Laser dust particle counting method assisted by optical tweezer technology

Abstract

The present application relates to the fields of precision optical measurement and environmental monitoring, and discloses a laser dust particle counting method assisted by optical tweezers technology, comprising the following steps: S1, constructing a particle detection channel, introducing a dust-containing gas to be measured into the particle detection channel at a constant flow rate, and generating a holographic optical potential well array moving along the airflow direction in the particle detection channel by using a spatial light modulator, so that the particles are dynamically bound by the holographic optical potential well array when passing through the detection area; S2, using a photodetector to continuously collect the scattering light signal generated by the particles under the action of the holographic optical potential well array. By introducing high-frequency position perturbation and digital phase-locked loop phase demodulation technology, active excitation detection of the physical properties of the particles is realized, the phase lag response of the particles to the high-frequency light field perturbation is used, the fluid dynamics characteristics which are difficult to directly measure are converted into accurate phase signals, and the multi-dimensional resolution of the measurement is significantly improved.

Description

Technical Field

[0001] This invention relates to the field of precision optical measurement and environmental monitoring technology, specifically to a laser dust particle counting method assisted by optical tweezers technology. Background Technology

[0002] Laser dust particle counters are core devices for monitoring cleanroom environments, air quality, and particulate matter concentration in industrial production processes. Their mainstream technology is usually based on the Mie scattering principle. The working logic of such devices is to use an air pump system to draw dust-laden gas into a dark chamber at a constant flow rate, and use a laser beam to irradiate the sampling airflow. When particles pass through the photosensitive area, they generate scattered light. A photodetector receives the scattered light and converts it into an electrical pulse signal. Then, the particle size and number are analyzed and inverted through the pulse height. Although this technology is relatively mature in the detection of micron-level particles, it still faces significant technical bottlenecks when dealing with the detection of nano-level ultrafine particles and complex unsteady flow field environments.

[0003] In existing technologies, with the increasing precision of semiconductor manufacturing processes and the improvement of environmental monitoring standards, the requirements for detection sensitivity are becoming increasingly stringent. However, traditional laser scattering methods are limited by Rayleigh scattering law. For particles with extremely small diameters, the intensity of scattered light decreases sharply with the sixth power of the particle size, resulting in extremely weak signals that are easily drowned out by background noise, leading to missed detections. In order to compensate for the insufficient signal strength, theoretically, the integration time of the detector needs to be increased, but this contradicts the need for high-throughput sampling in industrial applications. In high-throughput sampling mode, the time for particles to pass through the beam is often on the order of microseconds. The extremely short exposure time limits the amount of photons collected, further deteriorating the signal-to-noise ratio.

[0004] Furthermore, since laser beams are typically Gaussian distributed, the energy density difference between the beam center and the edge is enormous. When a particle only passes by the edge of the beam, the amplitude of the scattered light pulse it generates is low, making it very easy for the system to misjudge it as a small particle. This edge effect seriously affects the accuracy of particle size distribution analysis. At the same time, nanoscale particles are significantly affected by Brownian motion in the airflow, and their trajectories are highly random. They are difficult to pass stably along the streamline through the focal plane of the detection area like large particles, resulting in poor illumination uniformity. Moreover, it is difficult to constrain their random diffusion through simple mechanical air path design.

[0005] Although the academic community has attempted to introduce optical tweezers technology to manipulate particles using optical gradient forces, existing applications of optical tweezers are mostly limited to static single-particle capture in biological cells or liquid environments, or simply passive guidance using static optical funnels. This static or passive optical manipulation method is inadequate when facing high-speed flowing aerosol sampling: static optical traps, although having strong binding forces, can block the gas path and cannot achieve continuous counting; while passive optical guidance has limited binding force on tiny particles with violent Brownian motion and cannot sense changes in environmental parameters. In practical applications, fluctuations in ambient temperature, humidity, and air pressure can change the viscosity coefficient of the fluid, and existing devices usually use fixed parameter models for inversion, lacking in-situ environmental sensing and adaptive calibration capabilities, resulting in large deviations in measurement data under non-standard operating conditions. Therefore, this invention designs a laser dust particle counting method using optical tweezers technology to address the above-mentioned problems. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a laser dust particle counting method assisted by optical tweezers technology. This method solves the problems of existing laser dust particle counting technologies, such as difficulty in achieving both high-throughput sampling and precise capture of tiny particles in high-speed flowing environments, and poor particle size inversion accuracy due to the inability to sense changes in the viscosity coefficient of the environmental fluid in real time.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a laser dust particle counting method assisted by optical tweezers technology, comprising the following steps:

[0008] S1. Construct a particle detection channel, introduce the dust-laden gas to be tested into the particle detection channel at a constant flow rate, and use a spatial light modulator to generate a holographic potential trap array that moves along the airflow direction in the particle detection channel, so that the particles are dynamically bound by the holographic potential trap array when passing through the detection area.

[0009] S2. Continuously collect the scattered light signal generated by the particles under the action of the holographic potential trap array using a photodetector;

[0010] S3. Based on the dynamic characteristics of particles under the interaction of fluid and light field, a state-space model in the light-fluid coupled field is constructed. The collected scattered light signal is processed using a state estimation algorithm to obtain the state estimate of the particles and the corresponding estimation error covariance in real time. The state estimate includes at least the dynamic viscosity coefficient of the fluid environment.

[0011] S4. Monitor the estimation error covariance in real time, and control the spatial light modulator based on the change of the estimation error covariance to dynamically adjust the light field parameters of the holographic potential trap array in order to maintain effective particle capture.

[0012] S5. After the particle passes through the detection area, the particle size is inverted using the converged state estimate, and the final counting data can be obtained based on the inversion result.

[0013] Preferably, in the step of constructing the particle detection channel in step S1, the moving speed of the holographic potential trap array along the airflow direction is set to be less than the flow rate of the gas, so that the particles have a relative motion tendency with respect to the holographic potential trap array.

[0014] Preferably, the holographic potential trap array, in addition to its basic motion moving along the airflow direction, is superimposed with high-frequency position perturbations; the instantaneous position of the center of the potential trap in the holographic potential trap array... The following relationship must be satisfied:

[0015]

[0016] in, Based on movement speed, Sampling time, The sampling interval is... For the amplitude of the perturbation, The frequency is the perturbation angular frequency.

[0017] Preferably, after the step of continuously acquiring the scattered light signal using a photodetector, the method further includes: using digital phase-locked loop technology at the micro-perturbation angular frequency. Using the reference frequency, the acquired scattered light signal is demodulated to extract the phase lag of the particle motion relative to the high-frequency position perturbation. The intensity value of the scattered light signal and the phase hysteresis Together they form the observation vector and are input into the state estimation algorithm.

[0018] Preferably, in the step of constructing the state-space model in the optical-fluid coupled field, the state-space model is based on the Langevin equation, and its discretized form includes:

[0019]

[0020] in, It is a state vector that includes particle position, particle velocity, and the dynamic viscosity coefficient. For process noise; state transition function It includes a fluid drag term and an optical gradient force term; the fluid drag term is positively correlated with the dynamic viscosity coefficient, and the optical gradient force term is related to the spatial distribution of the holographic optical potential trap array.

[0021] Preferably, the processing of the acquired scattered light signal using the state estimation algorithm specifically includes: establishing a phase lag observation equation:

[0022]

[0023] in, Where is the particle radius, For the stiffness of the potential trap, The dynamic viscosity coefficient, To eliminate observation noise, the extended Kalman filter algorithm is used to correct the predicted state vector in conjunction with the observation vector, thereby obtaining the optimal state estimate and the posterior estimation error covariance matrix.

[0024] Preferably, the step of real-time monitoring of the estimation error covariance and feedback control of the spatial light modulator specifically includes: extracting the position variance component in the estimation error covariance matrix; comparing the position variance component with a preset safety threshold; if the position variance component is greater than the safety threshold, determining that the particle has an escape risk, and generating a feedback command to adjust the phase hologram of the spatial light modulator.

[0025] Preferably, the dynamic adjustment of the optical field parameters of the holographic potential trap array is specifically achieved by increasing the stiffness of the potential trap based on the degree to which the position variance component exceeds the safety threshold. Adjusted stiffness :

[0026]

[0027] in, Based on the stiffness, For the location variance components, As a safety threshold, This is the feedback gain coefficient.

[0028] Preferably, the method of dynamically adjusting the optical field parameters of the holographic optical potential trap array further includes: controlling the spatial light modulator to generate a lateral optical barrier in the normal direction of the particle's predicted trajectory to limit the lateral diffusion of the particles.

[0029] Preferably, in the step of inverting particle size, the specific calculation method is as follows: extract the converged estimate of the dynamic viscosity coefficient. and phase lag mean The particle radius is calculated using the following formula.

[0030]

[0031] The count data is output only when the trace of the estimated error covariance during particle passage is consistently below a preset divergence threshold.

[0032] This invention provides a laser dust particle counting method assisted by optical tweezers technology. It has the following beneficial effects:

[0033] 1. This invention effectively solves the contradiction between high-throughput sampling and high signal integration time in traditional laser particle counting by constructing a dynamic holographic potential trap array that moves along the airflow direction at a speed lower than the flow velocity. It can significantly extend the residence time of particles in the photosensitive area while maintaining continuous airflow, thereby increasing the effective integration time of the photodetector and greatly improving the signal-to-noise ratio of weak scattering signals without sacrificing sampling throughput.

[0034] 2. This invention achieves active excitation detection of particle physical properties by introducing high-frequency position perturbation and digital phase-locked loop (PLL) phase demodulation technology. It breaks away from the limitation of traditional methods that rely solely on the intensity of scattered light for passive measurement. By utilizing the phase lag response of particles to high-frequency light field perturbation, it transforms the difficult-to-measure hydrodynamic characteristics into accurate phase signals, providing an independent observation dimension other than light intensity for the particle size inversion of microparticles, and significantly improving the multidimensional resolution of the measurement.

[0035] 3. This invention establishes a state-space model under an optical-fluid coupled field and employs the Extended Kalman Filter (EKF) algorithm, effectively overcoming the randomness of the trajectory and signal noise interference caused by the strong Brownian motion of nanoscale particles. Unlike the traditional pulse height analysis method based on threshold judgment, this invention transforms the particle counting process into a state estimation process, uses the state transition equation to predict the particle trajectory, and uses observation data to correct the estimated value. It can accurately separate the real particle signal from the weak signal submerged by noise, thereby significantly reducing the sensitivity to background noise and improving the detection limit for extremely small particles.

[0036] 4. This invention constructs an active closed-loop control system to prevent particle escape by monitoring and estimating the error covariance matrix and controlling the spatial light modulator (SLM) in real time. The covariance matrix is ​​used to quantify the uncertainty of the system regarding the particle position. When the uncertainty increases, the optical trap stiffness is dynamically enhanced or a lateral auxiliary barrier is generated in milliseconds. This on-demand optical field reconstruction mechanism not only ensures the effective locking of unstable particles, but also avoids the thermal effect problems that may be caused by continuous high-power irradiation.

[0037] 5. This invention calculates the current fluid viscosity in real time by analyzing the motion response of particles in a dynamic light field, and corrects the Stokes drag model and particle size inversion formula based on the real-time parameters, thereby realizing in-situ adaptive calibration of the equipment and greatly improving the measurement accuracy and stability of the equipment in non-steady-state industrial environments. Attached Figure Description

[0038] Figure 1 This is one of the schematic diagrams of the method flow of the present invention;

[0039] Figure 2 This is a second schematic diagram of the method flow of the present invention;

[0040] Figure 3 This is the third schematic diagram of the method flow of the present invention;

[0041] Figure 4 This is the fourth schematic diagram of the method flow of the present invention;

[0042] Figure 5 This is the fifth schematic diagram of the method flow of the present invention;

[0043] Figure 6 This is the sixth schematic diagram of the method flow of the present invention;

[0044] Figure 7 This is the seventh schematic diagram of the method flow of the present invention. Detailed Implementation

[0045] 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.

[0046] Please see the appendix Figure 1 - Appendix Figure 7 This invention provides a laser dust particle counting method using optical tweezers technology, comprising the following steps:

[0047] S1. Construct a particle detection channel. The dust-laden gas to be tested is introduced into the particle detection channel at a constant flow rate. A spatial light modulator is used to generate a holographic potential trap array moving along the airflow direction within the particle detection channel. This causes the particles to be dynamically constrained by the holographic potential trap array as they pass through the detection area. In the particle detection channel construction step, the moving speed of the holographic potential trap array along the airflow direction is set to be less than the gas flow rate, causing the particles to have a relative motion tendency with respect to the holographic potential trap array. In addition to the basic motion of the holographic potential trap array moving along the airflow direction, a high-frequency position perturbation is superimposed. The instantaneous position of the center of the potential trap in the holographic potential trap array... The following relationship must be satisfied:

[0048]

[0049] in, Based on movement speed, Sampling time, The sampling interval is... For the amplitude of the perturbation, Following the step of continuously acquiring scattered light signals using a photodetector, the process further includes: using digital phase-locked loop technology to determine the perturbation frequency at a specific angular frequency. Using the reference frequency, the acquired scattered light signal is demodulated to extract the phase lag of the particle motion relative to the high-frequency position perturbation. The intensity value and phase lag of the scattered light signal Both are used as observation vectors and input into the state estimation algorithm;

[0050] Specifically, the physical detection environment is first constructed by using microfluidic technology to fabricate a particle detection channel with a light-transmitting observation window. A precision air pump system is then used to pump the dust-laden gas to be tested at a constant flow rate. In the particle detection channel, the stability of the flow field is the basis for subsequent measurements. At the same time, a holographic optical path system is built. The coherent light emitted by the laser source is expanded and collimated before illuminating the liquid crystal target surface of the spatial light modulator. The SLM, as the core optical field controller, loads the initial phase hologram generated by the computational holographic algorithm. After the hologram is Fourier transformed by the objective lens, a series of optical potential trap arrays arranged along the flow direction are generated on the central axis of the detection channel. These optical potential trap arrays use optical gradient force to provide non-contact physical confinement for the particles about to enter the detection area.

[0051] By refreshing the phase map of the SLM at a high frame rate, the optical potential trap is controlled to move along the airflow direction. To utilize fluid resistance to stabilize the particles, the base translational velocity of the optical potential trap is set to... And must meet This velocity difference causes the particles to be dragged by the optical potential trap as they move with the airflow, thus softly binding them near the center of the trap and prolonging their residence time in the detection area. To actively detect the particle's physical properties (such as mass and viscous drag), a high-frequency position perturbation is superimposed on the basic translational motion of the optical potential trap. Each sampling time (sampling interval is) The instantaneous position of the center of the optical potential trap in the holographic optical potential trap array. Controlled to satisfy the following relationship:

[0052]

[0053] in, For perturbation amplitude (typically submicron level), The perturbation angular frequency (typically on the order of kilohertz) serves as the excitation source for the system.

[0054] S2. Continuously collect the scattered light signal generated by particles under the action of a holographic potential trap array using a photodetector;

[0055] Specifically, when a particle passes through the detection area, it generates scattered light due to the dynamic optical potential trap. A high-sensitivity photodetector (such as a PMT or APD) is used to continuously collect the intensity signal of the scattered light. At this time, the original signal collected contains a mixture of the low-frequency envelope signal generated by the particle passing through the beam and the high-frequency modulation signal caused by high-frequency perturbations. This step uses digital phase-locked loop technology for signal processing.

[0056] Reference signal input: The perturbation angular frequency in step S1 As a reference frequency for the PLL;

[0057] Quadrature demodulation: The scattered light signal is mixed with the reference signal and low-pass filtered;

[0058] Phase extraction: Extracting the phase lag of particle motion relative to the optical potential trap perturbation. .

[0059] Particles in viscous fluids possess inertia and resistance, and their motion cannot perfectly follow the high-frequency jitter of the optical trap, resulting in phase hysteresis. It directly reflects the relaxation time of the particles and is a key observation for subsequent decoupling of particle size and environmental viscosity.

[0060] S3. Based on the dynamic characteristics of particles interacting with fluid and light fields, a state-space model in the optical-fluid coupled field is constructed. The collected scattered light signal is processed using a state estimation algorithm to obtain the particle's state estimate and the corresponding estimation error covariance in real time. The state estimate must include at least the dynamic viscosity coefficient of the fluid environment. In the step of constructing the state-space model in the optical-fluid coupled field, the state-space model is based on the Langevin equation, and its discretized form includes:

[0061]

[0062] in, It is a state vector containing particle position, particle velocity, and dynamic viscosity coefficient. For process noise; state transition function It includes fluid drag and optical gradient force terms; the fluid drag term is positively correlated with the dynamic viscosity coefficient, and the optical gradient force term is related to the spatial distribution of the holographic optical potential trap array. The collected scattered light signal is processed using a state estimation algorithm, specifically including: establishing the phase lag observation equation.

[0063]

[0064] in, Where is the particle radius, For the stiffness of the potential trap, The dynamic viscosity coefficient, To mitigate observation noise, the extended Kalman filter algorithm is used to correct the predicted state vector based on the observation vector, yielding the optimal state estimate and the posterior estimation error covariance matrix. In the particle size inversion step, the specific calculation method is as follows: extract the converged dynamic viscosity coefficient estimate. and phase lag mean The particle radius is calculated using the following formula.

[0065]

[0066] The count data is output only when the trace of the estimation error covariance during the particle's passage is consistently below a preset divergence threshold.

[0067] Specifically, in order to accurately reconstruct particle trajectories and environmental parameters from noise, this embodiment constructs a state-space model based on the Langevin equation and employs the Extended Kalman Filter (EKF) algorithm; the system state vector is defined. ,in For particle position, For particle velocity, Let be the dynamic viscosity coefficient of the fluid environment (representing the enhanced state to be estimated). The discretized state transition equation is:

[0068]

[0069] in, To simulate the effects of Brownian motion on process noise, the state transition function is... The dynamic evolution of particles under the combined action of fluid drag and optical gradient force is described:

[0070] Fluid drag term: follows Stokes' law, described as ,in Where is the particle radius, here These are state variables that evolve over time;

[0071] The light gradient force term is approximately approximated by Hooke's law and is described as follows: ,in For the stiffness of the potential trap, Establish the observation vector for the instantaneous position of the optical trap determined in step S1. With state vector The nonlinear mapping relationship, the observation vector includes the scattered light intensity and phase lag The phase lag observation equation is as follows:

[0072]

[0073] This equation clarifies the relationship between phase lag and dynamic viscosity coefficient. The functional relationship;

[0074] Algorithm execution prediction-correction loop:

[0075] Prediction: Based on the state estimate from the previous time step, use the state transition equation to predict the prior state at the current time step;

[0076] Correction: Calculate the Kalman gain using the current observation vector. The prior state is corrected to output the optimal state estimate. and the posterior estimation error covariance matrix .

[0077] S4. Real-time monitoring of the estimation error covariance and feedback control of the spatial light modulator based on changes in the estimation error covariance to dynamically adjust the optical field parameters of the holographic potential trap array to maintain effective particle capture. The steps of real-time monitoring of the estimation error covariance and feedback control of the spatial light modulator specifically include: extracting the position variance component from the estimation error covariance matrix; comparing the position variance component with a preset safety threshold; if the position variance component is greater than the safety threshold, determining that the particle has an escape risk, generating a feedback command to adjust the phase hologram of the spatial light modulator, and dynamically adjusting the optical field parameters of the holographic potential trap array. Specifically, this is done by increasing the stiffness of the potential trap according to the degree to which the position variance component exceeds the safety threshold. Adjusted stiffness satisfy:

[0078]

[0079] in, Based on the stiffness, For the location variance components, As a safety threshold, To provide feedback on the gain coefficient, other methods for dynamically adjusting the optical field parameters of a holographic potential trap array include:

[0080] The spatial light modulator is controlled to generate a lateral optical barrier in the normal direction of the predicted particle trajectory to limit the lateral diffusion of the particles.

[0081] Specifically, in order to prevent tiny particles from escaping the optical trap due to strong Brownian motion, a closed-loop feedback mechanism is introduced in this step.

[0082] Uncertainty monitoring: Real-time extraction of the posterior estimation error covariance matrix from the EKF output. Location variance components This value quantifies the uncertainty of the algorithm's current estimate of the particle's position;

[0083] Escape risk assessment: Setting a safety threshold ,like This indicates that the particles are severely affected by random forces and are at risk of escaping.

[0084] Active optical field reconstruction: The system generates feedback commands based on the judgment results to adjust the phase hologram of the SLM. The adjustment strategies include: stiffness enhancement: dynamically increasing the stiffness of the optical potential trap according to the degree of deviation. To enhance binding force, the formula is adjusted as follows:

[0085]

[0086] in To provide feedback gain coefficients, a lateral barrier is generated: a high-intensity optical barrier is generated in the normal direction (lateral) of the particle's predicted trajectory, physically restricting the lateral diffusion of the particle. Through this step, the measurement system transforms from traditional passive observation to active control, ensuring that the particle is always in the optimal detection area.

[0087] S5. After the particle passes through the detection area, the particle size is inverted using the converged state estimate, and the final counting data can be obtained based on the inversion result.

[0088] Specifically, after the particles pass through the detection area, the final calculation is performed based on the data from the entire process;

[0089] Validity verification: Examine the covariance trace Tr during particle traversal. If the value is consistently lower than the preset divergence threshold, it indicates that the EKF algorithm has converged throughout the entire process, the trajectory has been successfully locked, and it is determined to be a valid counting event.

[0090] Parameter inversion: Extracting the final convergent estimate of the dynamic viscosity coefficient from the state vector. and the mean of phase lag ,because This is the true viscosity obtained by the system adaptively tracking environmental changes. Using it for particle size calculation can eliminate environmental errors. The inversion formula is:

[0091]

[0092] Count output: Based on the calculated particle radius The particles are then categorized into the corresponding particle size channels, and the final particle count results and particle size distribution information are output.

[0093] This embodiment achieves a deep integration of optical tweezers dynamics, fluid mechanics and control theory through the above steps, effectively solving the problems of inaccurate counting and capture of tiny dust particles in complex environments in existing technologies.

[0094] In summary, this invention provides a laser dust particle counting method assisted by optical tweezers technology. By constructing a dynamic holographic potential trap array that moves along the airflow direction at a speed lower than the flow velocity, it effectively solves the contradiction between high-throughput sampling and high signal integration time in traditional laser particle counting. While maintaining continuous airflow, it significantly extends the residence time of particles in the photosensitive area, thereby increasing the effective integration time of the photodetector. By introducing high-frequency position perturbation and digital phase-locked loop (PLL) phase demodulation technology, it achieves active excitation detection of particle physical properties, providing an independent observation dimension other than light intensity for particle size inversion, significantly improving the multidimensional resolution of the measurement. Furthermore, by analyzing the particle's motion response in the dynamic light field, the current fluid viscosity is calculated in real time, and the Stokes drag model and particle size inversion formula are corrected based on this real-time parameter, enabling in-situ environmental adaptive calibration of the equipment. This greatly improves the measurement accuracy and stability of the equipment in non-steady-state industrial environments.

[0095] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A laser dust particle counting method assisted by optical tweezers technology, characterized in that, Includes the following steps: S1. Construct a particle detection channel, introduce the dust-laden gas to be tested into the particle detection channel at a constant flow rate, and use a spatial light modulator to generate a holographic potential trap array that moves along the airflow direction in the particle detection channel, so that the particles are dynamically bound by the holographic potential trap array when passing through the detection area. S2. Continuously collect the scattered light signal generated by the particles under the action of the holographic potential trap array using a photodetector; S3. Based on the dynamic characteristics of particles under the interaction of fluid and light field, a state space model in the light-fluid coupled field is constructed. The collected scattered light signal is processed by the state estimation algorithm, and the state estimate of the particle and the corresponding estimation error covariance are calculated in real time. The state estimate includes at least the dynamic viscosity coefficient of the fluid environment; S4. Monitor the estimation error covariance in real time, and control the spatial light modulator based on the change of the estimation error covariance to dynamically adjust the light field parameters of the holographic potential trap array in order to maintain effective particle capture. S5. After the particle passes through the detection area, the particle size is inverted using the converged state estimate, and the final counting data can be obtained based on the inversion result.

2. The laser dust particle counting method using optical tweezers technology assisted according to claim 1, characterized in that, In the step of constructing the particle detection channel described in step S1, the moving speed of the holographic potential trap array along the airflow direction is set to be less than the flow rate of the gas, so that the particles have a relative motion tendency with respect to the holographic potential trap array.

3. The laser dust particle counting method using optical tweezers technology assisted according to claim 2, characterized in that, The holographic potential trap array, in addition to its basic motion moving along the airflow direction, is also superimposed with high-frequency position perturbations; the instantaneous position of the center of the potential trap in the holographic potential trap array... The following relationship must be satisfied: in, Based on movement speed, Sampling time, The sampling interval is... For the amplitude of the perturbation, The frequency is the perturbation angular frequency.

4. The laser dust particle counting method using optical tweezers technology assisted according to claim 3, characterized in that, Following the step of continuously acquiring scattered light signals using a photodetector, the method further includes: using digital phase-locked loop technology at the micro-perturbation angular frequency. Using the reference frequency, the acquired scattered light signal is demodulated to extract the phase lag of the particle motion relative to the high-frequency position perturbation. The intensity value of the scattered light signal and the phase hysteresis Together they form the observation vector and are input into the state estimation algorithm.

5. The laser dust particle counting method using optical tweezers technology assisted according to claim 1, characterized in that, In the step of constructing the state-space model in the optical-fluid coupled field, the state-space model is based on the Langevin equation, and its discretized form includes: in, It is a state vector that includes particle position, particle velocity, and the dynamic viscosity coefficient. For process noise; state transition function It includes a fluid drag term and an optical gradient force term; the fluid drag term is positively correlated with the dynamic viscosity coefficient, and the optical gradient force term is related to the spatial distribution of the holographic optical potential trap array.

6. The laser dust particle counting method using optical tweezers technology assisted according to claim 5, characterized in that, The process of processing the acquired scattered light signal using a state estimation algorithm specifically includes: establishing a phase lag observation equation: in, Where is the particle radius, For the stiffness of the potential trap, The dynamic viscosity coefficient, To eliminate observation noise, the extended Kalman filter algorithm is used to correct the predicted state vector in conjunction with the observation vector, thereby obtaining the optimal state estimate and the posterior estimation error covariance matrix.

7. The laser dust particle counting method using optical tweezers technology assisted according to claim 1, characterized in that, The step of real-time monitoring of the estimation error covariance and feedback control of the spatial light modulator specifically includes: extracting the position variance component in the estimation error covariance matrix; comparing the position variance component with a preset safety threshold; if the position variance component is greater than the safety threshold, determining that the particle has an escape risk, and generating a feedback command to adjust the phase hologram of the spatial light modulator.

8. The laser dust particle counting method using optical tweezers technology assisted according to claim 7, characterized in that, The dynamic adjustment of the optical field parameters of the holographic potential trap array is specifically achieved by increasing the stiffness of the potential trap based on the degree to which the position variance component exceeds the safety threshold. Adjusted stiffness satisfy: in, Based on the stiffness, For the location variance components, As a safety threshold, This is the feedback gain coefficient.

9. The laser dust particle counting method using optical tweezers technology assisted according to claim 7, characterized in that, The method for dynamically adjusting the optical field parameters of the holographic potential trap array also includes: The spatial light modulator is controlled to generate a lateral optical barrier in the normal direction of the predicted particle trajectory to limit the lateral diffusion of the particles.

10. The laser dust particle counting method using optical tweezers technology assisted according to claim 6, characterized in that, In the step of inverting particle size, the specific calculation method is as follows: extract the converged estimate of the dynamic viscosity coefficient. and phase lag mean The particle radius is calculated using the following formula. : The count data is output only when the trace of the estimated error covariance during particle passage is consistently below a preset divergence threshold.

Images