A mobile fire detection system and method

Abstract

The application discloses a mobile fire detection system and method, and belongs to the technical field of fire detection. The system realizes over-distance accurate positioning and sub-pixel level precision. Through a unique "polynomial fitting + extreme value calculation" algorithm, the system can perform three times curve fitting on the signals of 2-3 detection units covered by the fire source imaging, and can analyze the sub-pixel level center coordinates by derivation. In combination with a preset coordinate mapping table, the fire source positioning precision can be improved to within 0.3 meters at a distance of 150 meters, and the specific tunnel post number and lane position can be accurately output, so that the limitation of the traditional detector that can only give an area alarm is completely changed. In terms of adaptability to complex environments, the application not only solves three major pain points of the traditional flame detector, i.e., positioning loss, weak anti-vibration and high false alarm rate, but also provides a complete set of high-reliability, intelligent and practical fire monitoring and emergency command tools for tunnel safety.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of fire detection, and in particular to a mobile fire detection system and method. BACKGROUND

[0002] As a key node of traffic infrastructure, the highway tunnel has the characteristics of long space (single hole length often exceeds 1km), closed environment (slow smoke and heat diffusion, low visibility), dense traffic flow (daily traffic volume exceeds ten thousand vehicles), and complex interference sources (vehicle vibration, high temperature exhaust, and lighting), which puts high requirements on the flame detection system:

[0003] (1) Need for precise positioning at a distance: tunnel fires spread quickly (5-10m per minute in a closed environment), and the specific stake number and lane position of the fire source need to be determined within 100m+ distance, otherwise rescue personnel need to check each section, delaying the golden opportunity for extinguishing the fire (every 1 minute delay, the fire loss expands by 30%);

[0004] (2) Strong anti-seismic performance is required: heavy truck traffic generates 10Hz-50Hz low-frequency vibration (amplitude 0.1mm-0.5mm), and tunnel structure resonance generates 20Hz-30Hz vibration (amplitude up to 0.8mm), which easily leads to optical deviation of the detector, increases signal noise, and affects detection accuracy;

[0005] (3) Low false alarm rate is required: interference sources such as vehicle exhaust pipes (800℃-1000℃, strong infrared radiation), brake pad overheating (2μm-2.5μm infrared wavelength), and sodium lamp / LED lighting (visible light-near infrared band) easily lead to false alarms (more than 5 false alarms per month) of traditional detectors, causing frequent closure of the tunnel and affecting traffic efficiency.

[0006] In the prior art, the traditional point-type infrared flame detector has three major defects:

[0007] (1) Lack of positioning capability: only the alarm signal of "fire in a certain area" can be output, and the stake number and lane of the fire source cannot be determined, which cannot meet the precise guidance requirements of tunnel rescue;

[0008] (2) Weak anti-seismic performance: no special vibration reduction design, optical coaxiality deviation exceeds 0.15mm in vibration environment, positioning error expands to more than 1m, and signal-to-noise ratio decreases to less than 22dB;

[0009] (3) Poor anti-interference ability: relying on single infrared radiation intensity judgment, unable to distinguish between moving fire source and vehicle high temperature / illumination interference, false alarm rate is as high as 5 times per month, and there is no adaptive design for tunnel temperature and humidity, and electromagnetic environment. SUMMARY

[0010] To solve the above problems, the technical solution provided by the present application is:

[0011] The mobile fire detection system comprises a lens module and a detection matrix module arranged in cooperation, the lens module collects light data and transmits the light data to the detection matrix module for photoelectric conversion, and the converted signal is sent to a data processing module, and the data processing process of the data processing module is as follows:

[0012] S100, signal preprocessing, mean filtering (window size 3) is performed on the original signal to eliminate high-frequency noise;

[0013] S200, peak detection, 2-3 adjacent units (continuous units with signal values higher than the threshold value) covered by the fire source imaging are identified;

[0014] S300, polynomial fitting, the least square method is used to solve the cubic polynomial coefficients a, b, c, and d;

[0015] S400, extreme value calculation, the derivative equation is solved to obtain the sub-pixel level center coordinates;

[0016] S500, coordinate conversion, according to the preset "array unit coordinate-tunnel pile number-lane position" mapping table, the actual position information is output.

[0017] Preferably, the step S300 is specifically

[0018] Let the detection unit coordinates be x0, x1, and x2 (distributed along the X axis), the corresponding signal values be y0, y1, and y2, and the cubic polynomial expression be y = ax³ + bx² + cx + d;

[0019] The least square fitting is performed using curve_fit, and p0 is the initial parameter guess value;

[0020] p0 = [0, -1, 0, max(y)] assumes that the curve opens downward, and the vertex is close to the maximum signal value;

[0021] popt, _ = curve_fit(self.cubic_polynomial, x, y, p0=[0, -1, 0, max(y)]);

[0022] The coefficients [a, b, c, d] obtained by fitting are returned.

[0023] Preferably, the step S400 is specifically

[0024] S410, fitting a cubic polynomial, obtaining coefficients a, b, c, and d,

[0025] a, b, c, d = self.fit_polynomial(x_units, signals)

[0026] S420, solve the point where the derivative is 0 as the extreme point

[0027] First derivative: y' = 3ax2 + 2bx + c

[0028] Let y' = 0, solve the quadratic equation: 3ax2 + 2bx + c = 0

[0029] Discriminant Δ = (2b)2 - 4 (3a) c;

[0030] If the discriminant is less than 0, there is no real root, which means that the curve has no obvious peak value, and the average coordinates of the covering unit are used as the center at this time;

[0031] S430, calculate the extreme point and filter the effective solution (located in the unit coordinate range)

[0032] x1 = (-2 b + np.sqrt(discriminant)) / (6 a);

[0033] x2 = (-2 b - np.sqrt(discriminant)) / (6 a);

[0034] candidates = [x for x in [x1, x2] if min(x_units) <= x <= max(x_units)];

[0035] S440, select the extreme point with the maximum signal value, which corresponds to the sub-pixel center coordinate (unit level).

[0036] Preferably, the coordinate conversion in step S500 is specifically

[0037] Convert the sub-pixel coordinates into tunnel post number and lane information, the sub-pixel coordinates including X-axis sub-pixel coordinates (unit level) and Y-axis sub-pixel coordinates (unit level), calculate the post number offset in X-axis direction (unit coordinate difference x unit interval x proportion coefficient) for the X-axis sub-pixel coordinates (unit level) and compare it with the reference post number to obtain the actual post number, and calculate the lane information for the Y-axis sub-pixel coordinates (unit level).

[0038] Preferably, the lens module comprises a shell and a micro-heating sheet arranged outside the shell, the shell comprises, in sequence, an auxiliary collimating mirror, a main lens, a narrow-band filter and a temperature sensor, and the shell is further provided with a laser positioning calibrator.

[0039] Preferably, the detection matrix module comprises a ceramic substrate and a plurality of sensing units arranged uniformly on the ceramic substrate, the ceramic substrate is provided with a buffer layer on one side surface of the sensing units, and the ceramic substrate is provided with a signal lead-out layer on the other side surface away from the sensing units, and the detection matrix module is arranged in close contact with the temperature sensor.

[0040] Preferably, the main lens is made of single-crystal germanium (Ge) and has a focal length of 300 mm, an aperture of 60 mm and a field of view angle of 8°; the surface of the main lens is coated with a Ge-ZnSe-Ge three-layer dielectric antireflection film; and the auxiliary collimating mirror (170) is a front concave collimating mirror with a focal length of -50 mm.

[0041] Preferably, the sensing unit adopts a LiTaO3-Nb2O5-ZrO2 ternary solid solution sensitive layer (Nb2O5 doping amount 5 mol%, ZrO2 doping amount 3 mol%) and has a thermoelectric coefficient ≥500 μC / (m² K); the sensing unit (220) has a size of 0.8 mm×0.8 mm, a voltage response rate ≥1.5 V / W (@4.3 μm, 1 kHz modulation frequency) and a noise equivalent power (NEP) ≤2×10 -11 W / √Hz.

[0042] A mobile fire detection method, which adopts the detection system described above, and the method comprises the following steps:

[0043] The lens module and the detection matrix module collect qualified light source signal data and send the data to a data processing module, and the data processing module performs the following operations: S100, signal preprocessing, mean filtering (window size 3) is performed on the original signal to eliminate high-frequency noise;

[0044] S200, peak detection, identifying 2-3 adjacent units (continuous units with signal values higher than a threshold value) covered by fire source imaging;

[0045] S300, polynomial fitting, solving the cubic polynomial coefficients a, b, c, d by using the least square method;

[0046] S400, extreme value calculation, obtaining the sub-pixel level center coordinates by solving the derivative equation;

[0047] S500, coordinate conversion, outputting actual position information according to a preset "array unit coordinate-tunnel pile number-lane position" mapping table.

[0048] Preferably, the radiation intensity threshold is dynamically adjusted according to the tunnel vehicle flow density, and the threshold is set to 0.03 W / cm2 during a vehicle flow peak and to 0.015 W / cm2 during a vehicle flow valley; and multiple mobile fire sources are identified and sorted according to radiation intensity to alarm, and a spread range within 1 minute is predicted in combination with a tunnel wind speed.

[0049] Compared with the prior art, the technical scheme provided by the application has the following beneficial effects:

[0050] The mobile fire detection system and method of the application realize over-distance accurate positioning and sub-pixel level accuracy: through a unique "polynomial fitting + extreme value calculation" algorithm, the system can perform cubic curve fitting on the signals of 2-3 detection units covered by the fire source imaging, and can analytically derive the sub-pixel level center coordinates (such as X-axis 17.23 units in the example) by derivation, and then in combination with a preset coordinate mapping table, the fire source positioning accuracy can be improved to within 0.3 meters at a distance of 150 meters, and the specific tunnel pile number (such as K1+199.9) and lane position can be accurately output, which completely changes the limitation of the traditional detector that can only "alarm in a region". In terms of adaptability in complex environments, the scheme not only solves the three major pain points of positioning loss, weak anti-vibration and high false alarm rate of the traditional flame detector, but also provides a complete set of high-reliability, intelligent and practical fire detection and emergency command tool for tunnel safety. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 Fig. 1 is a structural schematic view of a lens module of the application;

[0052] Figure 2 Fig. 2 is a sectional view of a detection matrix module of the application;

[0053] Figure 3 Fig. 3 is a top view of the detection matrix module of the application.

[0054] Explanation of reference numerals in the schematic view:

[0055] 110, housing; 120, micro heating sheet; 130, narrow-band filter; 140, temperature sensor; 150, laser positioning calibrator; 160, dustproof glass; 170, auxiliary collimating mirror; 180, main lens;

[0056] 210, buffer layer; 220, sensing unit; 230, ceramic substrate; 240, signal lead-out layer. DETAILED DESCRIPTION

[0057] In order to make the personnel in the technical field better understand the scheme of the present application, the technical scheme in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor should belong to the scope of protection of the present application.

[0058] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0059] In the present application, the terms "upper", "lower", "left", "right", "front", "back", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "lateral", "longitudinal", and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. These terms are mainly used to better describe the present application and its embodiments, and are not used to limit the indicated devices, elements or components to have a specific orientation, or to be constructed and operated in a specific orientation.

[0060] In addition, in addition to being used to indicate the orientation or positional relationship, the above-mentioned part of the terms can also be used to indicate other meanings, for example, the term "upper" can also be used to indicate a certain dependent relationship or connection relationship in some cases. For those of ordinary skill in the art, the specific meaning of these terms in the present application can be understood according to the specific circumstances.

[0061] In addition, the terms "mount", "set", "provided with", "connected", "connected", "sleeved" should be broadly understood. For example, it can be a fixed connection, a detachable connection, or a monolithic structure; it can be a mechanical connection, or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or internal communication between two devices, elements or components. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0062] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other in the case of no conflict. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0063] Embodiment 1

[0064] Referring to the accompanying drawings Figure 1 、 Figure 2 and Figure 3 , the mobile fire detection system of the embodiment comprises a lens module and a detection matrix module arranged in cooperation with each other, the lens module collects light data and transmits the light data to the detection matrix module for photoelectric conversion, and the converted signal is sent to a data processing module, and the data processing process of the data processing module is as follows:

[0065] S100, signal preprocessing, mean filtering (window size 3) is performed on the original signal to eliminate high-frequency noise;

[0066] S200, peak detection, 2-3 adjacent units (continuous units with signal values higher than the threshold value) covered by the fire source imaging are identified;

[0067] S300, polynomial fitting, the least square method is used to solve the cubic polynomial coefficients a, b, c, and d;

[0068] S400, extreme value calculation, the derivative equation is solved to obtain the sub-pixel level center coordinates;

[0069] S500, coordinate conversion, according to the preset "array unit coordinate-tunnel pile number-lane position" mapping table, the actual position information is output.

[0070] In addition, the lens module and the detection matrix module are arranged in one-to-one cooperation, and are arranged according to the interval of 120-150m of the highway tunnel or the interval of 80-100m of the urban tunnel, so as to ensure the accuracy of detection.

[0071] The lens module comprises a shell 110 and a micro-heating sheet 120 arranged on the outer side of the shell 110, the shell 110 comprises an auxiliary collimating mirror 170, a main lens 180, a narrow-band filter 130 and a temperature sensor 140 arranged in sequence in the shell 110, and a laser positioning calibrator 150 is further arranged on the outer side of the shell 110. The main lens 180 is made of single-crystal germanium (Ge) material, has a focal length of 300mm, an aperture of 60mm, and a field of view angle of 8°; the surface of the main lens is coated with a Ge-ZnSe-Ge three-layer dielectric antireflection film; and the auxiliary collimating mirror 170 is a front concave collimating mirror with a focal length of -50mm. The centers of the optical elements of the auxiliary collimating mirror 170, the main lens 180 and the narrow-band filter 130 are on the same straight line (i.e. the optical axis).

[0072] During the working process of the lens module, the material of the main lens 180, "single crystal germanium", is very sensitive to temperature, and its refractive index will change with temperature, resulting in focal length drift (thermal defocus), affecting the clarity of the image. In cold or humid environments, water vapor or frost will condense on the surface of the optical elements, completely blocking the infrared signal. The heating sheet raises the temperature of the shell and the internal lens, making it higher than the dew point of the environment, thereby preventing fogging / crusting and ensuring normal operation of the system at all times. Therefore, the miniature heating sheet 120 can maintain a relatively constant temperature for the module, stabilize the performance of the optical system, and reduce the performance drift caused by temperature fluctuations.

[0073] The auxiliary collimating mirror 170 is a front concave collimating mirror that can "pre-converge" the incident light from a larger field of view (8°) so that it can enter the main lens 180 at a smaller angle, thereby achieving a 8° field of view without increasing the size and complexity of the main lens. In cooperation with the main lens 180, it can effectively correct the spherical aberration, coma, and other aberrations of the entire optical system, and improve the imaging quality of the edge field of view.

[0074] The aperture of the main lens 180 is 60mm, which can ensure that the intensity of the collected target signal (infrared light) is within the preset range, and the focal length of 300mm determines that the magnification and resolution of the system meet the requirements.

[0075] Material selection: single crystal germanium - germanium is a core material in the field of infrared optics, with extremely high transmittance in the 3-12μm infrared waveband, but not transparent in the visible light waveband. This itself serves as a natural filter. Its high refractive index also allows for fewer lenses and more curved surfaces to achieve long focal length, simplifying the system structure. Coating - Ge-ZnSe-Ge three-layer dielectric antireflection coating: This is a key process to improve performance. Due to the high refractive index of germanium, about 36% of the incident light is reflected and lost. The antireflection coating greatly reduces this reflection loss through the principle of light interference cancellation, thereby significantly improving the transmittance of the entire system (from 64% to more than 95%).

[0076] The narrowband filter 130 of the present embodiment is configured to filter out light other than 4.3μm, so that stray infrared radiation such as sunlight and light in the background will be filtered out, and only the laser signal carrying information can reach the detector, thereby extracting the extremely weak effective signal in strong background noise, greatly improving the signal-to-noise ratio and detection distance of the system.

[0077] The detection matrix module of the embodiment includes a ceramic substrate 230 and a plurality of sensing units 220 uniformly arranged on the ceramic substrate 230. The ceramic substrate 230 is provided with a buffer layer 210 on one side surface of the sensing units 220, and a signal lead-out layer 240 on the side surface opposite to the sensing units 220. The detection matrix module is arranged in combination with the temperature sensor 140. The sensing unit 220 adopts a LiTaO3-Nb2O5-ZrO2 ternary solid solution sensitive layer (Nb2O5 doping amount: 5 mol%, ZrO2 doping amount: 3 mol%). The thermoelectric coefficient is greater than or equal to 500 μC / (m² K). The size of the sensing unit 220 is 0.8 mm x 0.8 mm. The voltage response rate is greater than or equal to 1.5 V / W (@4.3 μm, 1 kHz modulation frequency). The noise equivalent power (NEP) is less than or equal to 2 x 10 -11 W / √Hz.

[0078] The ceramic substrate 230 provides a firm and flat mounting base for the entire detection matrix. All the sensing units 220 are arranged in high precision and uniformity based on the ceramic substrate 230, which directly determines the spatial consistency of imaging. The ceramic substrate 230 is internally integrated with complex metal wires to connect the weak electrical signals generated by each sensing unit 220 to the signal lead-out layer 240. The high insulation of the ceramic substrate 230 ensures that the signals between thousands of sensing units 220 do not interfere with each other. The ceramic substrate 230 has excellent thermal stability and thermal conductivity. It helps to homogenize the temperature of the entire detector array and dissipate excess heat, reducing measurement errors caused by local thermal gradients.

[0079] The sensing unit 220 is used to directly convert infrared thermal radiation into a voltage signal. The selected material is a LiTaO3-Nb2O5-ZrO2 ternary solid solution sensitive layer.

[0080] LiTaO3 (lithium tantalate): It is an excellent thermoelectric material base with high thermoelectric coefficient and good physical and chemical stability.

[0081] Nb2O5 (dibismuth pentoxide) doping (5 mol%): The purpose of this doping is usually to optimize the dielectric properties and thermoelectric properties of the material, which may be used to increase the thermoelectric coefficient or adjust the Curie temperature, so that the material has optimal performance in the working temperature range.

[0082] ZrO2 (zirconium oxide) doping (3 mol%): The purpose is to enhance the mechanical strength and thermal stability of the material, and it may also help to suppress grain growth, improve the microstructure of the material, and thus improve the reliability and consistency of the device.

[0083] The detection matrix module is arranged in combination with the temperature sensor 140, so that the detection matrix module can accurately and timely measure real-time data.

[0084] Further, step S300 of the embodiment is specifically

[0085] Suppose the coordinates of the detection units are x0, x1, x2 (distributed along the X axis), and the corresponding signal values are y0, y1, y2. The cubic polynomial expression is: y = ax³ + bx² + cx + d.

[0086] Least square fitting is performed using curve_fit, and p0 is the initial parameter guess value.

[0087] p0 = [0, -1, 0, max(y)] Assuming the curve opens downward, and the vertex is close to the maximum signal value.

[0088] popt, _ = curve_fit(self.cubic_polynomial, x, y, p0=[0, -1, 0, max(y)]);

[0089] The coefficients [a, b, c, d] obtained by fitting are returned.

[0090] The step S400 is specifically

[0091] S410, fit a cubic polynomial to obtain coefficients a, b, c, d,

[0092] a, b, c, d = self.fit_polynomial(x_units, signals);

[0093] S420, solve the point where the derivative is 0 (extreme point)

[0094] First derivative: y’ = 3ax² + 2bx + c

[0095] Let y’ = 0, solve the quadratic equation: 3ax² + 2bx + c = 0

[0096] Discriminant Δ = (2b)² - 4 (3a) c

[0097] If the discriminant is less than 0, there is no real root, indicating that the curve has no obvious peak value. At this time, use the average coordinate of the covering unit as the center.

[0098] S430, calculate the extreme point and filter the valid solution (located within the unit coordinate range)

[0099] x1 = (-2 b + np.sqrt(discriminant)) / (6 a);

[0100] x2 = (-2 b - np.sqrt(discriminant)) / (6 a);

[0101] candidates = [x for x in [x1, x2] if min(x_units) <= x <= max(x_units)];

[0102] S440. Select the extreme point with the largest signal value. This extreme point corresponds to the sub-pixel center coordinates (cell level).

[0103] The coordinate transformation in step S500 is specifically as follows:

[0104] Subpixel coordinates are converted into tunnel station numbers and lane information. Subpixel coordinates include X-axis subpixel coordinates (cell level) and Y-axis subpixel coordinates (cell level). The station number offset in the X-axis direction (cell coordinate difference × cell spacing × scale coefficient) is calculated for the X-axis subpixel coordinates (cell level) and compared with the reference station number to obtain the actual station number. Lane information is calculated for the Y-axis subpixel coordinates (cell level).

[0105] Example 2

[0106] This embodiment provides a method for detecting moving fire, employing the detection system of Embodiment 1. The method is as follows:

[0107] The light source signal data that meets the conditions is collected by the lens module and the detection matrix module and sent to the data processing module. The data processing module performs the following operations: S100, signal preprocessing, the original signal is subjected to mean filtering (window size 3) to eliminate high frequency noise;

[0108] S200, peak detection, identifies 2-3 adjacent units (continuous units with signal values ​​higher than the threshold) covered by fire source imaging.

[0109] S300, Polynomial fitting, using the least squares method to solve for the coefficients a, b, c, d of the cubic polynomial;

[0110] S400, extreme value calculation, solving the derivative equation to obtain sub-pixel level center coordinates;

[0111] S500, coordinate transformation, outputs actual location information according to the preset "array unit coordinates-tunnel station number-lane position" mapping table.

[0112] Also includes according to the tunnel traffic density dynamic adjustment of the radiation intensity threshold, traffic peak set to 0.03W / cm², trough set to 0.015W / cm²; And at the same time identify multiple mobile fire and radiation intensity sorting alarm, and combined with the tunnel wind speed prediction 1 minute within the spread range.

[0113] Examples as follows:

[0114] Scenario: in 150m tunnel lane 2 has a mobile fire, imaging coverage X axis unit [16, 17, 18] and Y axis unit [31, 32, 33], the signal value as shown in the table:

[0115] direction unit coordinate signal value (V) X axis 16 0.8 X axis 17 1.5 X axis 18 1.2 Y axis 31 1.0 Y axis 32 1.6 Y axis 33 0.9

[0116] Calculation process:

[0117] X axis sub-pixel center calculation:

[0118] Fitting cubic polynomial: y = -0.1x³ + 5.1x² - 52.3x + 174.5

[0119] Derivation: y' = -0.3x² + 10.2x - 52.3

[0120] Solve equation -0.3x² + 10.2x - 52.3 = 0, get effective solution x = 17.23 (unit level)

[0121] Y axis sub-pixel center calculation:

[0122] Fitting cubic polynomial: y = -0.2x³ + 12.8x² - 258.5x + 1710.3

[0123] Derivation: y' = -0.6x² + 25.6x - 258.5

[0124] Solve equation get effective solution y = 32.15 (unit level)

[0125] Convert to tunnel location:

[0126] X axis offset: (17.23 - 32) × 0.8mm × 0.01m / mm = -0.118m (reference stake K1+200);

[0127] Actual stake: K1+200 - 0.118m = K1+199.9;

[0128] Y axis 32.15 units corresponding to lane 2;

[0129] Output results:

[0130] "subpixel_x": 17.23,

[0131] "subpixel_y": 32.15,

[0132] "stake_number": "K1+199.9",

[0133] "lane": "Second lane",

[0134] "position_error": 0.04mm (spatial error at 150m ≈ 0.3m).

[0135] In summary, this application presents an innovative solution to the severe challenges of fire detection in highway tunnels, possessing significant technical advantages and application value. Its core breakthrough lies in achieving long-range precise positioning and sub-pixel accuracy: through a unique "polynomial fitting + extreme value calculation" algorithm, the system can perform cubic curve fitting on the signals of 2-3 detection units covered by the fire source imaging, and analyze the sub-pixel-level center coordinates (e.g., X-axis unit 17.23 in the example) by differentiation. Combined with a preset coordinate mapping table, the fire source positioning accuracy can be improved to within 0.3 meters at 150 meters, accurately outputting the specific tunnel station number (e.g., K1+199.9) and lane position, completely changing the limitation of traditional detectors that can only provide "area alarms". In terms of adaptability to complex environments, the system demonstrates outstanding robustness: the lens module integrates a miniature heating element and a laser positioning calibrator, effectively overcoming temperature fluctuations and optical offsets within the tunnel; the detection matrix employs a ceramic substrate and buffer layer design, successfully suppressing the impact of 10-50Hz mechanical vibrations; and the combination of a 4.3μm narrowband filter and a high-sensitivity sensing unit system effectively eliminates interference from vehicle exhaust, lighting, etc., achieving a near-zero monthly false alarm rate. Furthermore, the system possesses multi-target processing and intelligent early warning capabilities, simultaneously identifying multiple moving fire sources and sorting alarms by radiation intensity. It can also dynamically predict the fire spread range within one minute based on tunnel wind speed. Through end-to-end optimization from optical design and material selection to signal processing, this solution not only addresses the three major pain points of traditional flame detectors—lack of positioning, weak seismic resistance, and high false alarm rate—but also provides tunnel safety with a complete set of highly reliable, intelligent, and practical fire monitoring and emergency command tools.

[0136] The above-described embodiments are merely illustrative of certain implementations of the present invention, and are described in a relatively specific and detailed manner. However, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements are all within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A mobile fire detection system, characterized in that: The system includes a lens module and a detection matrix module that work together. The lens module collects light data and sends it to the detection matrix module for photoelectric conversion. The converted signal is then sent to a data processing module. The data processing flow of the data processing module is as follows: S100, Signal preprocessing: Mean filtering is performed on the original signal to eliminate high-frequency noise; S200, peak detection, identifies 3 adjacent units covered by fire source imaging; S300, Polynomial fitting, using the least squares method to solve for the coefficients a, b, c, d of the cubic polynomial; S400, extreme value calculation, solving the derivative equation to obtain sub-pixel level center coordinates; S500, coordinate transformation, outputs actual location information according to the preset "array unit coordinates-tunnel station number-lane position" mapping table.

2. The mobile fire detection system according to claim 1, characterized in that: The specific steps S300 are as follows: Let the coordinates of the detection unit be x0, x1, x2, and the corresponding signal values ​​be y0, y1, y2. 2， The expression for a cubic polynomial is: y = ax³ + bx² + cx + d; Use curve_fit for least squares fitting, where p0 is the initial parameter guess value; p0 = [0, -1, 0, max(y)] Assume the curve opens downwards and the vertex is close to the maximum signal value; popt, _ = curve_fit(self.cubic_polynomial, x, y, p0=[0, -1, 0, max(y)]); Returns the fitted coefficients [a, b, c, d].

3. A mobile fire detection system according to claim 2, characterized in that: The specific step S400 is as follows: S410. Fit a cubic polynomial and obtain the coefficients a, b, c, d. a, b, c, d = self.fit_polynomial(x_units, signals); S420. Find the points where the derivative is 0 as the extreme points. First derivative: y' = 3ax² + 2bx + c Let y' = 0, solve the quadratic equation: 3ax² + 2bx + c = 0 The discriminant Δ = (2b)² - 4 (3a) c If the discriminant is less than 0, there are no real roots, indicating that the curve has no obvious peak. In this case, the average coordinates of the covering unit are used as the center. S430. Calculate the extreme points and filter out valid solutions; S440. Select the extreme point with the largest signal value, which corresponds to the sub-pixel center coordinates.

4. A mobile fire detection system according to claim 3, characterized in that: The coordinate transformation in step S500 is specifically as follows: Subpixel coordinates are converted into tunnel station numbers and lane information. Subpixel coordinates include X-axis subpixel coordinates and Y-axis subpixel coordinates. The station number offset in the X-axis direction is calculated for the X-axis subpixel coordinates and compared with the reference station number to obtain the actual station number. Lane information is calculated for the Y-axis subpixel coordinates.

5. A mobile fire detection system according to claim 1, characterized in that: The lens module includes a housing (110) and a miniature heating element (120) attached to the outside of the housing (110). The housing (110) includes an auxiliary collimating lens (170), a main lens (180), a narrowband filter (130), and a temperature sensor (140) arranged in sequence. A laser positioning calibrator (150) is also provided on the outside of the housing (110).

6. A mobile fire detection system according to claim 5, characterized in that: The detection matrix module includes a ceramic substrate (230) and a plurality of sensing units (220) uniformly disposed on the ceramic substrate (230). A buffer layer (210) is provided on the side surface of the ceramic substrate (230) where the sensing units (220) are located, and a signal lead-out layer (240) is provided on the side surface of the ceramic substrate (230) away from the sensing units (220). The detection matrix module is attached to the temperature sensor (140).

7. A mobile fire detection system according to claim 5, characterized in that: The main lens (180) is made of single-crystal germanium (Ge) material, with a focal length of 300mm, an aperture of 60mm, and a field of view of 8°; the surface of the main lens is coated with a Ge-ZnSe-Ge three-layer dielectric anti-reflection film; the auxiliary collimating lens (170) is a front concave collimating lens with a focal length of -50mm.

8. A mobile fire detection system according to claim 6, characterized in that: The sensing unit (220) adopts a ternary solid solution sensing layer of LiTaO3-Nb2O5-ZrO2 with a pyroelectric coefficient ≥500μC / (m²). K); The sensor unit (220) has a size of 0.8mm × 0.8mm, a voltage response rate ≥ 1.5V / W, and a noise equivalent power. .

9. A method for detecting a moving fire, characterized in that: Using the detection system according to any one of claims 1-8, the method is as follows: The lens module and the detection matrix module acquire light source signal data that meets the conditions and send it to the data processing module. The data processing module performs the following operations: S100, Signal preprocessing: Mean filtering is performed on the original signal to eliminate high-frequency noise; S200, peak detection, identifies 3 adjacent units covered by fire source imaging; S300, Polynomial fitting, using the least squares method to solve for the coefficients a, b, c, d of the cubic polynomial; S400, extreme value calculation, solving the derivative equation to obtain sub-pixel level center coordinates; S500, coordinate transformation, outputs actual location information according to the preset "array unit coordinates-tunnel station number-lane position" mapping table.

10. A method for detecting a moving fire according to claim 9, characterized in that: It also includes dynamically adjusting the radiation intensity threshold based on the tunnel traffic density, setting it to 0.03W / cm² during peak traffic hours and 0.015W / cm² during off-peak hours; it also simultaneously identifies multiple mobile fire sources and sorts them by radiation intensity to trigger alarms, and combines tunnel wind speed to predict the spread range within 1 minute.

Images