Intelligent detection early warning device

Abstract

The invention relates to the technical field of fire safety monitoring, in particular to an intelligent detection and early warning device which comprises an airflow generation module used for outputting a pneumatic modulation flow field with the flow speed periodically changing along with time to a monitoring area in response to a pneumatic modulation instruction so as to drive suspended particles in the monitoring area to do variable-speed following motion; the optical sensing module is used for constructing a grating field with a fixed space period in the monitoring area and collecting scattered light signals generated when the suspended particles penetrate through the grating field, and variable motion of the suspended particles is modulated into instantaneous frequency fluctuation in the scattered light signals through the grating field. According to the invention, particle phase lag parameters are obtained through pneumatic modulation and phase-locked demodulation, and then smoke and interference particles are accurately distinguished based on fluid inertia difference, so that the problem of high false alarm rate in a complex environment due to incapability of identifying physical attributes of particles due to single light intensity threshold detection adopted by most traditional detectors is solved.

Description

Technical Field

[0001] This invention relates to the field of fire safety monitoring technology, and in particular to an intelligent detection and early warning device. Background Technology

[0002] Photoelectric smoke detectors are currently the most mainstream front-end detection devices in automatic fire alarm systems. Their working principle is mainly based on the light scattering effect. Traditional detection devices typically contain a light-proof optical dark chamber, inside which are installed infrared or blue light emitters and photoelectric receivers. When airborne particles enter the dark chamber, the emitted light is scattered by the particles, and the scattered light is captured by the receiver and converted into an electrical signal. When the signal amplitude exceeds a preset alarm threshold, the device issues a fire alarm signal.

[0003] However, most traditional detectors use a single light intensity threshold for detection, which leads to a high false alarm rate in complex environments because they cannot identify the physical properties of particles. Summary of the Invention

[0004] To overcome the above shortcomings, this invention provides an intelligent detection and early warning device, which aims to improve the problem that traditional detectors mostly use a single light intensity threshold for detection, and because they cannot identify the physical properties of particles, they have a high false alarm rate in complex environments.

[0005] This invention provides the following technical solution: an intelligent detection and early warning device, comprising the following modules:

[0006] The airflow generation module is used to respond to the aerodynamic modulation command and output an aerodynamically modulated flow field with a flow velocity that changes periodically with time to the monitoring area, so as to drive the suspended particles in the monitoring area to perform variable speed following motion.

[0007] An optical sensing module is used to construct a grating field with a fixed spatial period in the monitoring area and collect the scattered light signal generated when the suspended particles pass through the grating field, wherein the variable speed motion of the suspended particles is modulated into instantaneous frequency fluctuations in the scattered light signal by the grating field.

[0008] The signal modulation module is used to generate the aerodynamic modulation command containing specific frequency characteristics and send it to the airflow generation module to establish the active excitation time base of the system;

[0009] The phase-locked demodulation module is used to acquire the scattered light signal, extract the instantaneous frequency change features in the scattered light signal, and perform cross-correlation demodulation operation between the instantaneous frequency change features and the aerodynamic modulation command to separate the motion response component that is synchronized with the specific frequency feature;

[0010] The dynamics determination module is used to calculate the phase lag parameter of the suspended particle relative to the aerodynamic modulation command based on the motion response component, determine the fluid inertial property of the suspended particle based on the phase lag parameter, and output a fire warning signal when the fluid inertial property meets the preset submicron aerosol characteristics.

[0011] By adopting the above technical solution, the particle phase lag parameter is obtained by pneumatic modulation and phase-locked demodulation, and then the smoke and interference particles are accurately distinguished based on the difference in fluid inertia. This improves the problem that most traditional detectors use a single light intensity threshold for detection, which results in a high false alarm rate in complex environments because they cannot identify the physical properties of particles.

[0012] Preferably, the airflow generating module includes:

[0013] Receive the pneumatic modulation command and modulate it into a pulse width modulation signal to drive the fan actuator to rotate;

[0014] By utilizing the rectifier structure located at the air outlet of the fan actuator, the turbulence generated by the fan is rectified into quasi-laminar flow through physical limiting;

[0015] A pulsating laminar flow field with a controllable fluctuation frequency that continuously outputs flow velocity over time to the monitoring area.

[0016] Preferably, the optical sensing module includes:

[0017] A collimating lens is used to calibrate the beam emitted by the laser emitter into a parallel beam;

[0018] The parallel beam is processed using a diffraction grating element to project an interference field with periodically distributed bright and dark fringes in the space of the monitoring area;

[0019] The detection field of view is locked onto the interference field by a focusing lens group, and the scattered light caused by suspended particles passing through bright and dark fringes is continuously received by a photoelectric conversion device.

[0020] The received scattered light intensity flicker is converted into a voltage fluctuation signal in the time domain, completing the mapping from mechanical motion to electrical signal.

[0021] Preferably, the signal modulation module includes:

[0022] A DC baseline component is generated in the digital domain to maintain unidirectional airflow.

[0023] A sinusoidal or cosine AC component is generated to produce periodic velocity fluctuations, and then superimposed on the DC baseline component.

[0024] The superimposed data is combined into a digital control sequence and sent to the airflow generation module as the aerodynamic modulation command.

[0025] Preferably, the signal modulation module further includes:

[0026] While generating the pneumatic modulation command, a light intensity strobe control command with a frequency higher than the specific frequency characteristic is generated simultaneously.

[0027] The light intensity flicker control command is sent to the optical sensing module to drive the light source to perform high-frequency brightness switching;

[0028] Through the synergistic effect of the aerodynamic modulation command and the optical intensity stroboscopic control command, a composite modulation field superimposed with low-frequency aerodynamic disturbance and high-frequency optical intensity stroboscopic is constructed in the monitoring area.

[0029] Preferably, the phase-locked demodulation module includes:

[0030] The scattered light signal is framed and truncated using a sliding window function.

[0031] Perform a Fourier transform operation on each frame of truncated signal to generate a time-frequency distribution matrix containing time, frequency, and energy dimensions;

[0032] Traverse each time slice of the time-frequency distribution matrix, search for the frequency point with the highest energy spectral density within the slice, and connect the highest energy frequency points of each time slice to reconstruct the instantaneous observation frequency curve;

[0033] Remove frequency abrupt noise points that exceed the preset bandwidth range from the instantaneous observed frequency curve.

[0034] Preferably, the phase-locked demodulation module further includes:

[0035] Two reference signals are generated based on the pneumatic modulation command: a sine reference signal with the same frequency and phase as the pneumatic modulation command, and a cosine reference signal with the same frequency as the pneumatic modulation command but with a phase offset of 90 degrees.

[0036] The instantaneous frequency change feature is multiplied point by point with the sine reference signal and the cosine reference signal, respectively;

[0037] The results of the multiplication are accumulated within a preset integration period, and the integral values ​​of the in-phase component and the quadrature component are output respectively.

[0038] Preferably, the dynamics determination module includes:

[0039] The arctangent operation logic is executed to calculate the arctangent value of the ratio of the integral value of the quadrature component to the integral value of the in-phase component, thereby obtaining the phase angle of the synthesized signal;

[0040] The inherent delay reference phase of the system is retrieved from the pre-memory, and the phase angle of the synthesized signal is subtracted from the inherent delay reference phase to obtain the phase lag angle, which characterizes the degree of lag of particle motion relative to the airflow driving command.

[0041] Preferably, the dynamics determination module further includes:

[0042] Retrieve the pre-stored aerosol particle size and phase hysteresis mapping table constructed based on Stokes' fluid dynamics equations;

[0043] The calculated phase lag parameter is substituted into the mapping table for matching;

[0044] When the matched phase lag parameter falls into the first numerical range characterizing the small inertia following characteristics, a determination result confirming it as smoke aerosol is generated, and a level flip signal is sent to the external alarm interface.

[0045] Preferably, the dynamics determination module further includes:

[0046] When the matched phase hysteresis parameter falls into the second numerical range characterizing the large inertial hysteresis, a determination result confirming that the particles are not fire particles is generated and the environmental pollution event is recorded.

[0047] The amplitude of the motion response component is monitored in real time. When the amplitude is lower than the system noise floor threshold, or when the vector synthesis magnitude of the integral values ​​of the in-phase component and the quadrature component is lower than the confidence threshold, a zeroing operation is performed to block the judgment process.

[0048] The present invention has the following beneficial effects:

[0049] 1. In this invention, the particle phase lag parameter is obtained by pneumatic modulation and phase-locked demodulation, and then the smoke and interference particles are accurately distinguished based on the difference in fluid inertia. This improves the problem that most traditional detectors use a single light intensity threshold for detection, which results in a high false alarm rate in complex environments because they cannot identify the physical properties of particles.

[0050] 2. In this invention, cross-correlation calculation is performed by the phase-locked demodulation module, thereby separating the signal component synchronized with the excitation from the noise. This improves the problem that traditional detection technologies mostly use direct signal amplification, which cannot filter out ambient light and circuit noise interference, resulting in poor detection capability for weak early smoke signals.

[0051] 3. In this invention, the airflow generation module outputs a periodically modulated flow field, thereby forcing particles to generate variable-speed motion carrying specific frequency characteristics. This improves the problem that most traditional detectors use passive diffusion sampling, and the response speed is slow in strong convection scenarios because the smoke entering the detection cavity is greatly affected by the ambient wind.

[0052] 4. In this invention, the mechanical motion is converted into a frequency signal by the optical sensing module using a grating field, and then detection is performed based on frequency characteristics rather than light intensity amplitude. This improves the problem that traditional photoelectric detection mostly uses analog light intensity measurement, and the sensitivity of the equipment decreases due to the decrease in lens transmittance caused by dust contamination. Attached Figure Description

[0053] Figure 1 This is an architectural diagram of an intelligent detection and early warning device proposed in this invention;

[0054] Figure 2 This is a flowchart of a detection method for an intelligent detection and early warning device proposed in this invention;

[0055] Figure 3 This is the core decision logic diagram of an intelligent detection and early warning device proposed in this invention. Detailed Implementation

[0056] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] Example 1: In the first embodiment of the present invention, the present invention provides an intelligent detection and early warning device, such as... Figures 1-3 As shown, it includes the following modules:

[0058] The airflow generation module is used to respond to aerodynamic modulation commands and output an aerodynamically modulated flow field with a flow velocity that changes periodically over time to the monitoring area, so as to drive the suspended particles in the monitoring area to perform variable speed following motion.

[0059] Furthermore, the airflow generation module includes:

[0060] Receives pneumatic modulation commands and modulates them into pulse width modulation signals to drive the fan actuator to rotate;

[0061] By using a rectifier structure installed at the air outlet of the fan actuator, the turbulence generated by the fan is regulated into quasi-laminar flow through physical limiting;

[0062] A pulsating laminar flow field with a controllable fluctuation frequency that continuously outputs flow velocity over time to the monitoring area.

[0063] Specifically, the airflow generation module, as the active excitation source of this device, has the core logic of establishing a precise mapping between electrical signals and fluid mechanical motion, projecting the frequency characteristics of the digital domain into the flow field of the physical domain. The operation of this module includes two processes: signal modulation driving and flow field physical shaping.

[0064] In the drive control and signal conversion process, the airflow generation module receives a pneumatic modulation command from the signal modulation module. This command is not a simple switching quantity, but a waveform sequence containing time-varying characteristics. The drive circuit parses this command and generates a corresponding pulse width modulation signal, i.e., a PWM signal, to control the instantaneous speed of the fan actuator.

[0065] In order to establish the expected pulsating flow field in the monitoring area, the instantaneous wind speed output by the fan is... It must strictly follow the changes in the modulation command. If the objective function of the aerodynamic modulation command is set to a sinusoidal wave form, then the theoretical wind speed model at the fan outlet is expressed as:

[0066] ;

[0067] in Characterization The instantaneous air velocity at the fan outlet; Characterizes the DC baseline wind speed, which is used to establish the basic airflow capable of transporting suspended particles to the detection area, ensuring that the flow field has unidirectional fluidity; The amplitude of the AC modulation represents the maximum deviation of the flow velocity fluctuation and determines the intensity of the inertial disturbance applied to the suspended particles. The aerodynamic modulation frequency, i.e. the system's carrier frequency, determines how fast the flow velocity changes.

[0068] The drive circuit is based on the target wind speed Adjusting the duty cycle of the PWM signal in real time Assuming that the fan speed and duty cycle have a linear relationship within the operating range, and that the wind speed is proportional to the speed, then the logic for generating the control signal follows the following relationship:

[0069] ;

[0070] in Characterization The PWM duty cycle applied to the fan motor at all times, with a value ranging from 0 to 1; The electromechanical conversion coefficient characterizes the motor torque constant, impeller aerodynamic efficiency, and drive circuit gain.

[0071] Through the above control logic, the module achieves the setting of a specific frequency. and amplitude The air medium is loaded, so that the output airflow carries a time fingerprint that can be demodulated by subsequent circuits.

[0072] Flow field shaping and physical constraint are crucial. The original airflow generated by the rotation of the fan impeller is usually accompanied by strong tangential velocity and vortices, which constitutes turbulence. If directly output to the monitoring area, random turbulent noise will overwhelm the modulation signal. Therefore, it is necessary to install a rectification structure at the fan outlet.

[0073] The rectifying structure employs honeycomb or grid-like physical confinement channels, with the axial length of the channels exceeding their radial diameter. When turbulent airflow passes through the rectifying structure, the radial velocity vector in the airflow is blocked and dissipated by the physical tube walls, leaving only the axial velocity vector. After rectification, the airflow streamlines are parallel to each other, the Reynolds number decreases, and it transforms into quasi-laminar flow.

[0074] This quasi-laminar flow field offers two major technical advantages: first, it reduces background airflow noise, improving the signal-to-noise ratio of subsequent photoelectric signals; second, it ensures consistent flow velocity at different locations within the monitoring area, ensuring that the Stokes drag force experienced by suspended particles in the flow field is only dependent on time. This is related to spatial location, but decoupled from it, ensuring the accuracy of dynamic parameter calculations.

[0075] In summary, the airflow generation module, through the synergy of electronic modulation and physical shaping, continuously outputs a pulsating laminar flow field with periodic and precise fluctuations in flow velocity over time to the monitoring area, forcing suspended particles in the area into a controlled variable-speed motion state, thus providing the necessary physical prerequisite for subsequent optical frequency shift detection.

[0076] The optical sensing module is used to construct a grating field with a fixed spatial period in the monitoring area and collect the scattered light signal generated when suspended particles pass through the grating field. The variable speed motion of the suspended particles is modulated into the instantaneous frequency fluctuation in the scattered light signal by the grating field.

[0077] Furthermore, the optical sensing module includes:

[0078] A collimating lens is used to calibrate the beam emitted by the laser emitter into a parallel beam;

[0079] Parallel beams are processed using diffraction grating elements to project an interference field with periodically distributed bright and dark fringes into the space of the monitoring area.

[0080] By using a focusing lens group to lock the detection field of view onto the interference field, the scattered light caused by suspended particles passing through bright and dark fringes is continuously received through a photoelectric conversion device.

[0081] The received scattered light intensity flicker is converted into a voltage fluctuation signal in the time domain, completing the mapping from mechanical motion to electrical signal.

[0082] Specifically, the optical sensing module constructs a physical conversion interface based on spatial light field modulation. Its core function is to utilize the interference properties of light to establish a micrometer-level spatial scale, linearly mapping the mechanical velocity of suspended particles to the instantaneous frequency of photoelectric signals. The operation of this module involves three physical processes: light field construction, particle scattering modulation, and photoelectric conversion.

[0083] In the construction and calibration of the spatial grating field, a laser emitter serves as a coherent light source, emitting a high-brightness divergent beam. Since the divergent light cannot directly form uniform interference fringes, a collimating lens is placed in front of the light source. After refraction by the collimating lens, the wavefront of the beam transforms from a spherical wave into a plane wave, forming a parallel beam. This parallel beam is incident on a diffraction grating element, where the light waves undergo diffraction and superposition under the influence of the grating structure, forming a periodically distributed interference field of alternating bright and dark areas in the monitoring area.

[0084] This interference field is the virtual grating, and its core physical parameter is the spatial period of the interference fringes, which is the distance between the centers of two adjacent bright or dark fringes. This spatial period determines the detection resolution of the system. We set this spatial period to a constant. Its value depends on the laser wavelength and the angle between the two interfering beams, and is a fixed physical quantity when the system leaves the factory.

[0085] The motion-frequency modulation mechanism is used when suspended particles are present in the monitoring area, and the particles are driven by the airflow generation module at an instantaneous velocity. When passing through the aforementioned interference field, the particles will successively pass through the bright fringe region with high light intensity and the dark fringe region with low light intensity.

[0086] When a particle is located in a bright stripe, the scattered light intensity reaches its peak; when a particle is located in a dark stripe, the scattered light intensity drops to its trough. The continuous passage of particles through the stripe array causes the originally constant scattered light intensity to exhibit periodic fluctuations in intensity over time. At this point, the mechanical velocity of the suspended particles is modulated into the fluctuation frequency of the scattered light signal. A photoelectric conversion device collects this scattered light through a focusing lens group and converts it into a voltage signal.

[0087] Based on the spatiotemporal correspondence, the instantaneous frequency of the output voltage signal With particle velocity Follows the following linear mathematical model:

[0088] ;

[0089] in, Characterization The instantaneous fluctuation frequency of the scattered light signal at any given moment is the fundamental frequency of the output voltage signal of the photodetector. Characterization The velocity component of the suspended particle perpendicular to the direction of the interference fringes at any given moment; The spatial period of bright and dark fringes in an interference field is characterized by the grating constant or fringe spacing.

[0090] The data input / output flow is as follows: the physical input of this module is the spatial motion vector of the suspended particles. When a particle tangentially crosses the grating field under the drive of airflow, each crossing action completes one physical sampling.

[0091] The physical output of this module is an analog voltage fluctuation signal. This signal preserves the complete dynamic characteristics of the particle's transit process; its waveform envelope reflects the particle's size and scattering cross-section, while the zero-crossing density or periodicity of the waveform directly corresponds to the formula above. Photoelectric conversion devices convert a weak stream of photons into an electric current, which is then converted into a voltage by a transimpedance amplifier circuit. Then, it is transmitted to the subsequent phase-locked demodulation module.

[0092] Through the above setup, the optical sensing module completes the lossless conversion from the invisible fluid velocity field to the processable electrical signal spectrum, providing the necessary carrier foundation for subsequent extraction of weak signals using the lock-in amplification principle.

[0093] The signal modulation module is used to generate aerodynamic modulation commands containing specific frequency characteristics and send them to the airflow generation module to establish the active excitation time base of the system.

[0094] Furthermore, the signal modulation module includes:

[0095] A DC baseline component is generated in the digital domain to maintain unidirectional airflow.

[0096] Generate a sinusoidal or cosine AC component to produce periodic velocity fluctuations, and superimpose it onto the DC baseline component;

[0097] The superimposed data is combined into a digital control sequence and sent as a pneumatic modulation command to the airflow generation module.

[0098] The signal modulation module also includes:

[0099] While generating the pneumatic modulation command, a light intensity strobe control command with a frequency higher than a specific frequency characteristic is simultaneously generated.

[0100] The light intensity flicker control command is sent to the optical sensing module to drive the light source to perform high-frequency brightness switching;

[0101] By combining aerodynamic modulation commands and optical intensity stroboscopic control commands, a composite modulation field superimposed with low-frequency aerodynamic disturbances and high-frequency optical intensity stroboscopic is constructed in the monitoring area.

[0102] Specifically, the signal modulation module, as the core of the system's timing control, is responsible for constructing and outputting excitation source signals with precise time-domain characteristics. This module establishes the active excitation time base of the entire detection system by synthesizing multi-dimensional control commands in the digital domain, achieving dual synchronous modulation of the air and light fields. Its operational logic includes three stages: aerodynamic composite waveform construction, light intensity stroboscopic generation, and coordinated output.

[0103] In the aerodynamic composite waveform construction and digital synthesis, the internal processor first establishes a digital model of the aerodynamic modulation command. To ensure that the airflow can stably transport the measured particles to the detection area and generate identifiable motion fluctuations, the control command is a linear superposition of DC and AC components.

[0104] The DC baseline component is used to set the fan's base speed, ensuring that the airflow always flows in one direction and preventing stagnation or reversal. The AC component is used to superimpose periodic variations onto the base wind speed, giving the airflow a specific frequency fingerprint. The processor generates an aerodynamic control sequence in the discrete time domain according to a preset sampling rate. Its mathematical expression is as follows:

[0105] ;

[0106] in Characterization The digital control quantity is constantly output to the airflow generation module, and this value is directly mapped to the duty cycle or voltage value of the fan drive circuit. The DC baseline coefficient is characterized by its magnitude, which determines the average wind speed in the monitoring area and ensures that laminar transport can be maintained by overcoming pipeline resistance. The AC modulation coefficient characterizes the magnitude of wind speed fluctuations, which determines the intensity of inertial perturbation applied to suspended particles. The fundamental frequency characterizing the aerodynamic modulation is a low-frequency signal, and this frequency is the main reference frequency for subsequent phase-locked demodulation. Characterizes the initial phase, used for phase alignment during system calibration.

[0107] The calculated The sequence is sent to the airflow generation module via a digital-to-analog converter interface or a PWM generator to drive the fan to generate airflow. Centered on, with It is a pulsating flow field with fluctuating frequency.

[0108] The module employs optical intensity strobe control and dual modulation, simultaneously generating optical intensity strobe control commands within the same clock cycle as the aerodynamic commands. To shift the effective signal to a higher frequency band in the spectrum to avoid low-frequency ambient light noise, the module generates a high-frequency carrier sequence.

[0109] This control command Typically a square wave pulse sequence, it is used to drive the light source in the optical sensing module to perform high-speed switching. Its mathematical logic is described as follows:

[0110] ;

[0111] in Characterization The drive level is constantly output to the optical sensing module; The intensity of the driving current when the light source is turned on determines the maximum light intensity of the detected light field; The rectangular gate function represents a single lighting action of the light source; The period characterizing high-frequency flicker is given by its reciprocal, which is the flicker frequency. And set Much greater than the aerodynamic frequency ; The pulse width, which characterizes a single flicker, determines the duty cycle of the light source.

[0112] Synergistic effect and composite field construction, the signal modulation module synchronously outputs the above through the bus. and In physical space, these two commands act on the air medium and the photonic medium respectively, constructing a composite modulation field in the monitoring area.

[0113] In this combined field, the trajectory of the suspended particles is affected by... Control, while particle scattering of light sampling is affected Control. For subsequent signal processing, this synergistic effect modulates the original baseband signal to a frequency range that is controlled by the baseband signal. As carrier, with In the sideband frequency band, the module's output is not just two electrical signals, but rather establishes a physical intermodulation mechanism based on low-frequency motion and high-frequency sampling for the entire system, enabling weak submicron particle signals to be completely separated from background noise in the frequency domain.

[0114] The phase-locked demodulation module is used to acquire the scattered light signal, extract the instantaneous frequency change features in the scattered light signal, and perform cross-correlation demodulation operation between the instantaneous frequency change features and the aerodynamic modulation command to separate the motion response component that is synchronized with the specific frequency feature.

[0115] Furthermore, the phase-locked demodulation module includes:

[0116] The scattered light signal is framed and truncated using a sliding window function.

[0117] Perform a Fourier transform operation on each frame of truncated signal to generate a time-frequency distribution matrix containing time, frequency, and energy dimensions;

[0118] Traverse each time slice of the time-frequency distribution matrix, search for the frequency point with the highest energy spectral density within the slice, and connect the highest energy frequency points of each time slice to reconstruct the instantaneous observation frequency curve.

[0119] Remove frequency abrupt noise points in the instantaneous observed frequency curve that exceed the preset bandwidth range.

[0120] The phase-locked demodulation module also includes:

[0121] Two reference signals are generated based on the aerodynamic modulation command: a sine reference signal with the same frequency and phase as the aerodynamic modulation command, and a cosine reference signal with the same frequency as the aerodynamic modulation command but with a phase offset of 90 degrees.

[0122] The instantaneous frequency change characteristics are multiplied point by point with the sine reference signal and the cosine reference signal, respectively;

[0123] The results of the multiplication are accumulated within a preset integration period, and the integral values ​​of the in-phase component and the quadrature component are output respectively.

[0124] Specifically, the phase-locked demodulation module, as the core computing unit of the system, performs the task of converting time-domain voltage signals into frequency-domain motion parameters. This module mainly includes two logical levels: instantaneous frequency extraction and orthogonal vector demodulation. It uses digital signal processing algorithms to separate the motion components that are only related to active aerodynamic excitation in a strong noise background.

[0125] The process of extracting instantaneous frequency features, and the scattered light voltage signal output by the optical sensing module. This is a non-stationary signal, and its frequency changes dynamically with time. The module first performs analog-to-digital conversion on the analog signal to obtain a discrete sequence. Using a sliding window function to Frame truncation is performed. The window function is introduced to suppress spectral leakage and ensure that the signal is approximately stationary within each frame.

[0126] For each truncated signal sequence, the module performs a Discrete Fourier Transform (DFT) to construct a time-frequency distribution matrix. This calculation process follows the mathematical model below:

[0127] ;

[0128] in Characterizing the first Frame in Complex spectral coefficients at each frequency point; Characterizing the discrete scattered optical voltage signal sequence of the input; The characterization length is A sequence of sliding window functions, such as the Hamming window or the Hanning window; The sliding step size determines the resolution on the time axis; The number of FFT points representing a single frame determines the resolution on the frequency axis; and These are the time frame index and the frequency index, respectively.

[0129] Obtain the time-frequency distribution matrix Then, the module iterates through each time index. Search for the energy spectral density in the corresponding spectral slice. Frequency index corresponding to the maximum value . Each frame Converted to physical frequencies and connected in chronological order, the instantaneous observed frequency curve is reconstructed. During this process, the module sets a bandwidth threshold logic. If the frequency jump amplitude calculated at a certain moment exceeds the physical limit of airflow modulation, it is determined to be electromagnetic interference noise and is removed. The curve is then completed using an interpolation algorithm.

[0130] Positive interactive demodulation process, extracting instantaneous observation frequency curves It contains two parts of information: one part is the regular fluctuation driven by airflow, and the other part is the random fluctuation caused by Brownian motion or ambient wind. In order to extract the former, the module introduces the lock-in amplification principle.

[0131] The module reads the digital control sequence generated by the signal modulation module and parses out the fundamental frequency of the aerodynamic modulation. Based on this frequency, the module internally generates two orthogonal reference signals: one is an in-phase reference signal. One path is a sine wave; the other is a quadrature reference signal. That is, a cosine wave.

[0132] The module will observe the discrete sequence of the instantaneous frequency curve. Each of the two reference signals is multiplied point by point, and the result is integrated within a preset integration period. The summation is performed within the range. This process implements narrowband bandpass filtering, and its core calculation formula is as follows:

[0133] ;

[0134] ;

[0135] in The integral value of the in-phase component represents the projection intensity of particle motion and airflow drive in the phase-consistent direction; The integral value of the orthogonal component represents the projection intensity of particle motion and airflow drive in the phase-perpendicular direction; It represents the number of sampling points within the integration period, and is numerically equal to the integration time multiplied by the sampling rate; The digital sampling rate characterizes the system; The fundamental frequency characterizing the aerodynamic modulation command.

[0136] Through the above calculations, any frequency is not equal to The noise component approaches zero over a long period of integration due to the cancellation of positive and negative signals, retaining only the motion response component that is strictly synchronized with the aerodynamic excitation. The module ultimately outputs a pair of scalar values. In the dynamics determination module, this pair of values ​​not only contains motion amplitude information, but also implicitly contains phase information reflecting fluid inertia.

[0137] The dynamics determination module is used to calculate the phase lag parameter of the suspended particles relative to the aerodynamic modulation command based on the motion response component, determine the fluid inertial properties of the suspended particles based on the phase lag parameter, and output a fire warning signal when the fluid inertial properties meet the preset submicron aerosol characteristics.

[0138] Furthermore, the dynamics determination module includes:

[0139] The arctangent operation logic is executed to calculate the arctangent value of the ratio of the integral value of the quadrature component to the integral value of the in-phase component, thereby obtaining the phase angle of the synthesized signal;

[0140] The inherent delay reference phase of the system is retrieved from the pre-memory, and the phase angle of the synthesized signal is subtracted from the inherent delay reference phase to obtain the phase lag angle, which characterizes the degree of lag of particle motion relative to the airflow driving command.

[0141] The dynamics determination module also includes:

[0142] Retrieve the pre-stored aerosol particle size and phase hysteresis mapping table constructed based on Stokes' fluid dynamics equations;

[0143] Substitute the calculated phase lag parameters into the mapping table for matching;

[0144] When the matched phase lag parameter falls into the first numerical range characterizing the small inertia following characteristics, a determination result confirming it as smoke aerosol is generated, and a level flip signal is sent to the external alarm interface.

[0145] The dynamics determination module further includes:

[0146] When the matched phase hysteresis parameter falls into the second numerical range characterizing the large inertial hysteresis, a determination result confirming that the particles are not fire particles is generated and the environmental pollution event is recorded.

[0147] The amplitude of the motion response components is monitored in real time. When the amplitude is lower than the system noise floor threshold, or when the vector synthesis magnitude of the integral values ​​of the in-phase component and the quadrature component is lower than the confidence threshold, a zeroing operation is performed to block the decision process.

[0148] Specifically, the dynamics determination module is the intelligent decision-making terminal of this device. Its fundamental task is to reverse-analyze the mathematical statistics output by the phase-locked demodulation module—the integral values ​​of the in-phase component and the quadrature component—into the physical properties of suspended particles. Based on the principles of aerosol dynamics, this module distinguishes the ability of particles to follow airflow through phase analysis, thereby achieving accurate classification of smoke, dust, and background noise. The module's operation encompasses three steps: phase vector analysis, inertial inversion matching, and multi-level logical decision-making.

[0149] Phase vector analysis and hysteresis calculation: The module first receives the input orthogonal vector pairs. and These two values ​​uniquely determine a signal vector in the complex plane. The argument of this vector contains information about the particle's motion lag, and the magnitude contains information about the particle's signal strength.

[0150] Perform coordinate transformation operations to convert the Cartesian coordinate system to a Cartesian coordinate system. Amplitude converted to polar coordinates and phase angle The phase angle is calculated using the four-quadrant arctangent function to ensure the uniqueness of the angle throughout the entire circumference. The mathematical expression is as follows:

[0151] ;

[0152] ;

[0153] in The observed phase angle characterizing the synthesized signal; The integral value representing the orthogonal component originates from the phase-locked demodulation module; The integral value representing the in-phase component originates from the phase-locked demodulation module; The vector synthesis magnitude, which characterizes the motion response components, represents the effective strength of the signal.

[0154] Due to inherent time delays in circuit transmission, photoelectric response, and mechanical transmission, the observed phase angle... This is not purely caused by particle inertia. Therefore, the module needs to retrieve the pre-stored system-inherent delay reference phase. Correction is performed. This reference phase is the system zero point obtained through calibration in a vacuum or pure air environment. The true phase hysteresis angle... The calculation is as follows:

[0155] ;

[0156] in The phase lag angle, which is characterized solely by the fluid inertia of the particle, is a core physical quantity for determining particle properties. The inherent delay reference phase constant of the system.

[0157] Stokes inertial inversion and table lookup matching are used to obtain the phase lag angle. Afterward, the module enters the physical property inversion stage. According to Stokes fluid dynamics theory, the motion behavior of suspended particles in an alternating flow field depends on their relaxation time. The longer the relaxation time, the greater the particle inertia, and the slower the response to changes in airflow velocity, which manifests as a phase lag angle. The larger.

[0158] The physical mapping relationship between the two follows the following dynamic equation:

[0159] ;

[0160] in Characterizes the aerodynamic modulation frequency, i.e., the excitation frequency of the system; The relaxation time of a particle is proportional to the square of its aerodynamic diameter.

[0161] Based on the above principles, the module pre-stores a table showing the mapping relationship between aerosol particle size and phase hysteresis. This table divides continuous phase hysteresis angles into different characteristic intervals. The module will then calculate... Substitute the values ​​into the table for retrieval and matching.

[0162] Multi-level logical decision and output: The module executes hierarchical decision logic based on the search results.

[0163] Level 1: Validity threshold determination; module real-time monitoring of vector synthesis modulus. .like Below the preset system noise floor threshold This indicates that there are no particles or the particle concentration is extremely low in the current monitoring area, and the signal is mainly composed of circuit thermal noise. At this time, the module forces a zeroing operation to block the subsequent judgment process and prevent noise-triggered false alarms.

[0164] Level 2: Fire smoke detection; if the signal is valid, and If the particle falls within the first numerical range, such as 0 to 15 degrees, this range corresponds to an extremely short relaxation time, indicating that the particle has extremely strong flow characteristics, consistent with the dynamic characteristics of submicron-sized smoke particles. Based on this, the module generates a determination result confirming the particle as smoke aerosol and drives the alarm interface circuit to switch levels, outputting a fire warning signal to the outside world.

[0165] Level 3: Non-fire interference exclusion, if If the particle falls into the second numerical range, such as 30 to 60 degrees, this range corresponds to a longer relaxation time, indicating that the particle has greater inertia and cannot keep up with changes in airflow frequency. This is consistent with the dynamic characteristics of large-diameter dust, water mist, or flying insects. Based on this, the module determines it as an environmental disturbance and does not trigger a fire alarm, but records an environmental pollution event in the internal log, thus achieving anti-interference filtering of non-fire particles.

[0166] Example 2: The application scenario focuses on complex industrial environments such as urban rail transit tunnels and underground utility tunnels. These scenarios are characterized by strong background airflow (such as train piston wind or forced ventilation), high concentrations of non-fire suspended particles (such as metal dust generated by braking and condensed water mist), and complex electromagnetic and optical interference. In such high-dynamic, high-interference scenarios, existing photoelectric detection technologies face severe challenges: traditional detection devices often rely on a single light scattering intensity threshold or static DC sampling logic, lacking the ability to analyze the hydrodynamic properties of suspended particles (such as inertia and relaxation time). This makes it impossible to effectively distinguish between submicron-level smoke generated by fire and micron-level dust inherent in the environment at the physical level, easily leading to high-frequency false alarms when dust accumulates or water vapor permeates the air. Simultaneously, traditional passive detection methods have extremely low signal-to-noise ratios under strong convective wind fields. Weak early smoldering signals are often masked by environmental wind noise or diluted by airflow, making it difficult for the detector to extract effective features from the background noise, resulting in missed detections or delayed responses to initial fires. To solve these problems, this invention provides an intelligent detection and early warning device, the structure of which is as follows... Figure 1 As shown. The specific implementation process of this device is as follows:

[0167] The active detection of the hydrodynamic properties of suspended particles is achieved through the collaborative operation of various modules. The signal modulation module generates specific frequency commands to establish the system excitation time base, and controls the airflow generation module to output a periodic pulsating flow field, forcing suspended particles to undergo variable-speed following motion, thus assigning a specific frequency fingerprint to the measured object. The optical sensing module uses a spatial grating field as a physical scale to linearly map the mechanical velocity of particles passing through the stripes into the instantaneous frequency fluctuation of the scattered light signal, completing the frequency modulation of the signal. The phase-locked demodulation module performs cross-correlation demodulation operations, comparing the received scattered light signal with the source excitation command, and using the coherent detection principle to filter out ambient light and background wind noise inconsistent with the modulation frequency, extracting the synchronous motion response component. The dynamics determination module, based on fluid dynamics principles, calculates the phase lag parameter according to the motion response component. Utilizing the significant difference in fluid inertia between submicron-sized smoke and large-diameter interference particles, this module accurately inverts particle properties by the magnitude of the phase lag angle, distinguishing fire smoke from non-fire particles from a physical source, overcoming the susceptibility to interference in traditional single-intensity threshold detection techniques.

[0168] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent detection and early warning device, characterized in that, Includes the following modules: The airflow generation module is used to respond to the aerodynamic modulation command and output an aerodynamically modulated flow field with a flow velocity that changes periodically with time to the monitoring area, so as to drive the suspended particles in the monitoring area to perform variable speed following motion. An optical sensing module is used to construct a grating field with a fixed spatial period in the monitoring area and collect the scattered light signal generated when the suspended particles pass through the grating field, wherein the variable speed motion of the suspended particles is modulated into instantaneous frequency fluctuations in the scattered light signal by the grating field. The signal modulation module is used to generate the aerodynamic modulation command containing specific frequency characteristics and send it to the airflow generation module to establish the active excitation time base of the system; The phase-locked demodulation module is used to acquire the scattered light signal, extract the instantaneous frequency change features in the scattered light signal, and perform cross-correlation demodulation operation between the instantaneous frequency change features and the aerodynamic modulation command to separate the motion response component that is synchronized with the specific frequency feature; The dynamics determination module is used to calculate the phase lag parameter of the suspended particle relative to the aerodynamic modulation command based on the motion response component, determine the fluid inertial property of the suspended particle based on the phase lag parameter, and output a fire warning signal when the fluid inertial property meets the preset submicron aerosol characteristics.

2. The intelligent detection and early warning device according to claim 1, characterized in that, The airflow generating module includes: Receive the pneumatic modulation command and modulate it into a pulse width modulation signal to drive the fan actuator to rotate; By utilizing the rectifier structure located at the air outlet of the fan actuator, the turbulence generated by the fan is rectified into quasi-laminar flow through physical limiting; A pulsating laminar flow field with a controllable fluctuation frequency that continuously outputs flow velocity over time to the monitoring area.

3. The intelligent detection and early warning device according to claim 1, characterized in that, The optical sensing module includes: A collimating lens is used to calibrate the beam emitted by the laser emitter into a parallel beam; The parallel beam is processed using a diffraction grating element to project an interference field with periodically distributed bright and dark fringes in the space of the monitoring area; The detection field of view is locked onto the interference field by a focusing lens group, and the scattered light caused by suspended particles passing through bright and dark fringes is continuously received by a photoelectric conversion device. The received scattered light intensity flicker is converted into a voltage fluctuation signal in the time domain, completing the mapping from mechanical motion to electrical signal.

4. The intelligent detection and early warning device according to claim 1, characterized in that, The signal modulation module includes: A DC baseline component is generated in the digital domain to maintain unidirectional airflow. A sinusoidal or cosine AC component is generated to produce periodic velocity fluctuations, and then superimposed on the DC baseline component. The superimposed data is combined into a digital control sequence and sent to the airflow generation module as the aerodynamic modulation command.

5. The intelligent detection and early warning device according to claim 1, characterized in that, The signal modulation module further includes: While generating the pneumatic modulation command, a light intensity strobe control command with a frequency higher than the specific frequency characteristic is generated simultaneously. The light intensity flicker control command is sent to the optical sensing module to drive the light source to perform high-frequency brightness switching; Through the synergistic effect of the aerodynamic modulation command and the optical intensity stroboscopic control command, a composite modulation field superimposed with low-frequency aerodynamic disturbance and high-frequency optical intensity stroboscopic is constructed in the monitoring area.

6. The intelligent detection and early warning device according to claim 1, characterized in that, The phase-locked demodulation module includes: The scattered light signal is framed and truncated using a sliding window function. Perform a Fourier transform operation on each frame of truncated signal to generate a time-frequency distribution matrix containing time, frequency, and energy dimensions; Traverse each time slice of the time-frequency distribution matrix, search for the frequency point with the highest energy spectral density within the slice, and connect the highest energy frequency points of each time slice to reconstruct the instantaneous observation frequency curve; Remove frequency abrupt noise points that exceed the preset bandwidth range from the instantaneous observed frequency curve.

7. The intelligent detection and early warning device according to claim 1, characterized in that, The phase-locked demodulation module further includes: Two reference signals are generated based on the pneumatic modulation command: a sine reference signal with the same frequency and phase as the pneumatic modulation command, and a cosine reference signal with the same frequency as the pneumatic modulation command but with a phase offset of 90 degrees. The instantaneous frequency change feature is multiplied point by point with the sine reference signal and the cosine reference signal, respectively; The results of the multiplication are accumulated within a preset integration period, and the integral values ​​of the in-phase component and the quadrature component are output respectively.

8. The intelligent detection and early warning device according to claim 7, characterized in that, The dynamics determination module includes: The arctangent operation logic is executed to calculate the arctangent value of the ratio of the integral value of the quadrature component to the integral value of the in-phase component, thereby obtaining the phase angle of the synthesized signal; The inherent delay reference phase of the system is retrieved from the pre-memory, and the phase angle of the synthesized signal is subtracted from the inherent delay reference phase to obtain the phase lag angle, which characterizes the degree of lag of particle motion relative to the airflow driving command.

9. The intelligent detection and early warning device according to claim 8, characterized in that, The dynamics determination module also includes: Retrieve the pre-stored aerosol particle size and phase hysteresis mapping table constructed based on Stokes' fluid dynamics equations; The calculated phase lag parameter is substituted into the mapping table for matching; When the matched phase lag parameter falls into the first numerical range characterizing the small inertia following characteristics, a determination result confirming it as smoke aerosol is generated, and a level flip signal is sent to the external alarm interface.

10. The intelligent detection and early warning device according to claim 8, characterized in that, The dynamics determination module further includes: When the matched phase hysteresis parameter falls into the second numerical range characterizing the large inertial hysteresis, a determination result confirming that the particles are not fire particles is generated and the environmental pollution event is recorded. The amplitude of the motion response component is monitored in real time. When the amplitude is lower than the system noise floor threshold, or when the vector synthesis magnitude of the integral values ​​of the in-phase component and the quadrature component is lower than the confidence threshold, a zeroing operation is performed to block the judgment process.

Images