;

[0033] Construct a spatial coupling model based on the preprocessed meteorological grid data and the topographic slope map, and predict the fire spread path and the speed of fire spread through the spatial coupling model;

[0034] Calculate the real-time distance between the fire area and the transmission line according to the power grid facility distribution map, and evaluate the threat level of the fire situation to the power grid facilities based on the calculated real-time distance.

[0035] Furthermore, the intelligent response module is connected to the spatio-temporal coupling monitoring module, and is used to perform hierarchical prevention and control operations according to the analysis results of the spatio-temporal characteristics of the fire situation, including:

[0036] When the fire arrival time reaches the set threshold, trigger hierarchical prevention and control instructions, including automatically disconnecting the high-voltage side circuit breaker, starting the tower base flame retardant spraying system, and performing an emergency load transfer based on the distribution terminal unit;

[0037] Use the DBSCAN clustering algorithm to divide the core fire area, the key prevention and control area, and the residential evacuation area, and perform prevention and control operations on the divided areas, including using drones to extinguish fires in the core area, dynamically adjusting the power in the prevention and control area, and giving early warning broadcasts and traffic control in the evacuation area.

[0038] Furthermore, when the fire arrival time reaches the set threshold, trigger hierarchical prevention and control instructions, including automatically disconnecting the high-voltage side circuit breaker, starting the tower base flame retardant spraying system, and performing an emergency load transfer based on the distribution terminal unit, including:

[0039] When the fire arrival time is greater than the first threshold, the current fire situation is in the incubation period, and the instruction for the incubation period is to monitor the fire point density and issue a yellow warning;

[0040] When the fire arrival time is less than the first threshold and greater than the second threshold, the current fire situation is in the initial combustion period, and the instruction for the initial combustion period is to start the tower base flame retardant spraying system, reduce the line load rate, and issue an orange warning;

[0041] When the fire arrival time is less than the second threshold, the current fire situation is in the spreading period, and the instruction for the spreading period is to issue a red warning and execute hierarchical prevention and control instructions.

[0042] Furthermore, the calculation formula for area power adjustment is as follows:

[0043]

[0044] Where: P new is the adjusted power, P old is the original line load power, D fire is the real-time distance between the fire field and the line, D safe is the safety radius, and k is the adjustment coefficient;

[0045] The broadcast content of the evacuation area warning broadcast includes the fire risk level, evacuation routes, and traffic control information. The traffic control includes the scope and strategy of the control.

[0046] Furthermore, the remote management platform is respectively connected to the spatio-temporal coupling monitoring module and the intelligent response module, and is used to output a wildfire disaster warning plan according to the analysis results of the spatio-temporal characteristics of the fire and the hierarchical prevention and control operation process, including:

[0047] Generate a fire risk heat map, a diffusion animation, and a power grid threat map according to the fire risk level and diffusion path of the spatio-temporal coupling monitoring module and the prevention and control operation records of the intelligent response module;

[0048] Divide the warning level according to the spatio-temporal characteristics of the fire, and generate a detailed warning plan including prevention and control instructions, evacuation plans, and traffic control;

[0049] Push warning information through text messages, APP notifications, and community broadcasts, and give priority to pushing to the person in charge of threatened equipment, residents in the evacuation area, and the traffic management department;

[0050] Dynamically update the fire field boundary and adjust the prevention and control strategy based on drone monitoring and sensor data.

[0051] From the above technical solutions, it can be seen that the embodiments of this application have the following advantages:

[0052] The monitoring and warning system for the wildfire disaster state of the distribution network integrates satellite, meteorological, terrain, and power grid data through a multi-source data processing module. The spatio-temporal coupling monitoring module dynamically analyzes the spatio-temporal characteristics of the fire, the intelligent response module executes hierarchical prevention and control operations, and the remote management platform generates and outputs a warning plan, fully considering the spatio-temporal heterogeneity of the fire spread. The system can accurately predict the fire spread path and speed, divide the core fire area, key prevention and control area, and resident evacuation area, and significantly improve the emergency evacuation efficiency through hierarchical prevention and control measures such as breaker disconnection, sprinkler activation, and load transfer, reducing the damage to power grid facilities and economic losses caused by the spread of the fire. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a structural block diagram of a monitoring and warning system for the wildfire disaster state of a distribution network in the present invention;

[0054] Figure 2 It is a schematic flowchart of an embodiment of the multi-source data processing module in the present invention;

[0055] Figure 3 It is a schematic flowchart of an embodiment of the spatio-temporal coupling monitoring module in the present invention;

[0056] Figure 4 It is a high-temperature change rate curve at different solar altitude angles in the present invention;

[0057] Figure 5 This is a schematic flowchart of the implementation process of the remote management platform in the present invention. Detailed implementation manners

[0058] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0059] In this embodiment, the distribution network mountain fire disaster status monitoring and early warning system is used to improve the emergency evacuation efficiency and early warning comprehensiveness of mountain fire disasters through the spatio-temporal heterogeneity of the spread of fire. The implementation method in this embodiment can be implemented in the system, can be implemented on the server, or can be implemented on the terminal, and no specific limitation is made.

[0060] Embodiment 1

[0061] Please refer to Figure 1 , a distribution network mountain fire disaster status monitoring and early warning system in the present invention includes a multi-source data processing module 1, a spatio-temporal coupling monitoring module 2, an intelligent response module 3 and a remote management platform 4. The multi-source data processing module 1 is used to integrate and process satellite data, meteorological data, terrain data and distribution network topology data; the spatio-temporal coupling monitoring module 2 is connected to the multi-source data processing module 1 and is used to dynamically analyze the spatio-temporal characteristics of the fire; the intelligent response module 3 is connected to the spatio-temporal coupling monitoring module 2 and is used to perform hierarchical prevention and control operations according to the analysis results of the spatio-temporal characteristics of the fire; the remote management platform 4 is respectively connected to the spatio-temporal coupling monitoring module 2 and the intelligent response module 3 and is used to output a mountain fire disaster early warning plan according to the analysis results of the spatio-temporal characteristics of the fire and the process of hierarchical prevention and control operations. The present invention integrates and standardizes the processing of satellite data, meteorological data, terrain data and distribution network topology data, provides a precise and comprehensive data basis for subsequent analysis by extracting corresponding feature vectors; dynamically analyzes the spatio-temporal evolution characteristics of the fire, can predict the fire spread path and threat level; performs hierarchical prevention and control operations according to the spatio-temporal characteristics of the fire, and dynamically optimizes the power grid operation; finally, integrates data, visualizes and generates an early warning plan through the platform.

[0062] Embodiment 2

[0063] Please refer to Figure 2 , the multi-source data processing module is used to integrate and process satellite data, meteorological data, terrain data and distribution network topology data, and includes the following steps:

[0064] S21. Perform format unification and standardization processing on the obtained satellite remote sensing data, meteorological data, terrain data and distribution network topology data;

[0065] S22. Extract feature vectors based on the processed satellite remote sensing data, meteorological data, terrain data, and distribution network topology data;

[0066] S23. Determine a standardized data set according to the extracted feature vectors. The standardized data set includes a spatiotemporally aligned brightness temperature sequence, meteorological grid data, a terrain slope map, and a power grid facility distribution map.

[0067] First, perform format unification and standardization processing on the acquired satellite remote sensing data, such as mid-infrared and far-infrared brightness temperatures; meteorological data, such as wind speed and humidity; terrain data, such as slope and aspect; and distribution network topology data, such as tower coordinates and line load rates, to ensure the consistency of the data spatiotemporal reference. Extract key feature vectors from the processed data, including the brightness temperature difference, fire risk meteorological index, terrain shielding coefficient, and power grid vulnerability score. Generate a standardized data set based on the extracted feature vectors, specifically including a spatiotemporally aligned brightness temperature sequence, meteorological grid data, a terrain slope map, and a power grid facility distribution map, providing a data basis for the spatiotemporal feature analysis and hierarchical prevention and control of fire situations.

[0068] Expression of the brightness temperature sequence:

[0069]

[0070] Where: T 3.9 is the brightness temperature in the mid-infrared band, T 11 is the brightness temperature in the far-infrared band, C 1 and C 2 are the first radiation constant and the second radiation constant, λ is the central wavelength of the band, λ 5 is the wavelength of band 5, L 3.9 is the radiance in the mid-infrared band, L 11 is the radiance in the far-infrared band.

[0071] Expression of the meteorological grid data:

[0072] FWI = I T (T) + I r (r) + I v (v) + I u (u)

[0073] Where: FWI is the fire risk meteorological index, I T (T) is the temperature index, based on the daily maximum temperature, I r (r) is the precipitation index, based on the daily precipitation, I v (v) is the wind speed index, based on the daily maximum wind speed, I u (u) is the humidity index, based on the daily minimum relative humidity.

[0074] Expression of the terrain slope map:

[0075]

[0076] Among them: S is the slope, indicating the steepness of the ground surface, and f x is the elevation change rate in the x direction, and f y is the elevation change rate in the y direction.

[0077] Expression of the power grid facility distribution map:

[0078]

[0079] Among them: D fire is the straight-line distance between the fire site and the power grid facilities, and x fire and y fire are the longitude and latitude coordinates of the fire site, and y fire and y grid are the longitude and latitude coordinates of the power grid facilities.

[0080] Embodiment III

[0081] Please refer to Figure 3 , the spatio-temporal coupling monitoring module is connected to the multi-source data processing module for dynamically analyzing the spatio-temporal characteristics of the fire situation, including the following steps:

[0082] S31. Perform data preprocessing on the standardized data set, and determine whether there are fire points in each area by combining the time series method and the spatial threshold method;

[0083] Determine the brightness temperature time series of the mid-infrared and far-infrared bands of the satellite according to the satellite brightness temperature sequence, with a time resolution of 10 minutes. Perform radiometric calibration, atmospheric correction, and geometric correction on the satellite remote sensing data to eliminate cloud cover and terrain distortion errors.

[0084] 1. Specifically, in the case of analyzing remote sensing data, calculate the brightness temperature change rate of each area (pixel) in the time series based on the satellite brightness temperature sequence. The calculation formula of the brightness temperature change rate is as follows:

[0085]

[0086] Among them: ΔΛ is the brightness temperature change rate, reflecting the dynamic fire situation, and are the mid-infrared brightness temperatures at times t + 1 and t respectively, and Δu is the time interval.

[0087] By analyzing the comparison between the change rate and the conventional background change rate, obtain whether there is other released heat source information in the identification area. The current geostationary meteorological satellite can obtain an observation once every 10 minutes, and the change of the surface temperature in this time period is small, generally less than 0.5K. Without a fire situation, the surface heating energy mainly comes from the sun. According to the heat conduction equation formula:

[0088]

[0089] In the formula, ν is the thermodynamic temperature, t is the time, κ is the thermal conductivity, which is related to the thermal conductivity, density, specific heat or heat capacity of the material. The change in the surface temperature is driven by the periodic solar radiation, which provides a periodic heat flux. The heating effect of solar radiation on the surface can be expressed by the following formula:

[0090]

[0091] In the formula, ν s is the effective sky long-wave radiation temperature, I is the incident solar radiation reaching the ground through the atmosphere, κ is the thermal conductivity, and x is the penetration distance into the surface. The first and third terms are the radiation fluxes incident from the sky and the sun respectively, and the second term is the emitted radiation flux output from the surface. The solar radiation I is a function of the surface reflectivity, solar declination, latitude and local slope in the solar spectral region (mostly visible and near-infrared regions), as shown in the following formula:

[0092] I(t) = (1 - A) * S 0 * C * H(t)

[0093] In the formula, A is the surface reflectivity, S 0 is the solar constant, and C is the factor of the attenuation of solar radiation by clouds.

[0094]

[0095] In the formula, Z′(t) is the local zenith angle of the inclined surface, Z(t) is the zenith angle, and M is the atmospheric attenuation, which is a function of the zenith angle Z. The time is related to the solar angle. For the clear-sky atmosphere during the day, the change in the surface temperature in the ideal state is mainly caused by solar irradiation. The formula for the effective solar radiation absorbed by the surface can be converted to the formula:

[0096] I(φ) = (1 - A) * S 0 * δ * ε * sin(φ)

[0097] In the formula, ε is the atmospheric transmittance, δ is the surface absorptance, and φ is the solar altitude angle.

[0098] For a fixed area at a specific time, assuming that A, ε and δ are constants and the underlying surface type remains unchanged, the energy absorbed by the surface is only related to the solar altitude angle φ. As the solar altitude angle increases, the absorbed energy also gradually increases, that is, the energy absorbed by the surface is proportional to the trigonometric function of the solar altitude angle. By comparing the temperature change rates at different times before and after with the background change rate in the non-fire state, it is determined whether there is a fire in the warming (or cooling) process of the identified area.

[0099] 2. Construct an ideal curve of the brightness temperature varying with time under clear sky conditions, set a dynamic threshold. When the change rate of the brightness temperature is greater than the first threshold, there is a fire point in the current area. When the change rate of the brightness temperature is less than the second threshold, the interference of natural cooling in the current area is excluded.

[0100] To determine whether the change in the brightness temperature of background pixels at any moment under clear sky conditions is an abnormal situation, it is necessary to obtain the regular temporal variation of the background brightness temperature and construct a temporal variation function of the background brightness temperature. The construction method of the background brightness temperature is to use the time series change of the background brightness temperature under clear atmospheric conditions to establish a normalized temporal variation function of the brightness temperature. The time scale generally requires continuous minute-level observations with a daily cycle of 24 hours. When the change in the brightness temperature exceeds the normal amplitude, it is considered that there are clouds or abnormal thermal pixels, and at this time, the background brightness temperature needs to be reconstructed. For a fixed area, the solar radiation received by the ground surface is related to the solar altitude angle. Combining the analysis of the mid-infrared brightness temperature change data of the ground surface under clear sky, the change in the ground surface brightness temperature and the change in the solar angle can be simplified into three time periods: (1) the daytime warming period, (2) the daytime cooling period, and (3) the nighttime cooling period. An ideal temperature change curve is obtained. Combining the basic principle of ground surface warming, that is, the radiation energy is positively correlated with the solar altitude angle, the relationship between the ground surface background brightness temperature and the solar altitude angle can be described by the formula:

[0101]

[0102] In the formula, T is the instantaneous brightness temperature; T max is the daily maximum brightness temperature; T min is the daily minimum brightness temperature; T 1 is the brightness temperature when the solar altitude angle drops to 0°.

[0103] In different seasons, there are differences in the absolute value of the brightness temperature change within a single day. To reduce the differences caused by different times and different regions, the brightness temperature difference value is converted into a brightness temperature change rate. Using the above formula for the relationship between the ground surface background brightness temperature and the solar altitude angle, the brightness temperature change rate of pixels at any time period can be estimated. According to the characteristic that the temperature change difference within a day does not reach 60K for different latitudes and different ground surface types, it is assumed that the condition of the maximum temperature difference of 60K for different regions and different seasons is satisfied. An ideal brightness temperature change rate curve for the daytime is established as Figure 4 shown. The daytime is equally divided into six intervals by the altitude angle, and the maximum assumed threshold for each stage is selected:

[0104] Assumption conditions: T max = 333 (60 °C), T min = 273 (0 °C), temperature difference 60K;

[0105] Warming stage: [0,30]: 0.35%; (30,60]: 0.3%; ; (60,90]: 0.2%; ;

[0106] Cooling stage: [90,60]: 0; (60,30]: 0.1%; (30,0]: 0.2%;

[0107] Taking a temperature difference of 60K as an example, the maximum change rate in the heating stage is less than 0.4%, and the maximum change rate in the cooling stage is less than 0.3%. During the time-series method identification process, in the heating stage, if the actual heating rate exceeds N times the maximum change rate, it can be considered that there is high-temperature anomaly information in the pixel (the heating speed is higher than the normal heating); in the cooling stage, if M times the actual cooling is lower than the maximum change rate, it can be considered that there is high-temperature anomaly information in the pixel (the cooling speed is slower than the normal cooling, and even shows a reverse increase). Here, N times and M times are the multiple thresholds for the heating and cooling processes under the condition of a temperature difference of 60K, respectively. This threshold multiple is used to determine whether there is a heating source at the identification moment.

[0108] S32. When there is a fire point, predict the time when the fire reaches, the fire spread path, the fire spread speed in the area where the fire point exists, and evaluate the threat level of the fire situation in the area where the fire point exists to the power grid facilities;

[0109] Specifically, S32 includes the following steps:

[0110] S321. Predict the time when the fire reaches in the area where the fire point exists based on the LSTM model;

[0111] The input for the LSTM model to predict the time when the fire reaches here includes the brightness temperature time series, meteorological data, and terrain data, and predicts the time when the fire reaches key facilities (such as poles and substations).

[0112] S322. Construct a spatial coupling model based on the preprocessed meteorological grid data and terrain slope map, and predict the fire spread path and fire spread speed through the spatial coupling model;

[0113] Based on the heat source point coordinates, meteorological grid data, and terrain slope map identified by the time-series method in the above steps, obtain the fire spread direction and spread speed. Among them, the spread direction is synthesized according to the slope aspect and wind direction according to the set weights, and the spread speed is determined according to the spread direction.

[0114] S323. Calculate the real-time distance between the fire area and the transmission line according to the power grid facility distribution map, and evaluate the threat level of the fire situation to the power grid facilities based on the calculated real-time distance.

[0115] From the expression of the above power grid facility distribution map, the real-time distance between the fire area and the transmission line is D fireBy constructing a power grid vulnerability scoring model to determine the threat level of a fire to power grid facilities, where the model parameters include real-time distance, load rate, and terrain occlusion, and a corresponding scoring model is obtained by setting weights for each parameter. Finally, the corresponding threat level is obtained according to the result of the scoring model. Here, the level is set from 1 to 5 levels, and a corresponding threshold is set for each level. When the calculated score reaches the set threshold, the corresponding threat level can be determined.

[0116] Embodiment 4

[0117] In the present invention, the intelligent response module is connected to the spatio-temporal coupling monitoring module and is used to perform hierarchical prevention and control operations according to the analysis results of the spatio-temporal characteristics of the fire, including the following:

[0118] S42. When the fire arrival time reaches the set threshold, a hierarchical prevention and control instruction is triggered, including automatically disconnecting the high-voltage side circuit breaker, starting the tower base flame retardant spraying system, and performing an emergency load transfer based on the distribution terminal unit;

[0119] 1. When the fire arrival time is greater than the first threshold, the current fire is in the incubation period, and the instruction for the incubation period is to monitor the fire point density and issue a yellow warning;

[0120] 2. When the fire arrival time is less than the first threshold and greater than the second threshold, the current fire is in the initial combustion period, and the instruction for the initial combustion period is to start the tower base flame retardant spraying system, reduce the line load rate, and issue an orange warning;

[0121] 3. When the fire arrival time is less than the second threshold, the current fire is in the spreading period, and the instruction for the spreading period is to issue a red warning and execute the hierarchical prevention and control instruction.

[0122] Specifically, the first threshold is set to 120 minutes here, and the second threshold is set to 30 minutes. When the predicted fire arrival time is greater than 120 minutes, the current fire is in the incubation period, and the instruction for the incubation period is to monitor the fire point density and issue a yellow warning; when the fire arrival time is less than 120 minutes and greater than 30 minutes, the current fire is in the initial combustion period, and the instruction for the initial combustion period is to start the tower base flame retardant spraying system, reduce the line load rate, and issue an orange warning; when the fire arrival time is less than 30 minutes, the current fire is in the spreading period, and the instruction for the spreading period is to issue a red warning and execute the hierarchical prevention and control instruction.

[0123] The hierarchical prevention and control instructions include the automatic disconnection of the high-voltage side circuit breaker, the activation of the tower base flame-retardant spraying system, and the emergency load transfer based on the distribution terminal unit. Among them, when the fire arrival time is less than 30 minutes and the fire site distance is less than 2 km, the high-voltage side circuit breakers of all distribution transformers within 2 km around the fire site are disconnected, and the status of the circuit breakers is monitored in real time through the SCADA system. When the fire arrival time is less than 60 minutes and the fire site distance is less than 1 km, the flame-retardant spraying systems of all poles and towers within 1 km around the fire site are activated, and the spraying intensity is dynamically adjusted according to the fire spread potential score. When the fire arrival time is less than 30 minutes and the fire site distance is less than 1.5 km, the load is transferred to the standby line through the DTU / FTU, giving priority to ensuring the power supply of hospitals and communication base stations.

[0124] 2. The DBSCAN clustering algorithm is used to divide the core fire area, the key prevention and control area, and the residential evacuation area. Prevention and control operations are carried out for the divided areas, including extinguishing the fire with drones in the core area, dynamically adjusting the power in the prevention and control area, and early warning broadcasts and traffic control in the evacuation area.

[0125] Specifically, the division results of the DBSCAN clustering algorithm are as follows: The core fire area is where the heat source point density ≥ 10 per km 2 , and the radius ≤ 500 m; it is the area with the most active fire, and the bright temperature change rate ΔΛ > 0.4·ΔΛ max . The key prevention and control area is within a 2-km range outside the fire site, with a spread potential score greater than 0.6; it is the area where the fire may spread to power grid facilities. The residential evacuation area is within a 5-km range outside the fire site, with a heat source point density < 5 per km 2 ; residents need to be evacuated in advance to prevent the further spread of the fire. The prevention and control operations include the following:

[0126] In the core area, fire extinguishing agents are dropped by drones to control the core area of the fire. The calculation formula for the amount of fire extinguishing agent dropped is as follows:

[0127] Q fire =A core ·ρ fire ·k extinguish

[0128] Where: Q fire is the amount of fire extinguishing agent dropped, A core is the area of the core fire area, ρ fire is the fire intensity coefficient, k extinguish is the fire extinguishing efficiency coefficient;

[0129] The calculation formula for power adjustment in the prevention and control area is as follows:

[0130]

[0131] Where: P new is the adjusted power, P oldis the original line load power, D fire is the real-time distance between the fire site and the line, D safe is the safety radius, and k is the adjustment coefficient;

[0132] The broadcast content of the evacuation area warning broadcast includes the fire level, evacuation route and traffic control information. The traffic control includes the control scope and strategy.

[0133] Embodiment Five

[0134] Please refer to Figure 5 , in the present invention, the remote management platform is respectively connected to the spatio-temporal coupling monitoring module and the intelligent response module, and is used to output a mountain fire disaster warning plan according to the analysis result of the spatio-temporal characteristics of the fire and the hierarchical prevention and control operation process, including the following:

[0135] S51. Generate a fire risk heat map, a diffusion animation and a power grid threat map according to the fire risk level, diffusion path of the spatio-temporal coupling monitoring module and the prevention and control operation records of the intelligent response module;

[0136] Based on the fire level distribution, the predicted result of the fire spread path and the threat level of the fire to the power grid facilities obtained by the spatio-temporal coupling monitoring module in the above steps, generate a fire risk heat map, a diffusion animation and a power grid threat map, and display them through the GIS platform.

[0137] S52. Divide the warning level according to the spatio-temporal characteristics of the fire, and generate a detailed warning plan including prevention and control instructions, evacuation plans and traffic control;

[0138] Here, the warning levels include level one, level two and level three. The prevention and control instructions include the circuit breaker disconnection list (equipment name, operation time); the area where the sprinkler system is started (coordinates, sprinkler intensity); the load transfer plan (transfer amount, standby line). The evacuation plan includes evacuation routes such as safety exits, assembly points, etc.; the traffic control information includes blocked roads, detour routes, etc.

[0139] S53. Push warning information through text messages, APP notifications and community broadcasts, and give priority to pushing to the person in charge of the threatened equipment, residents in the evacuation area and the traffic management department;

[0140] The text message push covers all the mobile phones of the residents in the evacuation area, and the content is concise and clear; the APP notification pushes the detailed warning plan through the special APP of the power company; the community broadcast broadcasts the warning information in real time through the loudspeaker. Priority ranking: The first-level push is to the person in charge of the threatened equipment and the emergency command center; the second-level push is to the residents in the evacuation area and the traffic management department; the third-level push is to the surrounding communities and media organizations. The push frequency is updated every 10 minutes to ensure the timeliness of the information.

[0141] S54. Dynamically update the fire site boundary and adjust the prevention and control strategy based on the drone monitoring and sensor data.

[0142] Dynamically update the boundary of the core fire area according to the change of fire field temperature and the effect of fire extinguishing agent delivery.

[0143] The above embodiments can accurately predict the spread path and speed of the fire, divide the core fire area, key prevention and control areas, and residential evacuation areas, and significantly improve the emergency evacuation efficiency through hierarchical prevention and control measures such as breaker disconnection, sprinkler activation, and load transfer, reducing the damage to power grid facilities and economic losses caused by the spread of the fire.

[0144] Those of ordinary skill in the art can realize that the units of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components of each example have been generally described according to their functions in the above description. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0145] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc. In addition, the functional units in each embodiment of the present invention can be integrated in one processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0146] If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: USB flash drives, read-only memory (ROM), random access memory (RAM), mobile hard disks, magnetic disks, or optical disks, etc., which can store program codes.

[0147] It can be understood that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present invention, and they should all be covered within the scope of the claims and the description of the present invention.

Claims

1. A distribution network wildfire disaster status monitoring and early warning system, comprising a multi-source data processing module, a spatiotemporal coupling monitoring module, an intelligent response module and a remote management platform, characterized in that: The multi-source data processing module is used to integrate and process satellite remote sensing data, meteorological data, terrain data and distribution network topology data; the spatiotemporal coupling monitoring module is connected to the multi-source data processing module and is used to dynamically analyze the spatiotemporal characteristics of the fire situation; The intelligent response module is connected to the spatiotemporal coupling monitoring module, and is used to perform graded prevention and control operations based on the spatiotemporal characteristic analysis results of the fire situation; the remote management platform is respectively connected to the spatiotemporal coupling monitoring module and the intelligent response module, and is used to output a wildfire disaster warning plan based on the spatiotemporal characteristic analysis results of the fire situation and the graded prevention and control operation process.

2. The distribution network wildfire disaster status monitoring and early warning system according to claim 1 is characterized in that: The multi-source data processing module is used to integrate and process satellite remote sensing data, meteorological data, terrain data and distribution network topology data, including: Unify and standardize the formats of acquired satellite remote sensing data, meteorological data, terrain data and distribution network topology data; Extract feature vectors based on processed satellite remote sensing data, meteorological data, terrain data and distribution network topology data; A standardized data set is determined according to the extracted feature vectors, wherein the standardized data set includes a satellite brightness temperature sequence aligned in time and space, meteorological grid data, a terrain slope map, and a power grid facility distribution map.

3. The distribution network wildfire disaster status monitoring and early warning system according to claim 2 is characterized in that: The standardized data set is determined according to the extracted feature vectors, and the standardized data set includes a brightness temperature sequence aligned in time and space, meteorological grid data, a terrain slope map, and a power grid facility distribution map, including: The expression of satellite brightness temperature series is: Where: T 3.9 is the brightness temperature in the mid-infrared band, T 11 is the brightness temperature of the far-infrared band, C1 and C2 are the first radiation constant and the second radiation constant, λ is the central wavelength of the band, and λ 5 is the wavelength of band 5, L 3.9 is the radiation brightness in the mid-infrared band, L 11 is the radiation brightness in the far infrared band; The expression of meteorological grid data: FWI=I T (T)+I r (r)+I v (v)+I u (u) Among them: FWI is the fire weather index, I T (T) is the temperature index, based on the daily maximum temperature, I r (r) is the precipitation index, based on daily precipitation, I v (v) is the wind speed index, based on the maximum daily wind speed, I u (u) is the humidity index, based on the daily minimum relative humidity; The expression of terrain slope map is: Where: S is the slope, indicating the steepness of the surface, f x is the rate of change of elevation in the x direction, f y is the rate of change of elevation in the y direction; The expression of the power grid facility distribution diagram is: Where: D fire is the straight-line distance between the fire site and the power grid facilities, x fire ,y fire is the latitude and longitude coordinates of the fire scene, y fire ,y grid are the latitude and longitude coordinates of the power grid facility.

4. The distribution network wildfire disaster status monitoring and early warning system according to claim 1 is characterized in that: The spatiotemporal coupling monitoring module is connected to the multi-source data processing module and is used to dynamically analyze the spatiotemporal characteristics of the fire, including: The standardized data set is preprocessed, and the presence of fire points in each area is determined by combining a time series method with a spatial threshold method; When there is a fire point, the fire arrival time, fire spread path, and fire spread speed of the fire area are predicted, and the threat level of the fire in the fire area to the power grid facilities is assessed.

5. The distribution network wildfire disaster status monitoring and early warning system according to claim 4 is characterized in that: The standardized data set is preprocessed, and the time series method and the spatial threshold method are combined to determine whether there are fire points in each area, including: Based on the satellite brightness temperature sequence, the brightness temperature change rate of each region in the time series is calculated using the following formula: Where: ΔΛ is the brightness temperature change rate, reflecting the dynamics of the fire situation, and are the mid-infrared brightness temperatures at time t+1 and t, respectively, and Δu is the time interval; An ideal curve of brightness temperature changing over time under clear sky conditions is constructed, and dynamic thresholds are set. When the brightness temperature change rate is greater than the first threshold, there is a fire point in the current area. When the brightness temperature change rate is less than the second threshold, the natural cooling interference in the current area is eliminated.

6. The distribution network wildfire disaster status monitoring and early warning system according to claim 5 is characterized in that: When a fire point exists, the fire arrival time, fire spread path, and fire spread speed of the area where the fire point exists are predicted, and the threat level of the fire situation in the area where the fire point exists to the power grid facilities is evaluated, including: Predict the arrival time of fire in the area with fire points based on the LSTM model; constructing a spatial coupling model based on the preprocessed meteorological grid data and the terrain slope map, and predicting the fire spread path and the speed of fire spread through the spatial coupling model; The real-time distance between the fire area and the transmission line is calculated according to the distribution map of power grid facilities, and the threat level of the fire to the power grid facilities is evaluated based on the calculated real-time distance.

7. The distribution network wildfire disaster status monitoring and early warning system according to claim 1 is characterized in that: The intelligent response module is connected to the spatiotemporal coupling monitoring module and is used to perform hierarchical prevention and control operations according to the spatiotemporal characteristic analysis results of the fire situation, including: When the fire arrival time reaches the set threshold, the hierarchical prevention and control instructions are triggered, including automatic disconnection of the high-voltage side circuit breaker, activation of the tower base flame retardant sprinkler system, and emergency load transfer based on the distribution terminal unit; The DBSCAN clustering algorithm is used to divide the core fire area, key prevention and control area and residential evacuation area, and prevention and control operations are performed in the divided areas, including using drones to extinguish fires in the core area, dynamically adjusting power in the prevention and control area, and early warning broadcasts and traffic control in the evacuation area.

8. The distribution network wildfire disaster status monitoring and early warning system according to claim 7 is characterized in that: When the fire arrival time reaches the set threshold, the hierarchical prevention and control instructions are triggered, including automatic disconnection of the high-voltage side circuit breaker, activation of the tower base flame retardant sprinkler system, and emergency load transfer based on the distribution terminal unit, including: When the fire arrival time is greater than the first threshold, the current fire is in the latent period, and the instruction of the latent period is to monitor the density of fire points and issue a yellow warning; When the fire arrival time is less than the first threshold and greater than the second threshold, the current fire is in the initial combustion stage. The instruction for the initial combustion stage is to start the tower base flame retardant spray system, reduce the line load rate, and issue an orange warning; When the fire arrival time is less than the second threshold, the current fire is in the spreading stage, and the instruction for the spreading stage is to issue a red warning and execute graded prevention and control instructions.

9. The distribution network wildfire disaster status monitoring and early warning system according to claim 7 is characterized in that: The DBSCAN clustering algorithm is used to divide the core fire area, key prevention and control area and resident evacuation area, and prevention and control operations are performed for the divided areas, including using drones to extinguish fires in the core area, dynamically adjusting power in the prevention and control area, and early warning broadcasting and traffic control in the evacuation area, including: The core area uses drones to drop fire extinguishing agents to control the core area of ​​the fire. The calculation formula for the amount of fire extinguishing agent dropped is as follows: Q fire =A core ·r fire ·k extinguish Where: Q fire is the amount of fire extinguishing agent released, A core is the core fire area, ρ fire is the fire intensity coefficient, k extinguish is the fire extinguishing efficiency coefficient; The calculation formula for power adjustment in the control area is as follows: Where: P new is the adjusted power, P old is the original line load power, D fire is the real-time distance between the fire scene and the line, D safe is the safety radius, k is the adjustment coefficient; The content of the evacuation area early warning broadcast includes the fire level, evacuation route and traffic control information. Traffic control includes the scope and strategy of control.

10. The distribution network wildfire disaster status monitoring and early warning system according to claim 1, characterized in that: The remote management platform is connected to the spatiotemporal coupling monitoring module and the intelligent response module respectively, and is used to output a wildfire disaster warning plan according to the spatiotemporal characteristic analysis results of the fire situation and the hierarchical prevention and control operation process, including: Generate a fire risk heat map, a diffusion animation and a power grid threat map according to the fire risk level and diffusion path of the spatiotemporal coupling monitoring module and the prevention and control operation record of the intelligent response module; Classify the warning level according to the spatiotemporal characteristics of the fire situation, and generate a detailed warning plan including prevention and control instructions, evacuation plans and traffic control; Push warning information via SMS, APP notifications and community broadcasts, with priority given to the person in charge of the threatened equipment, residents in the evacuation area and traffic management departments; Dynamically update fire boundaries and adjust prevention and control strategies based on drone monitoring and sensor data.

Images