Tunnel fire location detection calibration method and system
By acquiring wind speed and temperature distribution data within the tunnel, the positioning error of the heat detector is calculated, and the location of the fire source is corrected. This solves the problem of large positioning errors in tunnel fire detectors, enabling rapid and accurate location of the fire source and improving the efficiency and safety of fire rescue.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHANGAN UNIV
- Filing Date
- 2026-03-18
- Publication Date
- 2026-06-05
AI Technical Summary
Due to the influence of longitudinal wind, tunnel fire detectors suffer from large errors in fire source location, making it difficult to quickly and accurately determine the location of the fire source.
By acquiring wind speed and temperature distribution data within the tunnel, the positioning error of the heat detector is calculated, and the detection location of the fire source is corrected.
It enables rapid and accurate location of fire sources within tunnels, provides timely fire information, and improves the efficiency and safety of fire rescue.
Smart Images
Figure CN122157449A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of fire detection technology, and in particular to a method and system for calibrating the location of a tunnel fire. Background Technology
[0002] Due to the complex structure, confined space, and discrepancy between internal and external lighting conditions, tunnel accidents are frequent. Fire is one of the most significant hazards faced by tunnels. The unique structure of tunnels leads to rapid fire spread and slow smoke dispersion, making evacuation difficult, fire detection and suppression challenging, and resulting in significant casualties and economic losses. Currently, tunnel fire detection primarily relies on heat detectors, which calculate the scale and location of a fire by monitoring the temperature of the tunnel arch. However, in addition to natural wind, tunnels often require ventilation fans. The longitudinal airflow within the tunnel alters thermal convection and smoke distribution, slowing the temperature rise directly above the ignition point and increasing the temperature downstream, thus affecting the temperature detection accuracy of fire detectors.
[0003] Tunnel fires are a common and serious safety hazard in modern transportation infrastructure. Rapid fire detection and location are crucial for fire suppression, personnel evacuation, and safety protection. Chinese patent application CN117131451A discloses a multi-spatial environment fire source localization method and system based on AttentionLSTM. This method locates the fire source through temperature detection; however, there is often a deviation between the fire source and the location of the highest temperature detected by the system. This deviation is called the fire source detection offset distance, which is the difference between the actual location of the fire source and the location of the highest temperature detected by the system. In practical engineering, distributed fiber optic temperature measurement systems can provide temperature data distribution throughout the tunnel, but linear temperature detectors, which locate fire sources through temperature detection, have a high margin of error and cannot determine the fire source location in a timely and accurate manner. Summary of the Invention
[0004] The purpose of this invention is to provide a method and system for detecting and calibrating the location of a fire in a tunnel. By acquiring wind speed and temperature distribution data in the tunnel, the positioning error of the heat sensor is calculated, thereby correcting the detection location of the fire source.
[0005] A method for calibrating the location of a tunnel fire, comprising: S1, with the initial alarm location as the origin, the downstream direction of the fire source as the positive coordinate axis and the upstream direction of the fire source as the negative coordinate axis, construct a relative position coordinate system; S2, obtain the current wind speed value in the tunnel and judge the current wind speed value. When the current wind speed value is less than the second wind speed setting value, execute S3. When the current wind speed value is greater than or equal to the second wind speed setting value, execute S8. S3, acquire the temperature field data of the tunnel within a first set time period, as the first temperature data; the first set time period is the current time and several time steps before it; S4, obtain the location data of the first temperature peak point and the location data of the first temperature abrupt change point based on the first temperature data; S5, determine the current wind speed value; if the current wind speed value is less than or equal to the first wind speed setting value, execute S6; if the current wind speed value is greater than the first wind speed setting value and less than the second wind speed setting value, execute S7; the second wind speed setting value is greater than the first wind speed setting value. S6, obtain the difference between the maximum and minimum positions in the first temperature peak point position data, denoted as P1, obtain the minimum position in the first temperature change point position data, denoted as P2, and execute S11; S7, obtain the average value of the difference between the first temperature peak point location data and the first temperature change point location data at each time step, denoted as P1, obtain the minimum position in the first temperature peak point location data, denoted as P2, and execute S11; S8, acquire the temperature field data of the tunnel within the second set time period as the second temperature data; the second set time period is from the first alarm time to the current time; S9, obtain the location data of the second temperature peak point and the location data of the second temperature abrupt change point based on the second temperature data; S10, obtain the average value of the difference between the second temperature peak point location data and the second temperature change point location data at each time step, denoted as P1, obtain the minimum position in the second temperature peak point location data, denoted as P2, and execute S11; S11, obtain the predicted fire source location based on P1 and P2, and obtain the predicted fire source offset distance based on the predicted fire source location and the first alarm location.
[0006] Optionally, the expression for obtaining the predicted fire source location based on P1 and P2 is: Y1 = P2 - P1; Where Y1 is the predicted location of the fire source; The expression for the fire source prediction offset distance obtained based on the predicted fire source location and the initial alarm location is as follows: D1 = abs(Y1 - Y0); Where D1 is the predicted offset distance of the fire source and Y0 is the location of the first alarm.
[0007] Optionally, the maximum value in the temperature field data within the tunnel at each time step is taken as the temperature peak point location at the current time step, thus obtaining the first temperature peak point location data or the second temperature peak point location data.
[0008] Optionally, based on the temperature data, the temperature gradient data inside the tunnel at each time step is obtained using the first-order difference method; The first point in the temperature gradient data that exceeds the dynamic threshold is taken as the temperature abrupt change point location at the current time step, thus obtaining the first temperature abrupt change point location data and the second temperature abrupt change point location data.
[0009] Optionally, the dynamic threshold expression is: ; in, This represents the normal temperature field matrix, where N is the length of either the first or second preset time period, and n is the total number of temperature sampling points within the tunnel. Let j be the temperature value at the j-th temperature sampling point at the i-th time step. As a multiplier factor, The gradient mean, , The horizontal gradient matrix, , This is a dynamic threshold.
[0010] The present invention also provides a tunnel fire location detection and calibration system, comprising: The coordinate system module is used to construct a relative position coordinate system with the initial alarm location as the origin, the downstream direction of the fire source as the positive coordinate axis, and the upstream direction of the fire source as the negative coordinate axis. The first wind speed judgment module is used to obtain the current wind speed value in the tunnel and judge the current wind speed value. When the current wind speed value is less than the second wind speed setting value, the first data acquisition module is executed. When the current wind speed value is greater than or equal to the second wind speed setting value, the second data acquisition module is executed. The first data acquisition module is used to acquire temperature field data of the tunnel within a first set time period, as the first temperature data; the first set time period is the current moment and several time steps before it. The first data processing module is used to obtain the location data of the first temperature peak point and the location data of the first temperature change point based on the first temperature data. The second wind speed determination module is used to determine the current wind speed value. When the current wind speed value is less than or equal to the first wind speed setting value, the first position module is executed. When the current wind speed value is greater than the first wind speed setting value and less than the second wind speed setting value, the second position module is executed. The second wind speed setting value is greater than the first wind speed setting value. The first position module is used to obtain the difference between the maximum and minimum positions in the first temperature peak point position data, denoted as P1, obtain the minimum position in the first temperature change point position data, denoted as P2, and execute the position revision module. The second position module is used to obtain the average value of the difference between the first temperature peak point position data and the first temperature change point position data at each time step, denoted as P1, obtain the minimum position in the first temperature peak point position data, denoted as P2, and execute the position revision module. The second data acquisition module is used to acquire the temperature field data of the tunnel within a second set time period, as the second temperature data; the second set time period is from the first alarm time to the current time. The second data processing module is used to obtain the location data of the second temperature peak point and the location data of the second temperature change point based on the second temperature data. The third position module is used to obtain the average value of the difference between the second temperature peak point position data and the second temperature change point position data at each time step, denoted as P1, obtain the minimum position in the second temperature peak point position data, denoted as P2, and execute the position revision module. The location revision module is used to obtain the predicted fire source location based on P1 and P2, and to obtain the predicted fire source offset distance based on the predicted fire source location and the initial alarm location.
[0011] Optionally, the expression for obtaining the predicted fire source location based on P1 and P2 is: Y1 = P2 - P1; Where Y1 is the predicted location of the fire source; The expression for the fire source prediction offset distance obtained based on the predicted fire source location and the initial alarm location is as follows: D1 = abs(Y1 - Y0); Where D1 is the predicted offset distance of the fire source and Y0 is the location of the first alarm.
[0012] Optionally, the maximum value in the temperature field data within the tunnel at each time step is taken as the temperature peak point location at the current time step, thus obtaining the first temperature peak point location data or the second temperature peak point location data.
[0013] Optionally, based on the temperature data, the temperature gradient data inside the tunnel at each time step is obtained using the first-order difference method; The first point in the temperature gradient data that exceeds the dynamic threshold is taken as the temperature abrupt change point location at the current time step, thus obtaining the first temperature abrupt change point location data and the second temperature abrupt change point location data.
[0014] Optionally, the dynamic threshold expression is: ; in, This represents the normal temperature field matrix, where N is the length of either the first or second preset time period, and n is the total number of temperature sampling points within the tunnel. Let j be the temperature value at the j-th temperature sampling point at the i-th time step. As a multiplier factor, The gradient mean, , The horizontal gradient matrix, , This is a dynamic threshold.
[0015] The effects of this invention are as follows: This invention relates to a tunnel fire location detection and calibration method. By acquiring current wind speed and temperature distribution data along the tunnel, it calculates the positioning error of the current temperature sensor, thereby correcting the detection location of the fire source. This enables rapid location of fire sources within the tunnel, providing more timely and accurate fire information for fire rescue operations, and is of great significance to the operational safety of tunnel engineering. Attached Figure Description
[0016] Figure 1 This is a flowchart of the tunnel fire location detection and calibration method of the present invention; Figure 2 This is a schematic diagram of the temperature peak point and abrupt change point of the present invention. Detailed Implementation
[0017] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
[0018] Figure 1 This is a flowchart of the tunnel fire location detection and calibration method of the present invention, shown in Figure 1. The present invention provides a tunnel fire location detection and calibration method, which includes: S1, with the initial alarm location as the origin, the downstream direction of the fire source as the positive coordinate axis and the upstream direction of the fire source as the negative coordinate axis, constructs a relative position coordinate system.
[0019] S2: Obtain the current wind speed value inside the tunnel and determine if the current wind speed value is less than the second wind speed setting value. If the current wind speed value is greater than or equal to the second wind speed setting value, execute S3. If the current wind speed value is greater than or equal to the second wind speed setting value, execute S8. Preferably, the second wind speed setting value is 2.5 m / s.
[0020] S3, acquire the temperature field data of the tunnel within a first set time period, as the first temperature data; the first set time period is the current moment and several time steps prior. Specifically, the temperature field data is the one-dimensional temperature distribution data along the tunnel's axial direction.
[0021] Preferably, the temperature field data acquisition process is as follows: 1) File verification First, check if the file at the specified path exists; if it does not exist, throw a FileNotFoundError. Extract the filename (excluding the path) for subsequent parsing.
[0022] 2) Filename parsing The expected filename format is: X_Y_devc.xlsx (e.g., 12_34_devc.xlsx); Use underscores to separate filenames and combine the first and second parts into a floating-point number; example: 12_34_devc.xlsx → 12.34; this value is interpreted as a wind speed value, and a ValueError is thrown if the filename is not in the correct format or cannot be converted to a floating-point number.
[0023] 3) Data Reading Use pandas' read_excel function to read an Excel file; The header=None parameter indicates that the file has no column header rows; Read the raw data into a DataFrame.
[0024] 4) Data format validation Verify that the DataFrame has a shape of 9 rows × 82 columns; If it is not the expected shape, throw a ValueError.
[0025] 5. Data Structure Extraction Extract from a 9×82 data matrix: Time series: All values in column 0 (9 time points in total); Temperature field data: All data from column 1 to column 81 (a temperature field matrix of 9 rows × 81 columns).
[0026] S4, obtain the location data of the first temperature peak point and the location data of the first temperature change point based on the first temperature data.
[0027] Specifically, such as Figure 2Example: Peak positions (coordinates): [27 28 28 27 25 27 24 25 25]; Peak values: [ 26.105768 34.260309 41.080576 45.982522 53.273636 58.751981 88.505479 157.13757 170.95738 ]; Dynamic gradient threshold: 1.29 (example value, dynamically adjusted with input data); Jump positions (coordinates): [24, 21, 20, 17, 12, 8, 14, 6, 1].
[0028] Specifically, the maximum value in the temperature field data within the tunnel at each time step is taken as the temperature peak point location at the current time step, thus obtaining the first temperature peak point location data or the second temperature peak point location data.
[0029] Based on temperature data, the temperature gradient data within the tunnel at each time step is obtained using the first-order difference method. This is used to identify the frontal zone of the heat wave propagation from the fire source, thereby calculating the temperature gradient.
[0030] The first point in the temperature gradient data that exceeds the dynamic threshold is taken as the location of the temperature abrupt change point in the current time step, thus obtaining the location data of the first and second temperature abrupt change points.
[0031] The dynamic threshold is calculated using a weighted average of the standard deviations of temperature changes to ensure adaptability to disturbance responses under different wind speeds and heat flux intensities. Specifically, the expression for the dynamic threshold is: ; in, This represents the normal temperature field matrix, where N is the length of either the first or second preset time period, and n is the total number of temperature sampling points within the tunnel. Let j be the temperature value at the j-th temperature sampling point at the i-th time step. As a multiplier factor, The gradient mean, , The horizontal gradient matrix, , This is a dynamic threshold.
[0032] S5: Determine the current wind speed value. If the current wind speed value is less than or equal to the first wind speed setting value, execute S6. If the current wind speed value is greater than the first wind speed setting value but less than the second wind speed setting value, execute S7. The second wind speed setting value is greater than the first wind speed setting value. Preferably, the first wind speed setting value is 1.5 m / s.
[0033] S6, obtain the difference between the maximum and minimum positions in the first temperature peak point position data, denoted as P1, obtain the minimum position in the first temperature abrupt change point position data, denoted as P2, and execute S11.
[0034] S7, obtain the average value of the difference between the first temperature peak point location data and the first temperature change point location data at each time step, denoted as P1, obtain the minimum position in the first temperature peak point location data, denoted as P2, and execute S11.
[0035] S8, acquire the tunnel temperature field data within a second set time period, as the second temperature data; the second set time period is from the first alarm time to the current time. Specifically, the temperature field data is acquired every 10 seconds.
[0036] S9: Obtain the location data of the second temperature peak point and the location data of the second temperature abrupt change point based on the second temperature data. Refer to S4 for the specific acquisition method.
[0037] S10: Obtain the average value of the difference between the second temperature peak point location data and the second temperature abrupt change point location data at each time step, denoted as P1; obtain the minimum position in the second temperature peak point location data, denoted as P2; and execute S11.
[0038] S11, based on P1 and P2, the predicted fire source location is obtained, and based on the predicted fire source location and the first alarm location, the predicted fire source offset distance is obtained.
[0039] Specifically, the expression for predicting the location of the fire source is: Y1 = P2 - P1; Where Y1 is the predicted location of the fire source.
[0040] The expression for the fire source prediction offset distance is: D1 = abs(Y1 - Y0); Where D1 is the predicted offset distance of the fire source and Y0 is the location of the first alarm.
[0041] The present invention also provides a tunnel fire location detection and calibration system, comprising: The coordinate system module is used to construct a relative position coordinate system with the initial alarm location as the origin, the downstream direction of the fire source as the positive coordinate axis, and the upstream direction of the fire source as the negative coordinate axis.
[0042] The first wind speed judgment module is used to obtain the current wind speed value in the tunnel and judge the current wind speed value. When the current wind speed value is less than the second wind speed setting value, the first data acquisition module is executed. When the current wind speed value is greater than or equal to the second wind speed setting value, the second data acquisition module is executed.
[0043] The first data acquisition module is used to acquire the temperature field data of the tunnel within a first set time period, which is used as the first temperature data; the first set time period is the current moment and several time steps before it.
[0044] The first data processing module is used to obtain the location data of the first temperature peak point and the location data of the first temperature change point based on the first temperature data.
[0045] The second wind speed judgment module is used to judge the current wind speed value. When the current wind speed value is less than or equal to the first wind speed setting value, the first position module is executed. When the current wind speed value is greater than the first wind speed setting value and less than the second wind speed setting value, the second position module is executed. The second wind speed setting value is greater than the first wind speed setting value.
[0046] The first position module is used to obtain the difference between the maximum and minimum positions in the first temperature peak point position data, denoted as P1, obtain the minimum position in the first temperature change point position data, denoted as P2, and execute the position revision module.
[0047] The second position module is used to obtain the average value of the difference between the first temperature peak point position data and the first temperature change point position data at each time step, denoted as P1, obtain the minimum position in the first temperature peak point position data, denoted as P2, and execute the position revision module.
[0048] The second data acquisition module is used to acquire the temperature field data of the tunnel within a second set time period, which is used as the second temperature data; the second set time period is from the time of the first alarm to the current time.
[0049] The second data processing module is used to obtain the location data of the second temperature peak point and the location data of the second temperature change point based on the second temperature data.
[0050] The third position module is used to obtain the average value of the difference between the second temperature peak point position data and the second temperature change point position data at each time step, denoted as P1, obtain the minimum position in the second temperature peak point position data, denoted as P2, and execute the position revision module.
[0051] The location revision module is used to obtain the predicted fire source location based on P1 and P2, and to obtain the predicted fire source offset distance based on the predicted fire source location and the initial alarm location.
[0052] Specifically, based on P1 and P2, the expression for predicting the fire source location is as follows: Y1 = P2 - P1; Where Y1 is the predicted location of the fire source; Based on the predicted fire source location and the initial alarm location, the expression for the predicted fire source offset distance is as follows: D1 = abs(Y1 - Y0); Where D1 is the predicted offset distance of the fire source and Y0 is the location of the first alarm.
[0053] Preferably, the maximum value in the temperature field data within the tunnel at each time step is taken as the temperature peak point location at the current time step, thus obtaining the first temperature peak point location data or the second temperature peak point location data.
[0054] Based on temperature data, the temperature gradient data inside the tunnel at each time step is obtained using the first-order difference method; The first point in the temperature gradient data that exceeds the dynamic threshold is taken as the location of the temperature abrupt change point in the current time step, thus obtaining the location data of the first and second temperature abrupt change points.
[0055] Furthermore, the dynamic threshold expression is: ; in, This represents the normal temperature field matrix, where N is the length of either the first or second preset time period, and n is the total number of temperature sampling points within the tunnel. Let j be the temperature value at the j-th temperature sampling point at the i-th time step. As a multiplier factor, The gradient mean, , The horizontal gradient matrix, , This is a dynamic threshold.
[0056] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.
Claims
1. A method for calibrating the location of a tunnel fire, characterized in that, It includes: S1, with the initial alarm location as the origin, the downstream direction of the fire source as the positive coordinate axis and the upstream direction of the fire source as the negative coordinate axis, construct a relative position coordinate system; S2, obtain the current wind speed value in the tunnel and judge the current wind speed value. When the current wind speed value is less than the second wind speed setting value, execute S3. When the current wind speed value is greater than or equal to the second wind speed setting value, execute S8. S3, acquire the temperature field data of the tunnel within a first set time period, as the first temperature data; the first set time period is the current time and several time steps before it; S4, obtain the location data of the first temperature peak point and the location data of the first temperature abrupt change point based on the first temperature data; S5, determine the current wind speed value; if the current wind speed value is less than or equal to the first wind speed setting value, execute S6; if the current wind speed value is greater than the first wind speed setting value and less than the second wind speed setting value, execute S7; the second wind speed setting value is greater than the first wind speed setting value. S6, obtain the difference between the maximum and minimum positions in the first temperature peak point position data, denoted as P1, obtain the minimum position in the first temperature change point position data, denoted as P2, and execute S11; S7, obtain the average value of the difference between the first temperature peak point location data and the first temperature change point location data at each time step, denoted as P1, obtain the minimum position in the first temperature peak point location data, denoted as P2, and execute S11; S8, acquire the temperature field data of the tunnel within the second set time period as the second temperature data; the second set time period is from the first alarm time to the current time; S9, obtain the location data of the second temperature peak point and the location data of the second temperature abrupt change point based on the second temperature data; S10, obtain the average value of the difference between the second temperature peak point location data and the second temperature change point location data at each time step, denoted as P1, obtain the minimum position in the second temperature peak point location data, denoted as P2, and execute S11; S11, obtain the predicted fire source location based on P1 and P2, and obtain the predicted fire source offset distance based on the predicted fire source location and the first alarm location.
2. The tunnel fire location detection and calibration method according to claim 1, characterized in that, The expression for predicting the fire source location based on P1 and P2 is as follows: Y1 = P2 - P1; Where Y1 is the predicted location of the fire source; The expression for the fire source prediction offset distance obtained based on the predicted fire source location and the initial alarm location is as follows: D1 = abs(Y1 - Y0); Where D1 is the predicted offset distance of the fire source and Y0 is the location of the first alarm.
3. The tunnel fire location detection and calibration method according to claim 1, characterized in that, The maximum value in the temperature field data within the tunnel at each time step is taken as the temperature peak point location at the current time step, thus obtaining the first temperature peak point location data or the second temperature peak point location data.
4. The tunnel fire location detection and calibration method according to claim 1, characterized in that, Based on temperature data, the temperature gradient data inside the tunnel at each time step is obtained using the first-order difference method; The first point in the temperature gradient data that exceeds the dynamic threshold is taken as the temperature abrupt change point location at the current time step, thus obtaining the first temperature abrupt change point location data and the second temperature abrupt change point location data.
5. The tunnel fire location detection and calibration method according to claim 4, characterized in that, The dynamic threshold expression is: ； in, This represents the normal temperature field matrix, where N is the length of either the first or second preset time period, and n is the total number of temperature sampling points within the tunnel. Let j be the temperature value at the j-th temperature sampling point at the i-th time step. As a multiplier factor, The gradient mean, , The horizontal gradient matrix, , This is a dynamic threshold.
6. A tunnel fire location detection and calibration system, characterized in that, It includes: The coordinate system module is used to construct a relative position coordinate system with the initial alarm location as the origin, the downstream direction of the fire source as the positive coordinate axis, and the upstream direction of the fire source as the negative coordinate axis. The first wind speed judgment module is used to obtain the current wind speed value in the tunnel and judge the current wind speed value. When the current wind speed value is less than the second wind speed setting value, the first data acquisition module is executed. When the current wind speed value is greater than or equal to the second wind speed setting value, the second data acquisition module is executed. The first data acquisition module is used to acquire temperature field data of the tunnel within a first set time period, as the first temperature data; the first set time period is the current moment and several time steps before it. The first data processing module is used to obtain the location data of the first temperature peak point and the location data of the first temperature change point based on the first temperature data. The second wind speed determination module is used to determine the current wind speed value. When the current wind speed value is less than or equal to the first wind speed setting value, the first position module is executed. When the current wind speed value is greater than the first wind speed setting value and less than the second wind speed setting value, the second position module is executed. The second wind speed setting value is greater than the first wind speed setting value. The first position module is used to obtain the difference between the maximum and minimum positions in the first temperature peak point position data, denoted as P1, obtain the minimum position in the first temperature change point position data, denoted as P2, and execute the position revision module. The second position module is used to obtain the average value of the difference between the first temperature peak point position data and the first temperature change point position data at each time step, denoted as P1, obtain the minimum position in the first temperature peak point position data, denoted as P2, and execute the position revision module. The second data acquisition module is used to acquire the temperature field data of the tunnel within a second set time period, as the second temperature data; the second set time period is from the first alarm time to the current time. The second data processing module is used to obtain the location data of the second temperature peak point and the location data of the second temperature change point based on the second temperature data. The third position module is used to obtain the average value of the difference between the second temperature peak point position data and the second temperature change point position data at each time step, denoted as P1, obtain the minimum position in the second temperature peak point position data, denoted as P2, and execute the position revision module. The location revision module is used to obtain the predicted fire source location based on P1 and P2, and to obtain the predicted fire source offset distance based on the predicted fire source location and the initial alarm location.
7. The tunnel fire location detection and calibration system according to claim 6, characterized in that, The expression for predicting the fire source location based on P1 and P2 is as follows: Y1 = P2 - P1; Where Y1 is the predicted location of the fire source; The expression for the fire source prediction offset distance obtained based on the predicted fire source location and the initial alarm location is as follows: D1 = abs(Y1 - Y0); Where D1 is the predicted offset distance of the fire source and Y0 is the location of the first alarm.
8. The tunnel fire location detection and calibration system according to claim 6, characterized in that, The maximum value in the temperature field data within the tunnel at each time step is taken as the temperature peak point location at the current time step, thus obtaining the first temperature peak point location data or the second temperature peak point location data.
9. The tunnel fire location detection and calibration system according to claim 6, characterized in that, Based on temperature data, the temperature gradient data inside the tunnel at each time step is obtained using the first-order difference method; The first point in the temperature gradient data that exceeds the dynamic threshold is taken as the temperature abrupt change point location at the current time step, thus obtaining the first temperature abrupt change point location data and the second temperature abrupt change point location data.
10. The tunnel fire location detection and calibration system according to claim 9, characterized in that, The dynamic threshold expression is: ； in, This represents the normal temperature field matrix, where N is the length of either the first or second preset time period, and n is the total number of temperature sampling points within the tunnel. Let j be the temperature value at the j-th temperature sampling point at the i-th time step. As a multiplier factor, The gradient mean, , The horizontal gradient matrix, , This is a dynamic threshold.