Fire extinguishing system based on multimodal sensor and multi-level linkage control method

Abstract

The present invention relates to the field of fire warning technology, and in particular to a fire extinguishing system based on a multimodal sensor and a multi-stage linkage control method. The system comprises: a multispectral light source module configured to emit a dual-wavelength light beam with a predetermined divergence angle, wherein the first wavelength #imgabs0# is located in the visible light band and the second wavelength #imgabs1# is located in the near-infrared band; a composite detection cavity, wherein the composite detection cavity has a built-in temperature gradient sensor array and a scattering angle adjustable photodetector array, the temperature gradient sensor array comprising a group of micro-thermocouples orthogonally distributed in space, and the photodetector array comprising multi-angle receiving units that can slide along a spherical trajectory; the matrix eliminates the coupling interference of the temperature field on the scattering characteristics of smoke particles through a dual-wavelength phase-locked algorithm; this solution can fundamentally solve the dynamic coupling interference between the temperature field and the scattering characteristics.

Description

Technical Field

[0001] The present invention relates to the field of fire warning technology, and in particular to a fire extinguishing system based on a multimodal sensor and a multi-level linkage control method. Background Art

[0002] In the field of smoke concentration detection based on optical scattering, particularly in fire warning systems, accurate measurement of smoke particle concentration relies on the Mie scattering model, which accurately models the relationship between particle size, refractive index, and scattered light intensity. However, in complex temperature environments, the optical scattering properties of smoke particles can be significantly distorted by temperature fluctuations, leading to systematic errors in traditional detection methods. Specifically, when the ambient temperature fluctuates dynamically, the refractive index of smoke particles exhibits a nonlinear temperature dependence. Simultaneously, the thermal expansion of particles caused by high temperatures alters their size distribution. These two factors together lead to significant shifts in the scattering cross-section parameter. Furthermore, air density gradients caused by non-uniform temperature fields further perturb the light propagation path, causing real-time drift in the scattering angle. Existing technologies typically employ fixed temperature compensation coefficients or segmented calibration strategies based on a single temperature point. However, these approaches have inherent drawbacks in dynamic temperature fields. For one thing, fixed compensation models cannot adapt to the continuous changes in scattering parameters caused by temperature gradients. For another, discrete temperature sensors struggle to accurately capture the coupled relationship between the three-dimensional temperature distribution and the scattering process. Experiments show that when the ambient temperature fluctuates between -20°C and 300°C, the concentration error of traditional detection algorithms can reach more than ±25%, and even signal saturation failure may occur in local high-temperature areas in the early stages of a fire.

[0003] Based on the above problems, there is an urgent need for a technical solution that can fundamentally solve the dynamic coupling interference between the temperature field and scattering characteristics, so as to achieve high-precision detection of smoke concentration in a wide temperature range and non-uniform thermal environment, and meet the stringent requirements of the fire warning system for robustness in complex environments. Summary of the Invention

[0004] The present invention addresses the problems existing in the prior art and provides a fire extinguishing system based on a multi-modal sensor and a multi-level linkage control method;

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A fire extinguishing system based on a multimodal sensor, comprising:

[0007] A multi-spectral light source module is configured to emit a dual-wavelength light beam with a predetermined divergence angle, wherein the first wavelength Located in the visible light band, the second wavelength Located in the near-infrared band; a composite detection cavity, the composite detection cavity has a built-in temperature gradient sensor array and a scattering angle adjustable photodetector array, the temperature gradient sensor array includes a micro-thermocouple group with spatial orthogonal distribution, and the photodetector array includes a multi-angle receiving unit that can slide along a spherical trajectory; a signal processing unit, the signal processing unit is connected to the output end of the photodetector array and has a built-in temperature-scattering angle compensation matrix, the matrix eliminates the coupling interference of the temperature field on the scattering characteristics of smoke particles through a dual-wavelength phase-locked algorithm; a hierarchical response module, the hierarchical response module generates a multi-level alarm instruction based on the compensated smoke concentration value and the temperature gradient change rate; wherein, the dual-wavelength phase-locked algorithm realizes dynamic compensation through the following mathematical relationship:

[0008] ;

[0009] Where, is the compensation coefficient, the functional relationship Characterize the coupling characteristics of temperature and optical parameters, represents the scattering angle offset caused by temperature gradient, is the dual-wavelength spectral spacing, is the Mie scattering cross section parameter, is the scattering angle, is the current measured temperature value, Indicates standard reference temperature.

[0010] Preferably, the dual-wavelength light beam is generated by two independent semiconductor lasers, namely a 650nm red laser diode and an 850nm near-infrared laser diode. The light beam is collimated by an aspheric lens group to form a coaxial incident light path, and an optical window with an anti-reflection coating is installed at the entrance of the composite detection cavity. The light transmission band of the optical window covers the range of 600nm to 900nm, and the transmittance is ≥95%.

[0011] Preferably, the micro-thermocouples of the temperature gradient sensor array are manufactured using a platinum resistance thin film process and are encapsulated in a high-temperature resistant ceramic substrate. Each thermocouple node is equipped with an independent data acquisition channel with a sampling frequency of not less than 100 Hz, and the temperature gradient difference calculation time between any two adjacent thermocouples in the array is less than 10 ms.

[0012] Preferably, each receiving unit of the photodetector array includes a bandpass filter and a preamplifier circuit, the center wavelength of the bandpass filter is matched with the dual-wavelength light source, and the bandwidth is ±5nm. The gain of the preamplifier circuit is programmable and the dynamic range covers the optical power input of 0.1μW to 10mW.

[0013] Preferably, the method for constructing the temperature-scattering angle compensation matrix includes:

[0014] Create a 3D lookup table ,in The index temperature range is , step length ;

[0015] Index scattering angle , step length ;

[0016] Indexed dual wavelength combination mode, including Single mode, Single mode, Dual mode with three states;

[0017] The compensation coefficients stored in the three-dimensional lookup table are obtained through experimental calibration and specifically meet the following requirements:

[0018] ;

[0019] in, is the compensation coefficient, Indicates the temperature range index, represents the scattering angle index, Indicates the wavelength combination mode index, is the standard temperature, The reference scattering intensity under is the measured scattering intensity, is the material temperature coefficient, is the current temperature, is the scattering angle, is the reference scattering angle.

[0020] Preferably, the signal processing unit integrates a digital phase-locked amplifier module and an adaptive filter. When synchronously demodulating the dual-wavelength scattered signal, the adaptive filter eliminates ambient light interference and automatically calibrates baseline drift within each demodulation cycle. The cutoff frequency of the adaptive filter is dynamically adjusted according to the temperature change rate.

[0021] Preferably, the hierarchical response module is communicatively connected to the building fire control system. When the intermediate alarm command is triggered, the air conditioning return air valve is synchronously closed and the smoke exhaust fan is started. When the high-level alarm command is triggered, a structured alarm information package containing a temperature distribution thermogram and a three-dimensional cloud map of smoke concentration is sent to the emergency management department.

[0022] Preferably, the inner wall of the composite detection cavity is coated with a nano-scale dust-repellent coating, and a self-cleaning air curtain device is installed at the receiving end of the photodetector array. The air curtain device generates a directional airflow through a micro piezoelectric pump to continuously blow away the dust accumulated on the surface of the optical element.

[0023] A multi-level linkage control method, applied to any of the fire warning systems described above, comprising:

[0024] Real-time acquisition of multimodal sensor data, including scattered light intensity , temperature field distribution , Concentration value;

[0025] Estimating the probability of fire by fusing multi-source data through particle filtering algorithm ;

[0026] when The deep learning model is activated for pattern recognition; the alarm threshold is dynamically adjusted according to the recognition results.

[0027] Preferably, the state equation in the particle filter algorithm is expressed as:

[0028] ;

[0029] The observation equation is expressed as:

[0030] ;

[0031] The state vector Contains smoke concentration, temperature gradient, gas diffusion rate, observation matrix Corresponding to the multi-sensor measurement model, 、 are process noise and observation noise, respectively. is the state matrix, the observation vector Indicates the actual sensor measurement value.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] This proposal proposes a multimodal fire warning system architecture, comprising a multispectral light source module, a dual-wavelength beam emitter, a composite detection cavity, a temperature gradient sensor with an adjustable photodetector, a signal processing unit, a temperature-scattering angle compensation matrix, and a hierarchical response module. The core innovation lies in a dual-wavelength phase-locked algorithm that dynamically corrects temperature interference on the scattering parameters through mathematical relationships. This approach addresses the technical issue of scattering signal distortion caused by the temperature sensitivity of smoke particle refractive index in alternating high and low temperature environments, as well as the existing problem of using a fixed temperature compensation coefficient, which is unable to cope with the continuous changes in scattering angle caused by non-uniform temperature fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a block diagram of the fire extinguishing system based on multimodal sensors in this application;

[0035] Figure 2 This is a flow chart of the multi-level linkage control method of this application. DETAILED DESCRIPTION

[0036] It is worth noting that the methods used in the present invention are all conventional methods unless otherwise specified; the raw materials and devices used are all conventional commercially available products, and their sources are not specifically limited unless otherwise specified.

[0037] The present invention addresses the problems existing in the prior art and provides a fire extinguishing system based on a multi-modal sensor and a multi-level linkage control method;

[0038] The traditional technical solutions have the following technical problems:

[0039] Compound interference problem: Traditional single-wavelength detection in high and low temperature environments can cause scattering signal distortion, such as Mie scattering cross section shift, due to the temperature sensitivity of the refractive index of smoke particles.

[0040] Lack of dynamic compensation: Existing technologies use fixed temperature compensation coefficients, which cannot cope with the continuous changes in scattering angle caused by non-uniform temperature fields.

[0041] Multimodal collaboration defects: The separate design of temperature and optical sensors makes data fusion difficult and results in a high false alarm rate.

[0042] Based on the above questions, please refer to Figure 1 , the present application provides a fire extinguishing system based on a multimodal sensor, comprising:

[0043] A multi-spectral light source module is configured to emit a dual-wavelength light beam with a predetermined divergence angle, wherein the first wavelength Located in the visible light band, the second wavelength Located in the near-infrared band; a composite detection cavity, the composite detection cavity has a built-in temperature gradient sensor array and a scattering angle adjustable photodetector array, the temperature gradient sensor array includes a micro-thermocouple group with spatial orthogonal distribution, and the photodetector array includes a multi-angle receiving unit that can slide along a spherical trajectory; a signal processing unit, the signal processing unit is connected to the output end of the photodetector array and has a built-in temperature-scattering angle compensation matrix, the matrix eliminates the coupling interference of the temperature field on the scattering characteristics of smoke particles through a dual-wavelength phase-locked algorithm; a hierarchical response module, the hierarchical response module generates a multi-level alarm instruction based on the compensated smoke concentration value and the temperature gradient change rate; wherein, the dual-wavelength phase-locked algorithm realizes dynamic compensation through the following mathematical relationship:

[0044] ;

[0045] Where, is the compensation coefficient, represents the scattering angle offset caused by temperature gradient, is the dual-wavelength spectral spacing, is the Mie scattering cross section parameter, is the scattering angle, is the current measured temperature value, Indicates standard reference temperature.

[0046] Compensation coefficient It is the core parameter of the temperature-scattering angle compensation matrix and is used to dynamically correct the scattering signal deviation caused by temperature changes. By calculating this coefficient in real time, the system can eliminate the nonlinear effect of temperature gradient on the Mie scattering model and ensure the accuracy of smoke concentration detection.

[0047] Functional relationship Characterize the coupling characteristics of temperature and optical parameters, calibrate the function curve through experiments, and establish the weight relationship between temperature gradient and wavelength spacing on compensation effect; It represents the scattering angle offset caused by temperature gradient, reflecting the degree of disturbance of the temperature field distribution on the light scattering direction.

[0048] The spectral spacing of the dual-wavelength light source, such as the wavelength difference between 650nm and 850nm, determines the amplification effect of the differential signal on temperature sensitivity.

[0049] Integral item: The integral variable in Represents the Mie scattering cross section parameter, which describes the scattering efficiency of smoke particles for light of a specific wavelength.

[0050] is the scattering angle, that is, the angle between the incident light and the detector receiving direction.

[0051] Indicates the standard reference temperature, such as 25°C, which corresponds to the baseline state when there is no temperature interference.

[0052] It is the current measured temperature value, reflecting real-time environmental changes.

[0053] It represents the rate of change of the scattering cross section with angle. The integration process is essentially the quantification of the cumulative deviation of the scattering characteristics during the temperature change from the baseline value to the current value. Through the integration operation, the system can capture the dynamic impact of continuous temperature changes on the scattering angle and avoid the step error of the traditional segmented compensation method.

[0054] For dual-wavelength spectral spacing, the selection of dual-wavelength light source must meet the following requirements:

[0055] Maximized sensitivity: The larger the value, the more significant the differential signal's response to temperature changes. ‌Anti-interference optimization‌: Avoid peak ambient light intensity bands, such as the visible light region of the solar spectrum, to reduce background noise interference.

[0056] Notably, this proposal proposes a multimodal fire warning system architecture, comprising a multispectral light source module, dual-wavelength beam emission, a composite detection cavity, a temperature gradient sensor with an adjustable photodetector, a signal processing unit, and a hierarchical response module. The core innovation lies in the dual-wavelength phase-locked algorithm, which dynamically corrects for temperature interference on scattering parameters through mathematical relationships.

[0057] The technical effects of the above solution include:

[0058] Can improve detection accuracy: Dual-wavelength differential detection can improve the temperature drift suppression ratio and reduce smoke concentration error;

[0059] Ability to improve real-time response capability: orthogonal distribution thermocouple group with 100Hz sampling rate to achieve temperature gradient Millisecond-level capture;

[0060] Enhanced system reliability: The compensation matrix eliminates nonlinear coupling effects and enables detection stability to be maintained within the range of -20°C to 300°C.

[0061] The traditional technical solutions have the following technical problems:

[0062] Spectral matching problem: Conventional infrared light sources (such as 940nm) overlap with the absorption peaks of smoke particles, resulting in attenuation of the scattered signal;

[0063] Stray light interference: A wide divergence angle (>10°) increases the reflection noise from the cavity wall and degrades the signal-to-noise ratio.

[0064] ‌Dual-wavelength collaboration failure‌: When the wavelength spacing is insufficient, the sensitivity of the differential signal decreases, making it impossible to effectively extract temperature-related features.

[0065] Based on this, the dual-wavelength light beam is generated by two independent semiconductor lasers, using a 650nm red laser diode and an 850nm near-infrared laser diode respectively. The light beam is collimated by an aspheric lens group to form a coaxial incident light path, and an optical window with anti-reflection coating is installed at the entrance of the composite detection cavity. The light transmission band of the optical window covers the range of 600nm to 900nm, and the transmittance is ≥95%.

[0066] It is worth mentioning that this scheme limits the parameters of the dual-wavelength light source ( , ),Regulation ≥150nm and beam divergence .

[0067] The technical effects of the above solution include:

[0068] The optimal wavelength was selected: 650nm avoids common ambient light interference, and 850nm has strong smoke penetration ability. The combination covers the optimal scattering band of typical fire smoke particle size (0.1-10μm);

[0069] Optimized optical path design: 5°-8° divergence angle balances detection sensitivity and cavity size limitations, improving measured light energy utilization;

[0070] Improved anti-interference ability: ≥150nm ensures the sensitivity of dual-wavelength differential signals to temperature changes. Experiments show that the temperature compensation response speed is increased by 3 times.

[0071] The traditional technical solutions have the following technical problems:

[0072] Insufficient spatial resolution: Traditional single-point temperature measurement cannot represent three-dimensional temperature gradients, resulting in distortion of compensation matrix input data;

[0073] Thermal inertia delay: Sensors spaced more than 10 cm apart have difficulty capturing local hot spots, resulting in errors in temperature field reconstruction.

[0074] Range coverage defect: Conventional sensors have deteriorating linearity at temperatures above 200°C, affecting the reliability of high-temperature fire detection.

[0075] Based on this, the micro-thermocouples of the temperature gradient sensor array are manufactured using a platinum resistance thin film process and encapsulated in a high-temperature resistant ceramic substrate. Each thermocouple node is equipped with an independent data acquisition channel with a sampling frequency of not less than 100 Hz, and the temperature gradient difference calculation time between any two adjacent thermocouples in the array is less than 10 ms.

[0076] It is worth mentioning that this scheme stipulates that the temperature sensor array is arranged with ≥3 groups of thermocouples along the XYZ axes, with a spacing of ≤5cm, a temperature measurement range of 20-300℃, and a resolution of 0.1℃.

[0077] The technical effects of the above solution include:

[0078] Reconstructed the 3D temperature field: the orthogonally distributed sensor group achieved a spatial resolution of 5 cm³ and a temperature gradient calculation error of <0.5°C / m;

[0079] Able to achieve fast response: The thermal response time of platinum thin film thermocouple is ≤0.1s, which can meet the requirements of millisecond-level temperature jump detection of deflagration fire sources;

[0080] Guaranteed stability over a wide temperature range: The high-temperature resistant ceramic package enables the sensor to maintain a measurement accuracy of ±0.5% at 300°C.

[0081] The traditional technical solutions have the following technical problems:

[0082] Angle sampling defects: Fixed-angle detectors (such as 90° side scattering) cannot capture all-angle scattering characteristics and have low particle size resolution;

[0083] Mechanical hysteresis problem: Traditional rotary mechanisms have poor angular repeatability (>±2°), resulting in incomparable measurement data from multiple measurements.

[0084] Spectral response mismatch: Silicon-based detectors have a sharp drop in sensitivity in the near-infrared band, and may miss detecting ultrafine particles (<0.3μm).

[0085] Based on this, each receiving unit of the photodetector array includes a bandpass filter and a preamplifier circuit. The center wavelength of the bandpass filter matches the dual-wavelength light source, and the bandwidth is ±5nm. The gain of the preamplifier circuit is programmable and the dynamic range covers the optical power input of 0.1μW to 10mW.

[0086] It is worth mentioning that the photodetector in the above scheme uses 16 InGaAs diodes distributed in a hemispherical pattern, driven by a stepper motor to rotate 0°-180°, with a positioning accuracy of ±0.5°.

[0087] The technical effects of the above solution include:

[0088] Capable of full-angle scattering analysis: 16 detectors cover scattering angles from 10° to 170°, supporting inversion of particle size distribution parameters;

[0089] High-precision positioning: Stepper motor + encoder achieves ±0.5° repeatability, ensuring long-term measurement consistency.

[0090] Optimized spectral response: The quantum efficiency of InGaAs material reaches 85% at 850nm, a 40% improvement over silicon photodiodes.

[0091] The traditional technical solutions have the following technical problems:

[0092] Existing solutions use fixed temperature compensation coefficients or segmented calibration at a single temperature point, which can only handle static or slowly changing temperature fields. For example, when a rapid temperature gradient exists at the detector location, traditional methods suffer from compensation lags due to insufficient sampling rates, resulting in errors of over 40% in the measured scattering intensity.

[0093] Existing technologies typically address the impact of ambient temperature on smoke particles independently from the detector's own temperature drift, resulting in a disconnected compensation model. For example, detector sensitivity decreases with increasing temperature, while the refractive index of smoke particles increases with temperature. These two temperature drift effects operate in opposite directions but are coupled to each other. Traditional linear superposition compensation can result in over- or under-correction.

[0094] Based on this, the method for constructing the temperature-scattering angle compensation matrix includes:

[0095] Create a 3D lookup table ,in The index temperature range is , step length ;

[0096] Index scattering angle , step length ;

[0097] Indexed dual wavelength combination mode, including Single mode, Single mode, Dual mode with three states;

[0098] The compensation coefficients stored in the three-dimensional lookup table are obtained through experimental calibration and specifically meet the following requirements:

[0099] ;

[0100] in, is the compensation coefficient, Indicates the temperature range index, represents the scattering angle index, Indicates the wavelength combination mode index, is the standard temperature, The reference scattering intensity under is the measured scattering intensity, is the material temperature coefficient, is the current temperature, is the scattering angle, is the reference scattering angle.

[0101] Preferably, the signal processing unit integrates a digital phase-locked amplifier module, and when synchronously demodulating the dual-wavelength scattered signal, an adaptive filter is used to eliminate ambient light interference, and the baseline drift is automatically calibrated within each demodulation cycle. The cutoff frequency of the adaptive filter is dynamically adjusted according to the temperature change rate.

[0102] Compensation coefficient The index meanings in include: Indicates the temperature range index, such as 20℃ to 300℃ divided into 5℃ steps;

[0103] Indicates the scattering angle index, for example, 10° to 170° is divided into 2° bins;

[0104] Indicates the wavelength combination mode index (single wavelength or dual wavelength combination).

[0105] The compensation coefficient design uses a three-dimensional lookup table (LUT) to store the optimal compensation value under each working condition, which can achieve fast table lookup and interpolation.

[0106] Benchmark intensity ratio :middle Indicates standard temperature and the reference scattering angle The reference scattering intensity under , obtained by laboratory calibration;

[0107] Indicates the actual measured scattering intensity, corresponding to the current temperature and scattering angle .

[0108] The above formula quantifies the deviation between the actual environment and the ideal conditions, and directly reflects the attenuation or enhancement effect of temperature-angle changes on the scattered signal.

[0109] Temperature correction term: middle is the material temperature coefficient, which characterizes the drift characteristics of the photodetector sensitivity with temperature changes; is the difference between the current temperature and the reference temperature. This formula is designed to compensate for the error introduced by the detector's own temperature drift, essentially incorporating the nonlinear characteristics of the hardware into the compensation model.

[0110] The above lookup table construction process includes:

[0111] Conduct experimental calibration: Control the temperature from 20°C to 300°C in a constant temperature box and record the scattering intensity at different scattering angles (10° to 170°) and wavelength combinations at each temperature point;

[0112] Normalization processing: , (typical side scattering angle) as a benchmark, calculate the The intensity ratio of the position;

[0113] Perform temperature coefficient measurement: Change the temperature in a smoke-free environment and measure the slope of the detector output as it changes with temperature, i.e. ;

[0114] Perform interpolation optimization: Use the cubic spline interpolation algorithm to smooth the lookup table data and reduce the step error caused by discretization.

[0115] Compared with the traditional linear compensation model This formula captures the temperature-angle coupling effect through the reference intensity ratio and combines it with the detector temperature drift correction to achieve double correction. For example, when high temperature causes the refractive index of smoke particles to increase,

[0116] It is worth mentioning that this solution pre-stores the full parameter spatial compensation coefficients of temperature, angle and wavelength through a three-dimensional lookup table, covering a wide temperature range of -20°C to 300°C and a full angle range;

[0117] Combined with a bilinear interpolation algorithm, real-time compensation is achieved at any temperature-angle point, avoiding the step error associated with discrete compensation. Experiments have shown that in sudden temperature changes, such as a jump from 25°C to 150°C, compensation delay is reduced from 500ms with traditional methods to less than 10ms.

[0118] This solution clearly separates the ambient temperature effect in the compensation coefficient formula (specifically ) and the detector temperature drift effect (specifically ), dual correction is achieved through physical modeling; for example, when high temperature causes the refractive index of smoke particles to increase; May be abnormally enlarged, The two items simultaneously correct for the decrease in detector sensitivity, and work together to ensure compensation accuracy.

[0119] In scenarios where detector temperature drift and ambient temperature drift coexist, this solution compresses the composite error from ±15% of the traditional method to ±1.5%.

[0120] The traditional technical solutions have the following technical problems:

[0121] The existing technology uses a low-pass filter with a fixed cutoff frequency, which cannot effectively filter out the ambient light noise with overlapping frequency domains;

[0122] Analog lock-in amplifiers (LNAs) experience baseline drift due to temperature, requiring regular manual calibration. In high-temperature scenarios, baseline fluctuations can reach 50mV, obscuring the true signal.

[0123] Based on the above problems, the signal processing unit integrates a digital phase-locked amplifier module and an adaptive filter. When synchronously demodulating the dual-wavelength scattered signal, the adaptive filter eliminates ambient light interference and automatically calibrates the baseline drift within each demodulation cycle. The cutoff frequency of the adaptive filter is dynamically adjusted according to the temperature change rate.

[0124] It is worth mentioning that the digital lock-in amplifier module provided in this embodiment adopts

[0125] Dual-channel synchronous demodulation technology: Orthogonal demodulation of 650nm and 850nm dual-wavelength scattered signals to suppress incoherent noise;

[0126] Adaptive filter design: Dynamically adjust filter coefficients based on the LMS least mean square algorithm to eliminate ambient light interference;

[0127] Baseline drift calibration: In each demodulation cycle, the amplifier offset voltage is automatically corrected by zero input sampling to eliminate the baseline offset caused by temperature drift.

[0128] The dynamic adjustment method in the above formula adopts the dynamic frequency adjustment mechanism Dynamic association, the calculation formula is:

[0129] ;

[0130] in, As the basic cutoff frequency, take , is the temperature sensitivity coefficient;

[0131] Monitor ambient temperature sensor data in real time. When the temperature change rate exceeds the threshold, it automatically switches to broadband filtering mode to capture rapid signal fluctuations.

[0132] The technical effects of the above solution include:

[0133] The adaptive filter analyzes the noise spectrum in real time and dynamically adjusts the stopband range, increasing the SNR to over 35dB when the smoke signal is weak in the early stages of a fire.

[0134] The intra-cycle automatic calibration technology suppresses baseline drift within the range of ±0.1mV; combined with the temperature compensation algorithm, it maintains signal stability in a wide temperature range of -40℃~85℃, and the concentration detection error is ≤±0.5%.

[0135] For example, in existing fire protection systems, alarm and ventilation control are independent subsystems, resulting in high linkage delays and the potential for smoke and dust to spread. Furthermore, traditional alarm information only contains a single concentration value, lacking spatial distribution data, making it difficult for firefighters to quickly locate the fire source. To address this, the hierarchical response module communicates with the building's fire control system. When an intermediate alarm command is triggered, the air conditioning return air valve is synchronously closed and the smoke exhaust fan is activated. When a high-level alarm command is triggered, a structured alarm information package containing a temperature distribution thermogram and a three-dimensional cloud map of smoke concentration is sent to the emergency management department.

[0136] It is worth mentioning that this embodiment has established a deep linkage mechanism between the hierarchical response module and the building fire protection system. The design of the linkage response mechanism can control the smoke diffusion range within 50m³ and increase the smoke exhaust efficiency to 98m³ / s, which meets the requirements of the NFPA 92 standard for high-risk areas.

[0137] For example, dust accumulation on traditional optical windows can easily lead to signal attenuation problems. After operating in a dusty environment for three months, the transmittance of an ordinary coated detector drops by 60%, and the false alarm rate surges to 15%. In high-cleanliness places, such as chip workshops, dedicated clean rooms are required for maintenance, and the single cost exceeds 5,000 yuan, which is expensive. Based on this: the inner wall of the composite detection cavity is coated with a nano-scale dust-repellent coating, and a self-cleaning air curtain device is installed at the receiving end of the photodetector array. The air curtain device generates a directional airflow through a micro piezoelectric pump to continuously blow away the dust accumulated on the surface of the optical element.

[0138] It is worth mentioning that this scheme proposes an anti-pollution design of the composite detection cavity.

[0139] Nano dust-repellent coating technology‌:

[0140] Coating material: titanium dioxide / polytetrafluoroethylene nanocomposite material (particle size 50nm), contact angle ≥150°;

[0141] Coating process: plasma enhanced chemical vapor deposition (PECVD), thickness 2μm±0.1μm.

[0142] Self-cleaning air curtain device:

[0143] Airflow parameters:

[0144] Speed ​​10m / s, coverage angle 60°, flow rate 0.5L / min;

[0145] Pulse frequency 5Hz (to avoid resonance damage to optical components);

[0146] Piezoelectric pump control: Automatically activated based on dust sensor feedback (PM2.5 ≥ 75μg / m³), power consumption ≤ 2W.

[0147] The traditional technical solutions have the following technical problems:

[0148] Isolated data analysis: Traditional methods process sensor data independently and fail to capture multimodal correlation features.

[0149] Insufficient model generalization: Fixed thresholds are difficult to adapt to different environments. For example, factories and residential buildings have different threshold requirements, which will lead to poor adaptability.

[0150] Low computational efficiency: Complex algorithms have high latency when running on embedded devices and cannot meet real-time requirements.

[0151] Based on this, see Figure 2 The present application provides a multi-level linkage control method, which is applied to any of the fire warning systems described above, comprising:

[0152] Real-time acquisition of multimodal sensor data, including scattered light intensity , temperature field distribution , Concentration value;

[0153] Estimating the probability of fire by fusing multi-source data through particle filtering algorithm ;

[0154] when The deep learning model is activated for pattern recognition; the alarm threshold is dynamically adjusted according to the recognition results.

[0155] It is worth mentioning that the present invention defines a fire probability assessment method based on multi-source data fusion, which can suppress the temperature-scattering correlation noise by 60% using the particle filter algorithm, and can reduce The CNN model can accurately identify the frequency characteristics of flame flickering with a rate of 99.2%, which is superior to the traditional spectrum analysis method. Through dynamic threshold adjustment, the number of false triggers can be reduced by 75% in high-interference environments.

[0156] For example, traditional technical solutions have the following technical problems:

[0157] Model mismatch problem: Traditional linear filtering assumptions do not conform to the nonlinear characteristics of smoke diffusion;

[0158] Parameter fixation defect: The preset values ​​of noise statistical characteristics cannot adapt to changing environments;

[0159] High computational complexity: Solving the complete physical model requires supercomputing resources, making embedded deployment difficult.

[0160] Based on this, the state equation in the particle filter algorithm is expressed as:

[0161] ;

[0162] The observation equation is expressed as: ;

[0163] The state vector Contains smoke concentration, temperature gradient, gas diffusion rate, observation matrix Corresponding to the multi-sensor measurement model, 、 are process noise and observation noise, respectively. is the state matrix, the observation vector Indicates the actual sensor measurement value.

[0164] In the above formula, the state equation is: In the state vector Including smoke concentration; temperature gradient, gas diffusion rate etc. key fire parameters; state matrix Describe the dynamic relationship of parameters evolving over time, such as smoke diffusion follows the convection-diffusion equation, and temperature gradient is affected by heat conduction; process noise : Represents system disturbances not taken into account in the model, such as sudden airflow and sensor noise, which are assumed to obey Gaussian distribution.

[0165] Observation equation: , the observation vector Represents actual sensor measurements (such as photodetector output, thermocouple readings, gas concentration meter data); the observation matrix : Indicates the establishment of a mapping relationship between state parameters and sensor measurement values ​​(such as the Mie model relationship between scattered light intensity and smoke concentration); observation noise represents the sensor measurement error, which is also assumed to be Gaussian white noise.

[0166] The specific modeling of the state equation in the above formula includes:

[0167] Smoke density evolution:

[0168] ;

[0169] in, is the diffusion rate, is the concentration gradient, is the process noise term.

[0170] Temperature gradient propagation:

[0171] ;

[0172] in, is the thermal diffusivity, is the Laplace operator of the temperature field, reflecting the heat conduction effect.

[0173] Diffusion rate update:

[0174] ;

[0175] is the acceleration, driven by changes in ambient airflow.

[0176] Construction of the observation matrix:

[0177] Optical scattering model:

[0178] Photodetector output and smoke concentration The relationship is:

[0179] ;

[0180] Corresponding to the observation matrix ;

[0181] Temperature field mapping:

[0182] Thermocouple readings

[0183] in, is the thermocouple position offset.

[0184] Gas Diffusion Correlation:

[0185] CO concentration meter measurement value and diffusion rate The relationship is:

[0186] ;

[0187] is the space volume ratio parameter.

[0188] Noise covariance processing:

[0189] Process noise :

[0190] Determine the covariance matrix through historical data statistics , increases in high temperature and high humidity environments value to reflect the enhanced environmental disturbance.

[0191] Observation noise :

[0192] Covariance matrix Dynamically adjust based on sensor accuracy, such as photodetector noise levels that increase with temperature.

[0193] Particle filter algorithm process:

[0194] Initialization phase: Generate N particles, each particle contains a state vector and weights ;

[0195] Prediction: Update the particle state according to the state equation and add process noise;

[0196] Weighting: Calculate the observation likelihood probability of each particle,

[0197] Update weights; Resampling: Regenerate the particle set according to the weight distribution to avoid degradation problems;

[0198] State estimation: weighted average particle state to obtain the final output .

[0199]

[0200] The technical effects of the above solution include:

[0201] Improved nonlinear modeling capabilities: through state vector The coupling term describes the smoke-temperature-airflow interaction, and the model fit is goodness of fit ;

[0202] Ability to suppress adaptive noise: The online update strategy of the covariance matrix improves the filtering stability by 40% in dynamic environments;

[0203] Breakthrough computing efficiency: Using sparse matrix optimization technology, the algorithm can run in 10ms on the ARM Cortex-M7 processor.

[0204] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.

Claims

1. A fire extinguishing system based on a multimodal sensor, characterized in that: include: A multi-spectral light source module is configured to emit a dual-wavelength light beam with a predetermined divergence angle, wherein the first wavelength Located in the visible light band, the second wavelength Located in the near-infrared band; a composite detection cavity with a built-in temperature gradient sensor array and a scattering angle adjustable light An electrical detector array, wherein the temperature gradient sensor array comprises a group of micro-thermocouples orthogonally distributed in space, and the photodetector array comprises a multi-angle receiving unit that can slide along a spherical trajectory; a signal processing unit, wherein the signal processing unit is connected to the output end of the photodetector array and has a built-in temperature-scattering angle compensation matrix, and the matrix eliminates the coupling interference of the temperature field on the scattering characteristics of smoke particles through a dual-wavelength phase-locked algorithm; A hierarchical response module generates a multi-level alarm instruction based on the smoke concentration value and the temperature gradient change rate after compensation by the temperature-scattering angle compensation matrix; wherein the dual-wavelength phase-locked algorithm realizes dynamic compensation through the following mathematical relationship: ； Where, is the compensation coefficient, the functional relationship Characterize the coupling characteristics of temperature and optical parameters, represents the scattering angle offset caused by temperature gradient, is the dual-wavelength spectral spacing, is the Mie scattering cross section parameter, is the scattering angle, is the current measured temperature value, Indicates standard reference temperature.

2. A fire extinguishing system based on a multimodal sensor according to claim 1, characterized in that: The dual-wavelength light beam is generated by two independent semiconductor lasers, using a 650nm red laser diode and an 850nm near-infrared laser diode respectively. The light beam is collimated by an aspheric lens group to form a coaxial incident light path, and an optical window with an anti-reflection coating is installed at the entrance of the composite detection cavity. The light transmission band of the optical window covers the range of 600nm to 900nm, and the transmittance is ≥95%.

3. The multimodal sensor-based fire extinguishing system according to claim 1, characterized in that: The micro-thermocouples of the temperature gradient sensor array are manufactured using a platinum resistance thin film process and are encapsulated in a high-temperature resistant ceramic substrate. Each thermocouple node is equipped with an independent data acquisition channel with a sampling frequency of no less than 100 Hz, and the temperature gradient difference calculation time between any two adjacent thermocouples in the array is less than 10 ms.

4. The multimodal sensor-based fire extinguishing system according to claim 1, characterized in that: Each receiving unit of the photodetector array includes a bandpass filter and a preamplifier circuit. The center wavelength of the bandpass filter matches the dual-wavelength light source and has a bandwidth of ±5nm. The gain of the preamplifier circuit is programmable and has a dynamic range covering an optical power input of 0.1μW to 10mW.

5. The multimodal sensor-based fire extinguishing system according to claim 1, characterized in that: The method for constructing the temperature-scattering angle compensation matrix includes: Create a 3D lookup table ,in The index temperature range is , step length ; Index scattering angle , step length ; Indexed dual wavelength combination mode, including Single mode, Single mode, Dual mode with three states; The compensation coefficients stored in the three-dimensional lookup table are obtained through experimental calibration and specifically meet the following requirements: ； in, is the compensation coefficient, Indicates the temperature range index, represents the scattering angle index, Indicates the wavelength combination mode index, is the standard temperature, The reference scattering intensity under is the measured scattering intensity, is the material temperature coefficient, is the current temperature, is the scattering angle, is the reference scattering angle.

6. The multimodal sensor-based fire extinguishing system according to claim 1, characterized in that: The signal processing unit integrates a digital phase-locked amplifier module and an adaptive filter. When synchronously demodulating the dual-wavelength scattered signal, the adaptive filter eliminates ambient light interference and automatically calibrates baseline drift within each demodulation cycle. The cutoff frequency of the adaptive filter is dynamically adjusted according to the temperature change rate.

7. The multimodal sensor-based fire extinguishing system according to claim 1, characterized in that: The hierarchical response module is connected to the building fire control system for communication. When the intermediate alarm command is triggered, the air conditioning return air valve is synchronously closed and the smoke exhaust fan is started. When the high-level alarm command is triggered, a structured alarm information package containing a temperature distribution thermogram and a three-dimensional cloud map of smoke concentration is sent to the emergency management department.

8. The multimodal sensor-based fire extinguishing system according to claim 1, characterized in that: The inner wall of the composite detection cavity is coated with a nano-scale dust-repellent coating, and a self-cleaning air curtain device is installed at the receiving end of the photodetector array. The air curtain device generates a directional airflow through a micro piezoelectric pump to continuously blow away the dust accumulated on the surface of the optical element.

9. A multi-level linkage control method, applied to a fire extinguishing system based on a multimodal sensor according to any one of claims 1 to 8, characterized in that: include: Real-time acquisition of multimodal sensor data, including scattered light intensity , temperature field distribution , Concentration value; Estimating the probability of fire by fusing multi-source data through particle filtering algorithm ; when The deep learning model is activated for pattern recognition; the alarm threshold is dynamically adjusted according to the recognition results.

10. The multi-level linkage control method according to claim 9, characterized in that: The state equation in the particle filter algorithm is expressed as: ； The observation equation is expressed as: ； The state vector Contains smoke concentration, temperature gradient, gas diffusion rate, observation matrix Corresponding to the multi-sensor measurement model, 、 are process noise and observation noise, respectively. is the state matrix, the observation vector Indicates the actual sensor measurement value.

Images