Fire investigation and survey system supported by cloud computing

Abstract

The invention relates to the technical field of fire investigation, in particular to a cloud computing supported fire investigation and investigation system, which comprises a data receiving and preliminary analysis module, a dynamic fire mapping module, a fire spreading prediction module and a loss evaluation and reason analysis module. According to the invention, environmental data and GIS are integrated, accurate real-time updating capability is provided for fire exploration, so that a dynamic fire map can reflect the position and spreading condition of a fire source in real time, emergency response is ensured to be executed based on latest and most accurate data, and through deep analysis and simulation of wind speed, terrain and vegetation type data, the real-time updating capability is provided for fire exploration. Different fire spreading scenes can be predicted, high-risk areas can be clearly identified, pertinence of prevention measures is improved, distribution and use of resources are optimized, direct and indirect losses caused by fire disasters are economically analyzed and comprehensively evaluated, fire disaster reasons are clearly determined, and efficiency and effectiveness of fire disaster management are greatly enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of fire investigation and detection, and in particular to a fire investigation and detection system supported by cloud computing. Background Art

[0002] The field of fire investigation technology involves the use of various methods and systems to evaluate and analyze fire scenes to determine the causes, effects and losses of fires. Advanced technologies, including but not limited to remote sensing technology, image analysis, data acquisition and modeling technology, help to quickly and accurately collect fire scene data and achieve accurate reconstruction of the fire scene to support subsequent insurance claims, legal proceedings and safety improvement measures. Fire investigation not only requires hardware equipment to perform on-site operations, but also relies on powerful software support, such as database management and analysis tools, to process and interpret the large amount of data collected.

[0003] Among them, the fire investigation and survey system refers to the system that performs fire data processing and storage. The main purpose of the system is to improve the efficiency and reliability of fire data processing, realize remote storage, sharing and analysis of data through cloud technology, and can collect various data at the fire scene in real time, such as temperature, smoke distribution and structural damage, and then upload the data to the cloud. The cloud server processes the data to help investigators remotely analyze the cause of the fire, assess the loss and develop a recovery plan. In addition, the system also supports multi-party collaboration, allowing insurance companies, fire departments and other relevant parties to access and use fire data in real time.

[0004] Existing fire investigation technologies have significant delays and deficiencies in processing and analyzing key data, especially in terms of the real-time and accuracy of fire source location and spread prediction. Traditional methods are often unable to quickly integrate and process sensor data distributed in different locations, resulting in untimely updates of fire information and failure to provide decision makers with real-time fire dynamics. In addition, existing technologies lack effective tools for economic impact analysis and behavioral pattern recognition, making it difficult to quickly extract key information from large data sets, limiting the depth and breadth of fire consequence analysis. This lack of processing power, especially in large-scale fire incidents, can lead to misallocation and improper dispatch of rescue resources, increasing the risk of casualties and property losses. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a fire investigation and survey system supported by cloud computing.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: A fire investigation and survey system supported by cloud computing includes:

[0007] The data receiving and preliminary analysis module collects data from temperature sensors and smoke detectors, performs time synchronization and calibrates wind direction data, pre-processes camera and satellite image data, adjusts resolution and contrast, and performs data fusion with calibration data to generate synchronized environmental data;

[0008] The dynamic fire mapping module integrates the environmental data with the GIS data based on the synchronized environmental data, locates the fire source, depicts the fire spread boundary according to the location result, performs geocoding, updates the regional map, and generates a dynamic fire map;

[0009] The fire spread prediction module collects wind speed, terrain and vegetation type data based on the dynamic fire map, performs parameter calculations, simulates the fire spread path, predicts different spread scenarios based on the simulation results, analyzes potential fire risk areas, and generates fire spread prediction results;

[0010] The loss assessment and cause analysis module collects economic activity data for economic impact analysis based on the fire spread prediction results, analyzes the behavior data of personnel and vehicles for behavior pattern analysis, distinguishes normal from abnormal behavior patterns, and generates fire cause and loss investigation records.

[0011] As a further solution of the present invention, the calibration step of the wind direction data is:

[0012] Collect data from temperature sensors and smoke detectors, set timestamps, apply formulas,

[0013] t sensor =t real -δt

[0014] Calculate the data acquisition time t sensor , generating sensor data with timestamps, where t real represents the actual acquisition time, δt represents the transmission delay of data from the sensor to the processing unit;

[0015] The sensor data with the timestamp is time synchronized and the formula is applied,

[0016] t sync =t sensor +Δt sync

[0017] Calculate the time point t after synchronization sync , generate synchronized time data, where Δt sync is the offset from local time to Coordinated Universal Time;

[0018] Using the synchronized time data to adjust the wind direction data, apply the formula,

[0019] Wcal =W raw +k(t sync -t wind ) 2

[0020] Calculate the calibrated wind direction W cal , generate calibrated wind direction data, where W raw is the original wind direction data, t wind is the wind direction data collection time, and k is the calibration coefficient, which is used to adjust the sensitivity of wind direction data calibration.

[0021] As a further solution of the present invention, the step of acquiring the synchronization environment data is:

[0022] Preprocess camera and satellite image data, adjust resolution and contrast, apply formulas,

[0023]

[0024] Calculate the pixel value I after contrast and brightness adjustment prep , generate preprocessed image data, where I raw is the average original pixel value of the original image data, a I 、b I 、c I is the adjustment coefficient, a I Used to adjust contrast, b I To adjust the brightness, c I Used to normalize processing results;

[0025] The preprocessed image data is time-aligned with the calibrated wind direction data, applying the formula,

[0026]

[0027] Calculate the integrated image value I aligned , generate time-aligned image data, where max(W cal ) represents the maximum wind direction after calibration;

[0028] The time-aligned image data is data-fused with the synchronized time data, and the formula is applied:

[0029]

[0030] Calculate the fused environmental data value D fusion , generate synchronous environment data, where I aligned is the time-aligned image data, t sync To synchronize time data, m represents the collection time point, and M represents the total number of collection time points.

[0031] As a further solution of the present invention, the step of locating the fire source is:

[0032] Based on the synchronized environmental data, the temperature T and smoke concentration S are extracted, and the formula is used:

[0033] E score =ω 1 T+ω 2 S

[0034] Calculation environment score E score , generate environmental score data, where ω 1 ,ω 2 are weight coefficients, which are used to adjust the weights of temperature and smoke concentration respectively;

[0035] Using the environmental score data, check data validity, apply formulas,

[0036]

[0037] Calculate the probability of fire source P fire , generate fire source probability data, where k E Represents the sensitivity coefficient, E thresh is the environmental score threshold;

[0038] According to the fire source probability data, a location with a probability higher than a threshold value θ is selected as the fire source location, using the formula:

[0039] L fire = {x|x∈P fire ,x>Y}

[0040] Determine the final fire source location L fire , generate fire source location data, where Y is the probability threshold.

[0041] As a further solution of the present invention, the steps of obtaining the dynamic fire map are:

[0042] According to the fire source location data, combined with wind speed W and terrain slope G, the formula is applied:

[0043] B spread =L fire +αWcos(θ)+βG

[0044] Calculate the fire spread distance value B spread , generate fire spread boundary data, where α represents the wind speed influence coefficient, β represents the terrain slope coefficient, and L fire , W and G represent the fire source location, wind speed and terrain slope, respectively, and θ represents the angle between the wind direction and the fire source location;

[0045] Using the fire spread boundary data, geocoding technology is used to update the regional map in the GIS system, using the formula,

[0046]

[0047] Calculate the updated map evaluation value M updated , get the updated regional map, where M current Indicates the current map evaluation value, d(B spread ,x) represents B spread The distance to point x on the map, λ is the attenuation coefficient;

[0048] Based on the updated area map, a dynamically displayed fire map is created using the formula,

[0049]

[0050] Calculate the cumulative effect value D of time change on fire firemap , generate a dynamic fire map, where T D represents the time parameter, and σ is the time influence coefficient.

[0051] As a further solution of the present invention, the steps of simulating the fire spread path are:

[0052] Collect wind speed W, terrain height H and vegetation type index V data, and use the formula,

[0053]

[0054] Calculate the integration index I env , generate the comprehensive environmental index, where ω W ,ω H and ω V are normalized weight coefficients, corresponding to the weights of wind speed, terrain height, and vegetation type index;

[0055] The fire spread potential is estimated using the environmental comprehensive index using the formula:

[0056]

[0057] Calculate the fire spread potential value P spread , get the fire spread potential data, where k P is the adjustment coefficient used to adjust the impact of the comprehensive environmental index, T P is the baseline threshold and is the sensitive threshold of fire spread potential.

[0058] Based on the fire spread potential data and combined with the dynamic fire map, the fire spread path is simulated using the formula:

[0059] S path =P spread ln(1+D firemap )

[0060] Calculate the fire spread path simulation score S path , generate fire spread path simulation results.

[0061] As a further solution of the present invention, the steps of obtaining the fire spread prediction result are:

[0062] Analyze the simulation results of the fire spread path, predict the spread scenarios under different environmental conditions, and use the formula,

[0063]

[0064] Calculate the difference spread scenario score C diff , generate differential spread scenario analysis data, where W, H are wind speed and terrain height data, α C , β C is the environmental condition sensitivity coefficient;

[0065] Analyze the data according to the different spread scenarios, check the data integrity and analyze the potential fire risk areas, using the formula,

[0066] R risk =∫C diff ·dx

[0067] Calculate the fire area risk score R risk , generate potential fire risk area data, where dx represents the length of the integrated area; using the potential fire risk area data, adopt the formula,

[0068]

[0069] Calculate the total fire spread risk score P final , generate fire spread prediction results.

[0070] As a further solution of the present invention, the steps for obtaining the fire cause and loss investigation record are:

[0071] Based on the fire spread prediction results, economic activity data are collected and integrated from the regional economic statistics department, and the formula is used.

[0072]

[0073] Calculate the economic activity index E activity , generate economic activity data, where A i represents the economic volume of the i-th activity, q iis the corresponding weight, n represents the number of economic activities;

[0074] Using the economic activity data, investigate the impact of fire on the regional economy. Using the formula,

[0075]

[0076] Calculate the estimated economic loss value L economic , generate economic loss data, where δ L represents the maximum potential loss, ξ L is the fire sensitivity parameter, which is used to adjust the impact of economic activity indicators;

[0077] Collect the behavior data of people and vehicles when a fire occurs, conduct behavior analysis, distinguish normal and abnormal patterns, and use formulas to

[0078]

[0079] Calculate the behavior pattern analysis value and generate pattern recognition results, where B data is the average behavior score, θ B and μ B are the response sensitivity parameter and behavioral excursion threshold, respectively;

[0080] Integrate the economic activity data, economic loss data and pattern recognition results to compile and output fire cause and loss investigation records.

[0081] Compared with the prior art, the advantages and positive effects of the present invention are:

[0082] In the present invention, environmental data is integrated with GIS to provide accurate real-time update capabilities for fire surveys, so that dynamic fire maps can reflect the location and spread of fire sources in real time, ensuring that emergency responses are based on the latest and most accurate data. Through in-depth analysis and simulation of wind speed, terrain and vegetation type data, different fire spread scenarios can be predicted, high-risk areas can be clearly identified, the targeted nature of preventive measures can be improved, the allocation and use of resources can be optimized, economic analysis can comprehensively evaluate the direct and indirect losses caused by fire, and the cause of fire can be clarified, greatly enhancing the efficiency and effectiveness of fire management. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 is a system flow chart of the present invention;

[0084] Figure 2 A unified flow chart of the time standard of the present invention;

[0085] Figure 3 A flowchart for obtaining synchronous environmental data of the present invention;

[0086] Figure 4 A flow chart for locating the fire source of the present invention;

[0087] Figure 5 This is a flow chart for obtaining a dynamic fire map of the present invention;

[0088] Figure 6 It is a simulation flow chart of the fire spreading path of the present invention;

[0089] Figure 7 A flowchart for obtaining fire spread prediction results of the present invention;

[0090] Figure 8 The present invention is a flow chart for obtaining fire cause and loss investigation records. DETAILED DESCRIPTION

[0091] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with 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 intended to limit the present invention.

[0092] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0093] See also Figure 1 , the fire investigation and survey system supported by cloud computing includes:

[0094] The data receiving and preliminary analysis module collects data from temperature sensors and smoke detectors, performs time synchronization and calibrates wind direction data, pre-processes camera and satellite image data, adjusts resolution and contrast, and performs data fusion with calibration data to generate synchronized environmental data;

[0095] The dynamic fire mapping module is based on synchronized environmental data, integrates environmental data with GIS data, locates the fire source, depicts the fire spread boundary according to the positioning results, performs geocoding, updates the regional map, and generates a dynamic fire map;

[0096] The fire spread prediction module collects wind speed, terrain and vegetation type data based on dynamic fire maps, performs parameter calculations, simulates fire spread paths, predicts different spread scenarios based on simulation results, analyzes potential fire risk areas, and generates fire spread prediction results;

[0097] The loss assessment and cause analysis module is based on the fire spread prediction results, collects economic activity data for economic impact analysis, analyzes the behavioral data of people and vehicles for behavioral pattern analysis, distinguishes between normal and abnormal behavior patterns, and generates fire cause and loss investigation records.

[0098] Synchronized environmental data includes time-standardized data, image quality adjustment records and comprehensive sensor output results; dynamic fire maps include fire source location coordinates, fire boundary delineation results and map update records; fire spread prediction results include fire spread path simulation diagrams, fire risk assessment reports and scenario difference analysis records; fire cause and loss investigation records include economic impact analysis reports, behavior anomaly analysis results and abnormal behavior classification records.

[0099] See also Figure 2 , the calibration steps of wind direction data are:

[0100] Collect data from temperature sensors and smoke detectors, set timestamps, apply formulas,

[0101] t sensor =t real -δt

[0102] Calculate the data acquisition time t sensor , generating sensor data with timestamps, where t real represents the actual acquisition time, δt represents the transmission delay of data from the sensor to the processing unit;

[0103] To time synchronize the sensor data with timestamps, apply the formula,

[0104] t sync =t sensor +Δt sync

[0105] Calculate the time point t after synchronization sync , generate synchronized time data, where Δt sync is the offset from local time to Coordinated Universal Time;

[0106] Adjust the wind direction data using the synchronized time data, applying the formula,

[0107] W cal =W raw +k(t sync -t wind )2

[0108] Calculate the calibrated wind direction W cal , generate calibrated wind direction data, where W raw is the original wind direction data, t wind is the wind direction data collection time, and k is the calibration coefficient, which is used to adjust the sensitivity of wind direction data calibration.

[0109] Assume the following data exists:

[0110] t real : Actual collection time, for example 12:00:00.

[0111] δt: The transmission delay of data from the sensor to the processing unit, which is set to 2 seconds, indicating the time required from acquisition to the processing unit.

[0112] Substitute into the formula to calculate:

[0113] t sensor =12:00:00-2s=11:59:58

[0114] The result 11:59:58 represents the time point when the data is actually collected from the sensor, taking into account the transmission delay and providing an accurate time tag for data synchronization.

[0115] Continue to use the above obtained t sensor =11:59:58, assuming a time offset of Δt sync is 15 seconds.

[0116] Substituting into the formula, we get:

[0117] t sync =11:59:58+15s=12:00:13

[0118] The result 12:00:13 represents the time point after synchronization to UTC (universal time), which is the benchmark for calibrating the time of all data sets to ensure the consistency and accuracy of data time.

[0119] Assume t wind =12:00:05, W raw =100 degrees, k=0.1.

[0120] Calculation process:

[0121] t sync -t wind =8s

[0122] W cal =100+0.1×(8) 2 =100+6.4=106.4

[0123] The result 106.4 degrees represents the wind direction adjusted according to time synchronization. This value takes into account the offset from the original wind direction data collection to the synchronization time. The square term is used to strengthen the impact of the time difference on the wind direction data, ensuring the accuracy and real-time nature of the wind direction data.

[0124] See also Figure 3 , the steps to obtain the synchronized environment data are:

[0125] Preprocess camera and satellite image data, adjust resolution and contrast, apply formulas,

[0126]

[0127] Calculate the pixel value I after contrast and brightness adjustment prep , generate preprocessed image data, where I raw is the average original pixel value of the original image data, a I 、b I 、c I is the adjustment coefficient, a I Used to adjust contrast, b I To adjust the brightness, c I Used to normalize processing results;

[0128] Temporally align the preprocessed image data with the calibrated wind direction data, applying the formula,

[0129]

[0130] Calculate the integrated image value I aligned , generate time-aligned image data, where max(W cal ) represents the maximum wind direction after calibration;

[0131] The time-aligned image data is fused with the synchronized time data, and the formula is applied.

[0132]

[0133] Calculate the fused environmental data value D fusion , generate synchronous environment data, where I aligned is the time-aligned image data, t sync To synchronize time data, m represents the collection time point, and M represents the total number of collection time points.

[0134] Assume a I = 1.5 to adjust the contrast, b I =10 adjust brightness, c I =1.2 is used to normalize the processing results.

[0135] Assume that the original image data I raw The average pixel value is 120.

[0136] Substituting into the formula, we get:

[0137]

[0138] The result 158.33 represents the pixel value after contrast and brightness adjustment. Through this processing, the recognizability of the fire image is improved, providing a clearer image basis for subsequent data fusion.

[0139] Combined with the above calculation results I prep =158.33 and W cal =106.4.

[0140] Assume max(W cal )=110.

[0141] Substituting into the formula, we get:

[0142]

[0143] The result 153.15 represents the image data after adjustment by wind direction data. This value represents the comprehensive data after the image data and wind direction data are time-aligned, which helps to consider the influence of wind direction factors in environmental monitoring and provide more accurate analysis results.

[0144] t sync :The synchronized time data obtained above is set to 12:00:13, which is converted to a total of 43213 seconds. aligned =153.15, substitute it into the formula and calculate:

[0145] D fusion =153.15×43213=6618070.95

[0146] The result 6618070.95 represents the fused data value, which combines the time-aligned image data and the synchronization time, providing a comprehensive data indicator for the entire monitoring system, which can be used to evaluate the fire environment conditions at a specific moment, such as the correlation between visibility and time in the monitoring area.

[0147] See also Figure 4 , the steps to locate the fire source are:

[0148] Based on the synchronous environmental data, the temperature T and smoke concentration S are extracted, and the formula is used:

[0149] E score =ω 1 T+ω2 S

[0150] Calculation environment score E score , generate environmental score data, where ω 1 ,ω 2 are weight coefficients, which are used to adjust the weights of temperature and smoke concentration respectively;

[0151] Using environmental score data, checking data validity, applying formulas,

[0152]

[0153] Calculate the probability of fire source P fire , generate fire source probability data, where k E Represents the sensitivity coefficient, E thresh is the environmental score threshold;

[0154] According to the fire source probability data, the location with probability higher than the threshold θ is selected as the fire source location, using the formula,

[0155] L fire = {x|x∈P fire ,x>Y}

[0156] Determine the final fire source location L fire , generate fire source location data, where Y is the probability threshold.

[0157] Assume that the actual data is: T = 35 degrees Celsius, S = 200ppm.

[0158] The weight coefficients are: ω 1 =0.7,ω 2 =0.3.

[0159] Calculate the environmental score:

[0160] E score =0.7×35+0.3×200

[0161] E score =24.5+60

[0162] E score =84.5

[0163] The resulting environmental score is 84.5, which takes into account the effects of temperature and smoke on the potential location of the fire source. A higher score indicates a location that is more likely to be the fire source.

[0164] Assume k E =1.8, E thresh =50, continue to use the above calculation result E score =84.5.

[0165] Calculate the probability of fire source:

[0166]

[0167] A fire source probability close to 1 indicates that under the current environmental score, the location is very likely to be a fire source.

[0168] Assume that the probability threshold Y=0.75.

[0169] judge:

[0170] L fire ={x|x>0.75}

[0171] Because P fire ≈1>0.75, so this location is confirmed as the fire source.

[0172] The final fire source location is based on the high fire source probability, which strengthens the important decision support for rapid fire response and resource allocation.

[0173] See also Figure 5 , the steps to obtain the dynamic fire map are:

[0174] According to the fire source location data, combined with wind speed W and terrain slope G, the formula is applied:

[0175] B spread =L fire +αWcos(θ)+βG

[0176] Calculate the fire spread distance value B spread , generate fire spread boundary data, where α represents the wind speed influence coefficient, β represents the terrain slope coefficient, and L fire , W and G represent the fire source location, wind speed and terrain slope, respectively, and θ represents the angle between the wind direction and the fire source location;

[0177] Using the fire spread boundary data, geocoding technology is used to update the regional map in the GIS system, using formulas,

[0178]

[0179] Calculate the updated map evaluation value M updated , get the updated regional map, where M current Indicates the current map evaluation value, d(B spread ,x) represents B spread The distance to point x on the map, λ is the attenuation coefficient;

[0180] Based on the updated area map, a dynamic fire map is created using the formula,

[0181]

[0182] Calculate the cumulative effect value D of time change on fire firemap , generate a dynamic fire map, where T D represents the time parameter, and σ is the time influence coefficient.

[0183] Assume the following data exists:

[0184] L fire : is position A, that is, positionA.

[0185] W: wind speed, assumed to be 20km / h.

[0186] θ: The angle between wind direction and fire source, assumed to be 45°.

[0187] G: The slope of the terrain, assumed to be 10°.

[0188] α: Wind speed influence coefficient, set to 0.5.

[0189] β: Terrain influence coefficient, set to 0.3.

[0190] Calculate the fire spread boundary:

[0191] B spread =positionA+0.5×20×cos(45°)+0.3×10

[0192] B spread =positionA+0.5×20×0.707+3

[0193] B spread =positionA+7.07+3

[0194] B spread =positionA+10.07

[0195] The spread boundary increases by 10.07 units from the fire source location. The results show the potential spread direction and distance of the fire under the influence of wind speed and terrain, which helps emergency services to arrange fire lines.

[0196] Assume M current =100, d(B spread ,x)=100 (spread 100 units in all directions), λ=0.1, then:

[0197]

[0198] The updated map value is still 100, indicating that under the set attenuation rate and distance conditions, the fire spread has a uniform overall impact on the map, without changing the data value of the original map, which helps to understand the spatial distribution of the fire-affected area.

[0199] Assume T D is the number of hours from the start of the fire, for example 5 hours, i.e. T D =5,σ=0.05.

[0200] Substituting into the formula, we get:

[0201]

[0202] D firemap =0+0.25=0.25

[0203] The dynamic fire map has a value of 0.25, which represents the cumulative effect of fire impacts over time, providing a real-time dynamic view to help emergency managers monitor the spread and speed of fires.

[0204] See also Figure 6 , the simulation steps of the fire spread path are:

[0205] Collect wind speed W, terrain height H and vegetation type index V data, and use the formula,

[0206]

[0207] Calculate the integration index I env , generate the comprehensive environmental index, where ω W ,ω H and ω V are normalized weight coefficients, corresponding to the weights of wind speed, terrain height, and vegetation type index;

[0208] The fire spread potential is estimated using the comprehensive environmental index, using the formula:

[0209]

[0210] Calculate the fire spread potential value P spread , get the fire spread potential data, where k P is the adjustment coefficient used to adjust the impact of the comprehensive environmental index, T P is the baseline threshold and is the sensitive threshold of fire spread potential.

[0211] Based on the fire spread potential data and combined with the dynamic fire map, the fire spread path is simulated and the formula is used.

[0212] S path =P spread ln(1+Dfiremap )

[0213] Calculate the fire spread path simulation score S path , generate fire spread path simulation results.

[0214] W: wind speed, assumed to be 10m / s.

[0215] H: Terrain height, assumed to be 200m.

[0216] V: Vegetation type index, assumed to be 50 (representing medium vegetation cover).

[0217] Weight coefficient ω W =0.4,ω H =0.3,ω V =0.3 corresponds to the weights of wind speed, terrain and vegetation respectively.

[0218] Substitute into the formula and calculate:

[0219]

[0220] I env ≈4+0.3×0.03+0.3×3.684

[0221] I env ≈4+0.009+1.105

[0222] I env =5.114

[0223] The obtained comprehensive environmental index is 5.114, which will be used to further calculate the fire spread potential. A higher value indicates that the current environmental conditions are more conducive to fire spread.

[0224] Assume k P =0.1, T P =50.

[0225] Continue to use I env =5.114, then:

[0226]

[0227] The calculated result is 0.9978, indicating that the fire spread potential is very high (close to 1), which means that the fire is very easy to spread under the current environmental conditions.

[0228] D firemap : The current data value of the dynamic fire map, assuming it is the current fire level, the value is 4.

[0229] Using P spread =0.9978, substitute it into the formula and calculate:

[0230] S path =0.9978×ln(1+4)

[0231] S path =0.9978×ln(5

[0232] S path ≈0.9978×1.609

[0233] S path ≈1.605

[0234] The simulated fire spread path score was 1.605, a value that represents the potential intensity and speed of fire spread, with higher values ​​indicating a high degree of fire activity and potential for spread.

[0235] See also Figure 7 , the steps to obtain the fire spread prediction results are:

[0236] Analyze the simulation results of the fire spread path, predict the spread scenarios under different environmental conditions, and use the formula,

[0237]

[0238] Calculate the difference spread scenario score C diff , generate differential spread scenario analysis data, where W, H are wind speed and terrain height data, α C , β C is the environmental condition sensitivity coefficient;

[0239] Analyze data based on generated differential spread scenarios, check data integrity and analyze potential fire risk areas, using formulas,

[0240] R risk =∫C diff ·dx

[0241] Calculate the fire area risk score R risk , generate potential fire risk area data, where dx represents the length of the integrated area; using the potential fire risk area data, the formula is used,

[0242]

[0243] Calculate the total fire spread risk score P final , generate fire spread prediction results.

[0244] α C and β C are the sensitivity coefficients of wind speed and terrain height, assumed to be 0.05 and 0.02 respectively, regulating the effects of wind speed and terrain on fire spread.

[0245] Assume that the wind speed W is 15m / s. Assume that the height H is 250m.

[0246] Use S path =1.605 for calculation

[0247]

[0248] C diff ≈1.601

[0249] The obtained differential spread scenario score is 1.601, indicating that under the considered wind speed and terrain conditions, the fire may be slightly slowed down under the influence of these conditions, but still maintain a high intensity of spread.

[0250] Assuming that the length of the integrated area dx is 100m (using a simplified example to show how to integrate the impact of fire spread), then:

[0251] R risk =1.601×100

[0252] R risk =160.1

[0253] The calculated total score for the potential fire risk areas is 160.1, which represents the overall fire risk level within the predicted risk areas under the current simulation conditions. This value is high, indicating that attention should be paid to fire prevention measures in these areas.

[0254] Continue to calculate the total fire spread risk score:

[0255] P final =160.1 2

[0256] P final =25632.01

[0257] The final fire spread prediction result is 25632.01. The value represents the total score of the predicted fire spread risk under the given simulation conditions and calculation parameters. A very high value indicates that there is a serious risk of fire spread in the corresponding area and prevention and control measures need to be taken immediately.

[0258] See also Figure 8 , the steps to obtain fire cause and loss investigation records are:

[0259] Based on the fire spread prediction results, economic activity data were collected and integrated from the regional economic statistics departments, and the formula was used to

[0260]

[0261] Calculate the economic activity index E activity , generate economic activity data, where A irepresents the economic volume of the i-th activity, q i is the corresponding weight, n represents the number of economic activities;

[0262] Using economic activity data, investigate the impact of fire on the regional economy. Using the formula,

[0263]

[0264] Calculate the estimated economic loss value L economic , generate economic loss data, where δ L represents the maximum potential loss, ξ L is the fire sensitivity parameter, which is used to adjust the impact of economic activity indicators;

[0265] Collect the behavior data of people and vehicles when a fire occurs, conduct behavior analysis, distinguish normal and abnormal patterns, and use formulas to

[0266]

[0267] Calculate the behavior pattern analysis value and generate pattern recognition results, where B data is the average behavior score, θ B and μ B are the response sensitivity parameter and behavioral excursion threshold, respectively;

[0268] Integrate economic activity data, economic loss data and pattern recognition results to compile and output fire cause and loss investigation records.

[0269] Assume there are three economic activities:

[0270] Manufacturing output valueA 1 =1000 units, weight q 1 =0.5.

[0271] Retail SalesA 2 =500 units, weight q 2 =0.3.

[0272] Service Industry ActivitiesA 3 =300 units, weight q 3 =0.2.

[0273] Substituting into the formula, we get:

[0274] E activity =0.5×1000+0.3×500+0.2×300

[0275] E activity =500+150+60

[0276] E activity =710

[0277] The final economic activity index is 710, reflecting a composite measure of the impact of each economic sector on the total economy after taking into account the weights.

[0278] δ L : Maximum potential loss, set as a certain proportion of the total economy, assumed to be 1000.

[0279] ξ L : Disaster sensitivity parameter, which controls the impact of economic activity indicators, is set to 0.005.

[0280] Substitute into the formula to calculate:

[0281] L economic =1000·(1-e -0.005×710 )

[0282] L economic =1000·(1-e -3.55 )

[0283] L economic ≈1000·(1-0.0287)

[0284] L economic =971.3

[0285] The economic loss estimate is 971.3, which represents the economic value that is expected to be lost under a given fire impact.

[0286] θ B : Response sensitivity coefficient for behavioral data, set to 0.1.

[0287] μ B : Behavior deviation threshold, set to 50.

[0288] B data : Average behavior score, assumed to be 45.

[0289] but:

[0290]

[0291] The result of the behavior pattern analysis is 0.3775, which is a low value, indicating that under the given data, the probability of a normal behavior pattern is low, and there may be many abnormal behaviors, which is conducive to the rapid analysis of the cause of the fire.

[0292] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A fire investigation and survey system supported by cloud computing, characterized in that: The system comprises: The data receiving and preliminary analysis module collects data from temperature sensors and smoke detectors, performs time synchronization and calibrates wind direction data, pre-processes camera and satellite image data, adjusts resolution and contrast, and performs data fusion with calibration data to generate synchronized environmental data; The dynamic fire mapping module integrates the environmental data with the GIS data based on the synchronized environmental data, locates the fire source, depicts the fire spread boundary according to the location result, performs geocoding, updates the regional map, and generates a dynamic fire map; The fire spread prediction module collects wind speed, terrain and vegetation type data based on the dynamic fire map, performs parameter calculations, simulates the fire spread path, predicts different spread scenarios based on the simulation results, analyzes potential fire risk areas, and generates fire spread prediction results; The loss assessment and cause analysis module collects economic activity data for economic impact analysis based on the fire spread prediction results, analyzes the behavior data of personnel and vehicles for behavior pattern analysis, distinguishes normal from abnormal behavior patterns, and generates fire cause and loss investigation records.

2. The cloud computing-supported fire investigation and survey system according to claim 1, characterized in that: The calibration steps of the wind direction data are as follows: Collect data from temperature sensors and smoke detectors, set timestamps, apply formulas, t sensor =t real -δt Calculate the data acquisition time t sensor , generating sensor data with timestamps, where t real represents the actual acquisition time, δt represents the transmission delay of data from the sensor to the processing unit; The sensor data with the timestamp is time synchronized and the formula is applied, t sync =t sensor +Δt sync Calculate the time point t after synchronization sync , generate synchronized time data, where Δt sync is the offset from local time to Coordinated Universal Time; Using the synchronized time data to adjust the wind direction data, apply the formula, W cal =W raw +k(t sync -t wind ) 2 Calculate the calibrated wind direction W cal , generate calibrated wind direction data, where W raw is the original wind direction data, t wind is the wind direction data collection time, and k is the calibration coefficient, which is used to adjust the sensitivity of wind direction data calibration.

3. The cloud computing supported fire investigation and survey system according to claim 2, characterized in that: The steps for acquiring the synchronization environment data are as follows: Preprocess camera and satellite image data, adjust resolution and contrast, apply formulas, Calculate the pixel value I after contrast and brightness adjustment prep , generate preprocessed image data, where I raw is the average original pixel value of the original image data, a I , b I 、c I is the adjustment coefficient, a I Used to adjust contrast, b I To adjust the brightness, c I Used to normalize processing results; The preprocessed image data is time-aligned with the calibrated wind direction data, applying the formula, Calculate the integrated image value I aligned , generate time-aligned image data, where max(W cal ) represents the maximum wind direction after calibration; The time-aligned image data is data-fused with the synchronized time data, and the formula is applied: Calculate the fused environmental data value D fusion , generate synchronous environment data, where I aligned is the time-aligned image data, t sync To synchronize time data, m represents the collection time point, and M represents the total number of collection time points.

4. The cloud computing supported fire investigation and survey system according to claim 3, characterized in that: The steps of locating the fire source are as follows: Based on the synchronized environmental data, the temperature T and smoke concentration S are extracted, and the formula is used: E score =ω1T+ω2S Calculation environment score E score , generate environmental score data, where ω1 and ω2 are weight coefficients, which are used to adjust the weights of temperature and smoke concentration respectively; Using the environmental score data, check data validity, apply formulas, Calculate the fire source probability P fire , generate fire source probability data, where k E Represents the sensitivity coefficient, E thresh is the environmental score threshold; According to the fire source probability data, a location with a probability higher than a threshold value θ is selected as the fire source location, using the formula: L fire ={x∣x∈P fire ,x>Y} Determine the final fire source location L fire , generate fire source location data, where Y is the probability threshold.

5. The cloud computing supported fire investigation and survey system according to claim 4, characterized in that: The steps for obtaining the dynamic fire map are: According to the fire source location data, combined with wind speed W and terrain slope G, the formula is applied: B spread =L fire +αWcos(θ)+βG Calculate the fire spread distance value B spread , generate fire spread boundary data, where α represents the wind speed influence coefficient, β represents the terrain slope coefficient, and L fire , W and G represent the fire source location, wind speed and terrain slope, respectively, and θ represents the angle between the wind direction and the fire source location; Using the fire spread boundary data, geocoding technology is used to update the regional map in the GIS system, using the formula, Calculate the updated map evaluation value M updated , get the updated regional map, where M current Indicates the current map evaluation value, d(B spread ,x) represents B spread The distance to point x on the map, λ is the attenuation coefficient; Based on the updated area map, a dynamically displayed fire map is created using the formula, Calculate the cumulative effect value D of time change on fire firemap , generate a dynamic fire map, where T D represents the time parameter, and σ is the time influence coefficient.

6. The cloud computing supported fire investigation and survey system according to claim 5, characterized in that: The simulation steps of the fire spread path are: Collect wind speed W, terrain height H and vegetation type index V data, and use the formula, Calculate the integration index I env , generate the comprehensive environmental index, where ω W ,ω H and ω V are normalized weight coefficients, corresponding to the weights of wind speed, terrain height, and vegetation type index; The fire spread potential is estimated using the environmental comprehensive index using the formula: Calculate the fire spread potential value P spread , get the fire spread potential data, where k P is the adjustment coefficient used to adjust the impact of the comprehensive environmental index, T P is the baseline threshold and is the sensitive threshold of fire spread potential. Based on the fire spread potential data and combined with the dynamic fire map, the fire spread path is simulated using the formula: S path =P spread ·ln(1+D firemap ) Calculate the fire spread path simulation score S path , generate fire spread path simulation results.

7. The cloud computing supported fire investigation and survey system according to claim 6, characterized in that: The steps for obtaining the fire spread prediction result are: Analyze the simulation results of the fire spread path, predict the spread scenarios under different environmental conditions, and use the formula, Calculate the difference spread scenario score C diff , generate differential spread scenario analysis data, where W, H are wind speed and terrain height data, α C , β C is the environmental condition sensitivity coefficient; Analyze the data according to the different spread scenarios, check the data integrity and analyze the potential fire risk areas, using the formula, R risk =∫C diff ·dx Calculate the fire area risk score R risk , generate potential fire risk area data, where dx represents the length of the integrated area; Using the potential fire risk area data, the formula is adopted: Calculate the total fire spread risk score P final , generate fire spread prediction results.

8. The cloud computing supported fire investigation and survey system according to claim 7, characterized in that: The steps for obtaining the fire cause and loss investigation records are as follows: Based on the fire spread prediction results, economic activity data are collected and integrated from the regional economic statistics department, and the formula is used. Calculate the economic activity index E activity , generate economic activity data, where A i represents the economic volume of the i-th activity, q i is the corresponding weight, n represents the number of economic activities; Using the economic activity data, investigate the impact of fire on the regional economy. Using the formula, Calculate the estimated economic loss value L economic , generate economic loss data, where δ L represents the maximum potential loss, ξ L is the fire sensitivity parameter, which is used to adjust the impact of economic activity indicators; Collect the behavior data of people and vehicles when a fire occurs, conduct behavior analysis, distinguish normal and abnormal patterns, and use formulas to Calculate the behavior pattern analysis value and generate pattern recognition results, where B data is the average behavior score, θ B and μ B are the response sensitivity parameter and behavioral excursion threshold, respectively; Integrate the economic activity data, economic loss data and pattern recognition results to compile and output fire cause and loss investigation records.

Images