A collaborative optimization control method for workshop cleanliness and air conditioning fan rotation frequency

Abstract

The present application provides a method for coordinated optimization control of workshop cleanliness and air-conditioning fan rotation frequency, comprising: sampling and analyzing dust of different particle size segments in workshops of different cleanliness levels, obtaining mass fraction data and particle size distribution characteristic parameters of dust in each particle size segment, and obtaining dust particle size distribution data; arranging multiple hot-bulb anemometers in workshops of different cleanliness levels to synchronously measure air flow velocity and turbulence intensity in different areas, analyzing the mapping relationship between air-conditioning fan rotation frequency and local wind speed and turbulence intensity, and obtaining air flow velocity and turbulence intensity distribution under different air-conditioning fan rotation frequencies; determining the target control concentration threshold for dust of different particle size segments, combining the sedimentation rate and diffusion distance data of dust of different particle sizes, obtaining the intersection of air-conditioning fan rotation frequency thresholds that simultaneously meet the constraints of sedimentation uniformity and diffusion distance, and determining the optimal control range of the air-conditioning fan rotation frequency.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method for coordinated optimization control of workshop cleanliness and air-conditioning fan rotation frequency. Background Art

[0002] In industrial production, different workshops have distinct individual characteristics in their demands for air conditioning systems and cleanliness control, which has led to a series of complex technical contradictions. Due to differences in production characteristics, precision electronics workshops and ordinary mechanical processing workshops have completely different requirements for the operating parameters of air conditioning fans. The former requires high-frequency operation to ensure an ultra-clean environment, while the latter can moderately reduce the frequency, which is like seeking a balance between precision clocks and rough machinery. The relationship between air conditioning fans and dust collection systems is even more subtle. Although the high-speed operation of the fan can improve the efficiency of air circulation, it also invisibly aggravates the diffusion of dust and increases the burden on the dust collection system; on the contrary, although reducing the fan frequency helps dust collection, it may form dead corners for dust accumulation in certain areas, just like swinging between cleaning and blockage. What is even more tricky is that the game between cleanliness and energy consumption, how to achieve energy conservation and emission reduction through intelligent control strategies without sacrificing production quality, has become a difficult problem facing industrial production. These intertwined factors, like an intricate network, make the coordinated control of workshop air conditioning systems extremely complex. How to find the best balance point under the framework of multi-objective optimization has become a technical challenge that needs to be solved urgently. Summary of the invention

[0003] The present invention provides a method for coordinated optimization control of workshop cleanliness and air conditioning fan rotation frequency, which mainly includes:

[0004] Sampling and analyzing dust of different particle size segments in workshops with different cleanliness levels, obtaining the mass fraction data and particle size distribution characteristic parameters of dust in each particle size segment, and obtaining dust particle size distribution data;

[0005] Track the movement trajectory of dust in different particle size segments and obtain the dust settling rate and diffusion distance data, including recording the dust falling process to calculate the settling rate, obtaining the dust concentration at different locations to determine the diffusion distance, and combining the dust particle size distribution data to establish the relationship between the dust particle size and the settling rate and diffusion distance under different cleanliness levels;

[0006] By placing multiple hot-bulb anemometers in workshops with different cleanliness levels to synchronously measure the air flow velocity and turbulence intensity in different areas, the mapping relationship between the air conditioning fan rotation frequency and the air flow velocity and turbulence intensity is analyzed to obtain the air flow velocity and turbulence intensity distribution under different air conditioning fan rotation frequencies;

[0007] The settling rates of dust particles of different particle sizes are combined with the turbulence intensity distribution under different air-conditioning fan rotation frequencies. Computational fluid dynamics software is used to simulate the dust movement trajectory under different rotation frequencies, and the standard deviation of dust settling distribution is calculated to determine whether the settling uniformity requirements are met. The air-conditioning fan rotation frequency threshold that meets the settling uniformity requirements is obtained.

[0008] The dust diffusion distances of different particle sizes and the air flow velocity data under different air-conditioning fan rotation frequencies are combined to simulate the dust diffusion process under different air-conditioning fan rotation frequencies. According to different cleanliness level standards, it is judged whether the diffusion distance exceeds the safety range threshold, and the air-conditioning fan rotation frequency threshold that meets the diffusion distance constraint is obtained.

[0009] For dust in different particle size segments, the target control concentration threshold is determined. Combined with the dust settling rate and diffusion distance data of different particle sizes, the intersection of the air-conditioning fan rotation frequency thresholds that simultaneously meet the constraints of settling uniformity and diffusion distance is obtained, and the optimal control range of the air-conditioning fan rotation frequency is determined.

[0010] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0011] The present invention discloses a method for collaborative optimization control of workshop cleanliness and air-conditioning fan rotation frequency. The method first uses a laser particle size analyzer to sample and analyze the dust in the workshop to obtain the particle size distribution data of the dust; then the falling process of the dust is recorded by a high-speed camera, its sedimentation rate is calculated, and samplers are placed at different positions to measure the dust concentration, thereby determining the diffusion distance of the dust; the air flow velocity and turbulence intensity in different areas are measured synchronously, and the mapping relationship between the air-conditioning fan rotation frequency and the air flow velocity and turbulence intensity is analyzed. In addition, the present invention also combines the dust sedimentation rate and turbulence intensity distribution to simulate the dust movement trajectory under different rotation frequencies, calculates the standard deviation of the dust sedimentation distribution, and determines whether the sedimentation uniformity requirements are met. Further, the present invention simulates the dust diffusion process under different air-conditioning fan rotation frequencies, combines the different cleanliness levels specified in the national standard, judges whether the diffusion distance exceeds the safety range threshold, and finds the air-conditioning fan rotation frequency threshold that meets this constraint. Finally, for dust of different particle size segments, the target control concentration threshold is determined, and the data of sedimentation rate and diffusion distance are integrated to obtain the intersection of the air-conditioning fan rotation frequency threshold that satisfies both the sedimentation uniformity and diffusion distance constraints, thereby determining the optimal control range of the air-conditioning fan rotation frequency. In general, the present invention can effectively control the dust distribution in the workshop, and by accurately adjusting the rotation frequency of the air-conditioning fan, it can ensure the uniform sedimentation of the dust and limit its diffusion range, meeting the environmental requirements of workshops of different cleanliness levels, thereby improving the air quality of the workshop and the safety of the working environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 The present invention is a flow chart of a method for collaboratively optimizing control of workshop cleanliness and air-conditioning fan rotation frequency.

[0013] Figure 2 It is a schematic diagram of a collaborative optimization control method of workshop cleanliness and air-conditioning fan rotation frequency of the present invention.

[0014] Figure 3 It is another schematic diagram of a coordinated optimization control method of a workshop cleanliness and air-conditioning fan rotation frequency of the present invention. DETAILED DESCRIPTION

[0015] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0016] like Figure 1-3 In this embodiment, a method for coordinated optimization control of workshop cleanliness and air conditioning fan rotation frequency may specifically include:

[0017] Step S101, sampling and analyzing dust of different particle size segments in workshops of different cleanliness levels, obtaining mass fraction data and particle size distribution characteristic parameters of dust of each particle size segment, and obtaining dust particle size distribution data.

[0018] A laser particle size analyzer is used to sample the air in workshops of different cleanliness levels to obtain the scattered light intensity signal of dust particles; the particle size of the dust particles is calculated according to the corresponding relationship between the scattered light intensity and the particle size, and the preset particle size segments are classified and counted to obtain the quantity distribution of dust in each particle size segment; the quantity distribution is converted into a mass distribution, and the mass fraction data of dust in each particle size segment is obtained by integral calculation to obtain the mass concentration distribution characteristics of dust in each workshop; the dust particle size distribution data of workshops of different cleanliness levels are compared and analyzed, the difference coefficient of the dust mass fraction in each particle size segment is calculated, the K-means clustering algorithm is used to classify the dust characteristics of different workshops, and the dust particle size cumulative distribution diagram and frequency distribution diagram are drawn.

[0019] Specifically, a laser particle size analyzer is used to sample the air in workshops of different cleanliness levels, and the scattered light intensity signal of dust particles is obtained through the principle of laser scattering. The particle size of dust particles is calculated based on the corresponding relationship between the scattered light intensity and the particle size, and the particle size segments of 0.1μm, 0.2μm, 0.3μm, 0.5μm, 1μm, 3μm, 5μm, 10μm, etc. are classified and counted to obtain the number distribution data of dust in each particle size segment. According to the number distribution data of dust in each particle size segment measured by the laser particle size analyzer, the Hatch-Choate equation is used to convert the number distribution into mass distribution. The equation takes into account the density and shape factors of the particles, and the mass fraction data of dust in each particle size segment is obtained by integral calculation. Data processing is performed for workshops of different cleanliness levels to obtain the mass concentration distribution characteristics of dust in each workshop. The geometric mean particle size and geometric standard deviation were calculated by the moment estimation method, and the log-normal distribution function was fitted by the least squares method to obtain the mathematical model parameters of the dust particle size distribution, including the geometric mean particle size, geometric standard deviation, median particle size, etc., so as to obtain the mathematical model of the dust particle size distribution. The dust particle size distribution data of workshops with different cleanliness levels were compared and analyzed, the difference coefficient of the dust mass fraction in each particle size segment was calculated, and the K-means clustering algorithm was used to classify the dust characteristics of different workshops. The cumulative distribution diagram and frequency distribution diagram of dust particle size were drawn to highlight the differences and characteristics between workshops of different levels, and to fully reflect the particle size distribution characteristics of dust in workshops of different cleanliness levels. When using a laser particle size analyzer to sample the air in workshops of different cleanliness levels, the sampling time was set to 30 minutes and the sampling flow rate was set to 28.3L / min. The laser particle size analyzer uses a semiconductor laser with a wavelength of 650nm to determine the particle size by measuring the intensity of scattered light. For particle size segments such as 0.1μm, 0.2μm, 0.3μm, 0.5μm, 1μm, 3μm, 5μm, and 10μm, the number of particles is recorded respectively. For example, in a Class 100 cleanroom, 1000 particles are detected in the 0.1μm particle size segment, 500 particles are detected in the 0.3μm particle size segment, and 100 particles are detected in the 1μm particle size segment. The Hatch-Choate equation, which converts the number distribution of particles into the mass distribution, is used to convert the number distribution into the mass distribution, assuming that the density of the dust is 2.5 grams per cubic centimeter and the shape factor is 1.5. Through integral calculation, the mass fraction of dust in each particle size segment is obtained. The results show that the mass fraction of the 0.1μm particle size segment is 0.5%, the 0.3μm particle size segment is 2%, and the 1μm particle size segment is as high as 10%. Furthermore, in order to quantify the central tendency and dispersion of the particle size distribution, the moment estimation method was used based on the statistical characteristics of the particle mass distribution to calculate the geometric mean particle size to be 0.8 μm and the geometric standard deviation to be 2.1.On this basis, the least squares method was used to fit the data to the log-normal distribution function, and the fitting parameters μ (i.e., logarithmic mean particle size) were obtained to be equal to 0.8 of the natural logarithm (approximately equal to -0.223), and σ (i.e., logarithmic standard deviation) was equal to 2.1 of the natural logarithm (approximately equal to 0.742). These parameters reveal the specific form of the particle size distribution. In order to evaluate the similarities and differences in the dust particle size distribution between workshops of different cleanliness levels, the dust particle size distribution data of workshops of different cleanliness levels were also compared and analyzed, and the coefficient of difference of the dust mass fraction in each particle size segment was calculated, which is helpful to understand the stability and consistency of the particle size distribution under different environments. The coefficient of difference between the 1000-class clean workshop and the 100-class clean workshop in the 0.5μm particle size segment is 1.5, indicating that the dust concentration in this particle size segment of the 1000-class workshop is higher. The K-means clustering algorithm is used to classify the dust characteristics of different workshops, and the number of clusters k is set to 3, representing the three pollution levels of low, medium and high. Through iterative calculation, we get the cluster centers, such as the low pollution level center is (0.2μm, 1000 / m³), the medium pollution level center is (0.5μm, 5000 / m³), and the high pollution level center is (1μm, 10000 / m³). Finally, we draw the dust particle size cumulative distribution diagram and frequency distribution diagram to intuitively display the particle size distribution characteristics of dust in workshops of different cleanliness levels.

[0020] Step S102, tracking the movement trajectory of dust in different particle size segments, obtaining the dust settling rate and diffusion distance data, including recording the dust falling process to calculate the settling rate, obtaining the dust concentration at different positions to determine the diffusion distance; combining the dust particle size distribution data, establishing the relationship between the dust particle size and the settling rate and diffusion distance under different cleanliness levels.

[0021] Obtain a dust falling video sequence, and use the OpenCV image processing library to perform frame analysis on the video sequence; based on the frame analysis results, identify dust particles of different particle size segments, and extract the position changes of the dust particles between adjacent frames; calculate the instantaneous speed of the dust particles based on the position changes and the time interval between adjacent frames, and perform statistics on multiple dust particles of the same particle size segment to obtain the average settling rate of the dust in the particle size segment; in workshops with different cleanliness levels, arrange multiple samplers in the vertical and horizontal directions, measure the dust concentration at different positions through the samplers, and calculate the spatial distribution of the dust concentration; based on the spatial distribution of the dust concentration and the dust particle size distribution data, establish a relationship model between the dust particle size and the settling rate, and use the relationship model to obtain the quantitative relationship between the dust particle size and the settling rate under different cleanliness levels; in combination with the dust particle size distribution data, establish a relationship model between the dust particle size and the diffusion distance, and determine the quantitative relationship between the dust particle size and the diffusion distance under different cleanliness levels.

[0022] Specifically, the falling process of dust of different particle size segments is continuously filmed by a high-speed camera, and the OpenCV image processing library is used to perform frame analysis on the filmed video sequence. The image segmentation and edge detection algorithms are used to identify dust particles of different particle size segments, extract the position changes of dust particles between adjacent frames, calculate the instantaneous velocity of dust particles according to the time interval between adjacent frames, and perform statistical averaging on multiple dust particles of the same particle size segment to obtain the average settling rate of dust in this particle size segment. In workshops with different cleanliness levels, multiple samplers are arranged vertically and horizontally, and the spacing between samplers is determined according to the size of the workshop. The initial concentration is measured at the dust release source, and the dust concentration at different locations is measured by the sampler. The spatial distribution of dust concentration is calculated using the Kriging interpolation method, and the distance when the dust concentration drops to half of the initial value is determined from the concentration distribution curve as the diffusion distance of the dust. Combined with the dust particle size distribution data, the polynomial regression analysis method is used to establish the relationship model between dust particle size and sedimentation rate. The parameters of the relationship curve are obtained by fitting the least squares method. The polynomial order with the largest determination coefficient R² is selected as the best fitting model. At the same time, the degree of fitting of the relationship model is evaluated by correlation analysis to obtain the quantitative relationship between dust particle size and sedimentation rate under different cleanliness levels. Based on the measured dust concentration spatial distribution data, the finite difference method is used to discretize the diffusion equation, and then the iterative algorithm is used to solve the diffusion coefficient. Combined with the dust particle size distribution data, the relationship model between dust particle size and diffusion distance is established. The model parameters are obtained by numerical calculation, so as to determine the quantitative relationship between dust particle size and diffusion distance under different cleanliness levels, and compare and analyze the differences between each cleanliness level. The diffusion equation refers to the mathematical equation that describes the diffusion process of dust in space. For the spatial distribution data of dust concentration, the following is the diffusion equation ∂C / ∂t=D∇²C, where C is the dust concentration, which is usually a function of time t and spatial position (x, y, z), that is, C(t, x, y, z), D is the diffusion coefficient, which indicates the rate at which dust diffuses in space, and ∇² is the Laplace operator, which indicates the second-order derivative in space, that is, ∇²C=∂²C / ∂x²+∂²C / ∂y²+∂²C / ∂z². In the implementation process, a high-speed camera continuously shoots the dust falling process at a rate of 1000 frames per second to obtain a video sequence with a resolution of 1920x1080. The Canny edge detection algorithm in the OpenCV library is used to identify the contours of dust particles, and the low threshold is set to 50 and the high threshold is set to 150. The particle size is calculated by contour area, and the dust is divided into three particle size segments: 0.1-1μm, 1-5μm, and 5-10μm. 100 particles are selected for each particle size segment, and their displacement between adjacent frames is calculated to obtain the average sedimentation rate. The average sedimentation rate of the 1-5μm particle size segment is 0.3mm / s. In a Class 100 clean room, a sampler is arranged every 1m in the vertical direction and every 2m in the horizontal direction.The initial concentration measured at the dust release source was 1000 particles / m³. The spatial distribution of dust concentration was calculated by using the Kriging interpolation method and setting the search radius to 5m. The distance at which the concentration drops to 500 particles / m³ was determined by the interpolation results as the diffusion distance. For the 1-5μm particle size segment, the diffusion distance is 3m. The relationship model between dust particle size and sedimentation rate was established by third-order polynomial regression analysis, and the relationship curve parameters were obtained by least squares fitting. The coefficient of determination R² of polynomials of different orders was calculated, and the one with the largest R² was selected as the best fitting model. For example, for a Class 100 cleanroom, the relationship v=0.05d³-0.2d²+0.3d+0.1 was obtained, where v is the sedimentation rate (mm / s) and d is the particle size (μm). The diffusion equation was discretized using the central difference format, with a grid spacing of 0.1m and a time step of 0.1s. The diffusion coefficient was solved by the Gauss-Seidel iterative algorithm, and the convergence threshold was set to 10. ⁻6 Combined with the dust particle size distribution data, a relationship model between dust particle size and diffusion distance was established. Comparative analysis showed that the diffusion distance of dust in the 1-5μm particle size range in the 1000-class cleanroom was longer than that in the 100-class cleanroom.

[0023] Step S103, by deploying multiple hot-bulb anemometers in workshops with different cleanliness levels to synchronously measure the air flow velocity and turbulence intensity in different areas, analyze the mapping relationship between the air conditioning fan rotation frequency and the air flow velocity and turbulence intensity, and obtain the air flow velocity and turbulence intensity distribution under different air conditioning fan rotation frequencies.

[0024] The positions of key measurement points in the workshop are obtained by pre-simulating the airflow distribution with computational fluid dynamics software. Multiple hot-bulb anemometers are arranged according to the positions of the key measurement points to synchronously collect the airflow velocity and turbulence intensity data of the hot-bulb anemometers. The airflow velocity and turbulence intensity data under different air-conditioning fan rotation frequencies are obtained. The support vector regression algorithm is used to establish a mapping relationship between the air-conditioning fan rotation frequency and the airflow velocity and turbulence intensity. The mapping relationship selects the fan rotation frequency as the independent variable, and the airflow velocity and turbulence intensity as the dependent variables. The airflow velocity and turbulence intensity values ​​of each measuring point under multiple air-conditioning fan rotation frequencies are calculated, and the spatial coordinate information of each measuring point is combined to perform interpolation calculations on the entire workshop space to obtain the airflow velocity and turbulence intensity spatial distribution data under different air-conditioning fan rotation frequencies.

[0025] Specifically, multiple hot-bulb anemometers are arranged in workshops with different cleanliness levels. Computational fluid dynamics software is used to simulate the airflow distribution in advance, and key measurement points are determined as the layout locations of the hot-bulb anemometers. A data acquisition card with a 16-bit resolution and a 100kHz sampling rate is used to synchronously collect the airflow velocity and turbulence intensity data of each measuring point. The sampling frequency is set to 10Hz and the sampling time is 10 minutes to obtain the airflow velocity and turbulence intensity time series data of each measuring point. The air-conditioning fan rotation frequency is controlled by a frequency converter. The frequency range is determined according to the design parameters of the air-conditioning system and the actual operation requirements. Starting from the lowest frequency of 20Hz, the frequency range is gradually increased to the highest frequency of 60Hz in steps of 5Hz. Maintain stable operation for 5 minutes at each frequency, record the current frequency value and synchronously collect the airflow velocity and turbulence intensity data of each measuring point. Repeat this process until the entire frequency range is covered to obtain the airflow velocity and turbulence intensity data set at different air-conditioning fan rotation frequencies. For the collected data set, the support vector regression algorithm is used to establish the mapping relationship between the rotation frequency of the air-conditioning fan and the air flow velocity and turbulence intensity. The fan rotation frequency is selected as the independent variable, and the air flow velocity and turbulence intensity are selected as the dependent variables. The kernel function parameters and penalty factors are optimized by the grid search method to obtain the regression model of each measuring point, so as to establish the mapping relationship database between the rotation frequency of the air-conditioning fan and the air flow velocity and turbulence intensity. Based on the established mapping relationship database, the radial basis function interpolation method is used to calculate the air flow velocity and turbulence intensity values ​​of each measuring point under multiple arbitrary air-conditioning fan rotation frequencies. Combined with the spatial coordinate information of the measuring points, the Kriging interpolation method is used to interpolate the entire workshop space to obtain the spatial distribution data of air flow velocity and turbulence intensity under different air-conditioning fan rotation frequencies. The air flow characteristics of workshops with different cleanliness levels are compared and analyzed, the difference coefficients between the levels are calculated, and the influence of the cleanliness level on the air flow distribution is quantitatively evaluated. In the implementation process, the ANSYS Fluent software is used to simulate the air flow distribution of the Class 100 clean workshop. The mesh division adopts hexahedral meshes with a mesh number of 1 million. The simulation results show that there are 6 key measurement points at a height of 1.5m from the ground, located in the four corners and the center of the workshop. TSI9565 hot ball anemometers are arranged at these locations, and data is collected through the NIPXIe-6363 data acquisition card. The sampling frequency is set to 10Hz and the sampling time is 10 minutes. The air-conditioning fan is controlled by the ABBACS580 inverter. According to the air-conditioning design parameters, the frequency range is determined to be 20Hz to 60Hz. At each frequency point, such as 25Hz, the data is recorded within 5 minutes to obtain 3000 sampling points. The SVR function in the scikit-learn library is used to implement support vector regression, and the RBF kernel function is selected. The parameters C and gamma are optimized by the grid search method, and the value ranges are [0.1,1,10] and [0.01,0.1,1] respectively.For the central area measurement point, the optimal parameters C=1, gamma=0.1 were obtained, and the coefficient of determination R² of the regression model reached 0.95. The Rbf function in the SciPy library was used for radial basis function interpolation, the multiquadric basis function was selected, and the smoothing factor ε was set to 1. Interpolation calculations were performed on a 100×100×50 three-dimensional grid to obtain the airflow distribution of the entire workshop. By comparing the airflow distribution of Class 100 and Class 1000 clean workshops at a frequency of 40Hz, the average wind speed difference coefficient was calculated to be 0.15, and the turbulence intensity difference coefficient was 0.08, which quantified the impact of cleanliness level on airflow distribution.

[0026] Step S104, combining the settling rates of dust of different particle sizes with the turbulence intensity distribution at different air-conditioning fan rotation frequencies, and using computational fluid dynamics software to simulate the dust movement trajectory at different rotation frequencies. Calculate the standard deviation of the dust settling distribution, determine whether the settling uniformity requirements are met, and obtain the air-conditioning fan rotation frequency threshold that meets the settling uniformity requirements.

[0027] According to the relationship between sedimentation rate and particle size, a particle motion equation is established, and the particle motion equation is used to describe the motion trajectory of dust particles. A three-dimensional model of the workshop is constructed, and the geometric structure and surface characteristics of the workshop are obtained by setting grid division parameters and boundary conditions. The airflow field distribution under different rotation frequencies is simulated; the airflow field distribution is combined with the particle motion equation to obtain the motion trajectory of dust particles. A statistical analysis is performed on the motion trajectory of the dust particles, and the statistical analysis results are used to calculate the spatial coordinates of the dust sedimentation distribution. A standard deviation threshold for sedimentation uniformity is set. If the calculated standard deviation is less than the standard deviation threshold, it is judged that the sedimentation uniformity under this rotation frequency meets the requirements; if the calculated standard deviation is greater than the standard deviation threshold, it is judged that the sedimentation uniformity under this rotation frequency does not meet the requirements.

[0028] Specifically, according to the sedimentation rate data of dust with different particle sizes and the turbulence intensity distribution data under the rotation frequency of the air-conditioning fan, a piecewise function is used to describe the relationship between sedimentation rate and particle size, and the particle motion equation is established. The Lagrangian particle tracking method is used to describe the dust motion trajectory, and the particle motion equation is discretized into a difference equation. The fourth-order Runge-Kutta method is used for numerical integration to obtain the position and velocity of dust particles at different times. The ANSYS Fluent software is used to build a three-dimensional model of the workshop, set the meshing parameters, set the boundary conditions considering the adsorption effect of the workshop wall, and set the initial dust distribution as the initial condition according to the measured data. The mapping relationship data between the rotation frequency of the air-conditioning fan and the turbulence intensity distribution is imported, and the k-ε turbulence model is used to simulate the airflow field distribution under different rotation frequencies. The airflow field data is combined with the particle motion equation to calculate the motion trajectory of dust with different particle sizes at each rotation frequency. The Monte Carlo simulation method is used to perform statistical analysis on the motion trajectory of a large number of dust particles. The uniform meshing of the workshop bottom is used to determine the sedimentation position, and the spatial coordinates of the dust sedimentation distribution are calculated. The standard deviation formula is used to calculate the standard deviation of the sedimentation distribution, and the functional relationship between the standard deviation and the rotation frequency of the air-conditioning fan is established by polynomial regression. The trend curve of the standard deviation changing with frequency is fitted by the least squares method. The standard deviation threshold required for sedimentation uniformity is set, and the standard deviation is compared with the threshold to determine whether the sedimentation uniformity at each frequency meets the requirements. The frequency range boundary that meets the requirements is searched by binary search. The initial search range is set to the lowest and highest operating frequencies of the fan. The iterative termination condition is that the search interval is less than the preset accuracy or reaches the maximum number of iterations. Finally, the threshold of the air-conditioning fan rotation frequency that meets the sedimentation uniformity requirements is determined, and the upper and lower limits of the threshold range are output. For dusts of different particle sizes, the above process is repeated to compare and analyze the influence of particle size on sedimentation uniformity. In the implementation process, firstly, for dust in the range of 0.1-10μm, a piecewise function is used to describe the relationship between sedimentation rate and particle size, such as the Stokes formula for the 0.1-1μm segment and the Allen formula for the 1-10μm segment. The Lagrangian particle tracking method was used, the time step was set to 0.01s, and the fourth-order Runge-Kutta method was used for numerical integration. A 100m³ clean workshop model was constructed in ANSYS Fluent, and the mesh was divided into hexahedral meshes with 1 million meshes. The wall adsorption effect was set to complete adsorption, and the initial dust distribution was set according to the log-normal distribution. The standard k-ε turbulence model was used, and the turbulence intensity was set to 5%. For the fan frequency in the range of 20-60Hz, the simulation was performed every 5Hz. The Monte Carlo method was used to simulate the motion trajectory of 10,000 particles, and the bottom surface of the workshop was divided into 100×100 grids. The calculated standard deviation and frequency relationship was fitted by a third-order polynomial, such as σ=0.001f³-0.015f²+0.05f+0.2, where σ is the standard deviation and f is the frequency.The standard deviation threshold is set to 0.5, and the binary search method is used to search for the frequency range that meets the requirements. The initial search range is 20-60Hz, the accuracy is set to 0.1Hz, and the maximum number of iterations is 20. The final frequency range that meets the requirements is 30-45Hz. The three particle sizes of 0.1μm, 1μm, and 10μm are analyzed separately. The results show that the larger the particle size, the narrower the frequency range that meets the uniformity requirements, such as 28-47Hz for 0.1μm and 28-47Hz for 10μm.

[0029] Step S105, combining the dust diffusion distances of different particle sizes and the air flow velocity data under different air-conditioning fan rotation frequencies to simulate the dust diffusion process under different air-conditioning fan rotation frequencies; judging whether the diffusion distance exceeds the safety range threshold according to different cleanliness grade standards, and obtaining the air-conditioning fan rotation frequency threshold that meets the diffusion distance constraint.

[0030] A dust diffusion model is established using the convection-diffusion equation, which describes the change of dust concentration with time and space; the dust diffusion model is numerically solved to obtain the dust concentration distribution; a three-dimensional workshop model is constructed, the three-dimensional workshop model sets grid division parameters and sets initial conditions and boundary conditions in combination with the internal structure of the workshop and the location of the dust source; the airflow field distribution under different rotation frequencies is simulated to obtain airflow field data; the airflow field data is combined with the dust concentration distribution to calculate the diffusion process of dust with different particle sizes under each rotation frequency; the diffusion distance when the dust concentration reaches half of the initial value is calculated, and a functional relationship between the diffusion distance and the rotation frequency of the air-conditioning fan is established; the least squares method is used to fit the functional relationship to obtain a trend curve of the diffusion distance changing with frequency; according to the standard provisions of different cleanliness levels, a diffusion distance safety range threshold is set; the frequency range boundary that meets the requirements is searched by binary search to obtain the air-conditioning fan rotation frequency threshold that meets the diffusion distance constraint.

[0031] Specifically, a dust diffusion model is established based on the data of the diffusion distance of dust with different particle sizes and the data of the air flow velocity under the rotation frequency of the air conditioner fan, and the convection diffusion equation is used to describe the change of dust concentration with time and space. For dust with different particle sizes, the diffusion coefficient related to the particle size is used, such as the diffusion coefficient of the particle size D=kT / (3πμd), where k is the Boltzmann constant, T is the temperature, μ is the air viscosity, and d is the particle size. The diffusion equation is discretized into a difference equation, and the dust concentration distribution is obtained by numerically solving it through the Crank-Nicolson implicit format. The three-dimensional model of the workshop is constructed using ANSYS Fluent software, the meshing parameters are set, and the initial conditions and boundary conditions are set considering the internal structure of the workshop and the location of the dust source. The mapping relationship data of the rotation frequency of the air conditioner fan and the air flow velocity distribution are imported, and the RNGk-ε turbulence model is used to simulate the air flow field distribution under different rotation frequencies. The air flow field data is combined with the dust diffusion model to calculate the diffusion process of dust with different particle sizes at each rotation frequency. The simulated dust diffusion process is analyzed by numerical simulation method, and the change of diffusion distance over time is calculated from the dust release source. The distance when the dust concentration reaches half of the initial value is obtained by cubic spline interpolation algorithm as the diffusion distance, and the functional relationship between the diffusion distance and the rotation frequency of the air-conditioning fan is established. The trend curve of the diffusion distance changing with frequency is obtained by least square fitting. According to the national standards for different cleanliness levels, the diffusion distance safety range threshold is set, and the calculated diffusion distance is compared with the safety range threshold to determine whether the diffusion distance at each frequency exceeds the safety range. The frequency range boundary that meets the requirements is searched by binary search. The initial search range is set to the lowest and highest operating frequencies of the fan. The iterative termination condition is that the search interval is less than the preset accuracy or reaches the maximum number of iterations. Finally, the air-conditioning fan rotation frequency threshold that meets the diffusion distance constraint is determined, and the upper and lower limits of the threshold range are output. Repeat the above process for different cleanliness levels to compare and analyze the influence of cleanliness level on the diffusion distance constraint. In the implementation process, the diffusion coefficient related to the particle size is first used for dust in the range of 0.1-10 μm. For example, for dust with a particle size of 1 μm, the calculated diffusion coefficient is about 2.8×10 ⁻7m² / s. The Crank-Nicolson implicit format is used for numerical solution, with the time step set to 0.1s and the space step set to 0.01m. A 100m³ clean workshop model is constructed in ANSYS Fluent, and the meshing adopts hexahedral mesh with 1 million meshes. Considering the point source release in the center of the workshop, the initial concentration is set to 1000 / m³. The RNGk-ε turbulence model is used, and the turbulence intensity is set to 5%. For the fan frequency in the range of 20-60Hz, the simulation is performed every 5Hz. The diffusion distance is calculated from the dust release source, and the distance when the concentration drops to 500 / m³ is determined by the cubic spline interpolation algorithm as the diffusion distance. The relationship between the diffusion distance and the frequency is obtained by least squares fitting, such as D=0.02f²-1.5f+40, where D is the diffusion distance (m) and f is the frequency (Hz). For a Class 100 clean room, the threshold of the diffusion distance safety range is set to 3m. The binary search method is used to search for a frequency range that meets the requirements. The initial search range is 20-60Hz, the accuracy is set to 0.1Hz, and the maximum number of iterations is 20. The final frequency range that meets the requirements is 35-50Hz. The above process is repeated for the Class 1000 clean room, and the safety range threshold is set to 5m. The frequency range that meets the requirements is 30-55Hz, indicating that lower cleanliness levels allow a wider fan frequency adjustment range.

[0032] Step S106, for dust of different particle size segments, determine the target control concentration threshold, combine the dust settling rate and diffusion distance data of different particle size segments, obtain the intersection of the air-conditioning fan rotation frequency thresholds that simultaneously meet the constraints of settling uniformity and diffusion distance, and determine the optimal control range of the air-conditioning fan rotation frequency.

[0033] The target control concentration data of clean rooms of different cleanliness levels are obtained, and the particle size segments that are not clearly specified are interpolated to obtain a complete particle size-target concentration correspondence data set; based on the complete particle size-target concentration correspondence data set, a target concentration threshold table containing key particle size segments is established; an optimization model is constructed, and the sedimentation uniformity and diffusion distance constraints are used as optimization objective functions, and the air conditioning fan rotation frequency is used as a decision variable; the Pareto optimal solution set is screened, and the standard deviation threshold required for sedimentation uniformity is set, and the solution that satisfies the standard deviation less than the standard deviation threshold and the diffusion distance less than the preset safety distance is selected; the frequency intersection interval corresponding to the dust of each particle size segment is analyzed to obtain a frequency set containing multiple sub-intervals; for the frequency set of the multiple sub-intervals, combined with the target control concentration threshold table, a frequency-concentration response analysis is performed; the polynomial regression method is used to establish a mapping relationship between frequency and dust concentration, and a relationship between concentration C and frequency f is fitted; the dust concentration at different frequencies is calculated using the relationship; the frequency interval that makes the dust concentration of each particle size segment closest to the target control concentration is selected, and the optimal frequency range is determined.

[0034] Specifically, according to the national standards for different cleanliness levels, target control concentration data sets for Class 100, Class 1000 and Class 10000 clean rooms are established respectively. The cubic spline interpolation algorithm is used to interpolate the particle size segments that are not clearly specified in the standard to obtain a complete particle size-target concentration correspondence data set. For key particle size segments such as 0.1μm, 0.3μm, 0.5μm, 1μm, and 5μm, a target concentration threshold table is established as a constraint condition for the subsequent optimization process. Using the NSGA-II multi-objective optimization algorithm, the settling uniformity and diffusion distance constraints are used as optimization objective functions to construct an optimization model. The air conditioning fan rotation frequency is used as the decision variable, and the upper and lower limits of the frequency are set to 20Hz-60Hz. The settling rate and diffusion distance data of dust of different particle sizes are substituted into the model, and the population size is set to 100, the evolutionary generation is 200, the crossover probability is 0.9, and the mutation probability is 0.1. The Pareto optimal solution set that meets multiple objectives is solved. The Pareto optimal solution set is screened, and the standard deviation threshold required for sedimentation uniformity is set to 0.1. The solution that satisfies both the standard deviation less than 0.1 and the diffusion distance less than the preset safety distance is selected. The frequency intersection interval corresponding to the dust of each particle size segment is determined by interval intersection operation. The specific method is to compare the frequency range of each particle size segment one by one and take their intersection. These intervals are used as candidate optimal control ranges to obtain a frequency set containing multiple sub-intervals. For the candidate optimal control range, combined with the target control concentration threshold, frequency-concentration response analysis is performed. The polynomial regression method is used to establish the mapping relationship between frequency and dust concentration. For each particle size segment, the relationship between concentration C and frequency f is fitted, C=af³+bf²+cf+d, where a is the coefficient of the cubic term, which determines the shape and direction of the curve, b is the coefficient of the quadratic term, which affects the concavity of the curve, c is the coefficient of the linear term, which determines the slope of the curve, and d is the constant term, which determines the vertical position of the curve. The dust concentration at different frequencies is calculated using the fitted polynomial function. The frequency interval that makes the dust concentration of each particle size segment closest to the target control concentration is selected, and the optimal frequency range is determined by minimizing the mean square error. Finally, the optimal control range of the air-conditioning fan rotation frequency is output, and the optimal ranges of different particle size segments are compared and analyzed. During the implementation process, a target control concentration data set is first established for a Class 100 clean room, such as 3520 particles / m³ for a particle size of 0.1μm, 1000 particles / m³ for 0.3μm, and 352 particles / m³ for 0.5μm. Using the cubic spline interpolation algorithm, the target concentration of a particle size of 0.2μm is calculated to be 2260 particles / m³. The NSGA-II algorithm is used for multi-objective optimization, with the population size set to 100 and the evolutionary generation set to 200. The optimization results show that for 0.5μm dust particles, the frequency range that meets the requirements is 35-50Hz.The standard deviation threshold of sedimentation uniformity is set to 0.1, and the common frequency range of the 0.1-1μm particle size segment is obtained by interval intersection operation as 38-47Hz. The frequency-concentration mapping relationship is established using third-order polynomial regression. For example, for a particle size of 0.5μm, C=0.02f³-2.5f²+98f-1200 is obtained, where C is the concentration (pieces / m³) and f is the frequency (Hz). Among them, a is the coefficient of the cubic term, which determines the shape and direction of the curve, and b is the coefficient of the quadratic term, which affects the concavity of the curve. c is the coefficient of the linear term, which determines the slope of the curve. d is the constant term, which determines the vertical position of the curve. By minimizing the mean square error, the final optimal frequency range is determined to be 40-45Hz. Comparative analysis shows that the optimal frequency range for smaller particle sizes (0.1-0.3μm) is wider (38-48Hz), while the range for larger particle sizes (1-5μm) is narrower (41-44Hz), reflecting the differentiated requirements of different particle sizes for air conditioning control. Finally, considering all particle size segments comprehensively, the optimal control range of the air conditioning fan rotation frequency is 41-44Hz, which can simultaneously meet the control requirements of dust in each particle size segment.

[0035] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for coordinated optimization control of workshop cleanliness and air conditioning fan rotation frequency, characterized in that: The method comprises: Sampling and analyzing dust of different particle size segments in workshops with different cleanliness levels, obtaining the mass fraction data and particle size distribution characteristic parameters of dust in each particle size segment, and obtaining dust particle size distribution data; Track the movement trajectory of dust in different particle size segments and obtain the dust settling rate and diffusion distance data, including recording the dust falling process to calculate the settling rate, obtaining the dust concentration at different locations to determine the diffusion distance, and combining the dust particle size distribution data to establish the relationship between the dust particle size and the settling rate and diffusion distance under different cleanliness levels; By placing multiple hot-bulb anemometers in workshops with different cleanliness levels to synchronously measure the air flow velocity and turbulence intensity in different areas, the mapping relationship between the air conditioning fan rotation frequency and the air flow velocity and turbulence intensity is analyzed to obtain the air flow velocity and turbulence intensity distribution under different air conditioning fan rotation frequencies; The settling rates of dust particles of different particle sizes are combined with the turbulence intensity distribution under different air-conditioning fan rotation frequencies. Computational fluid dynamics software is used to simulate the dust movement trajectory under different rotation frequencies, and the standard deviation of dust settling distribution is calculated to determine whether the settling uniformity requirements are met. The air-conditioning fan rotation frequency threshold that meets the settling uniformity requirements is obtained. The dust diffusion distances of different particle sizes are combined with the air flow velocity data under different air-conditioning fan rotation frequencies to simulate the dust diffusion process under different air-conditioning fan rotation frequencies. According to different cleanliness level standards, it is judged whether the diffusion distance exceeds the safety range threshold, and the air-conditioning fan rotation frequency threshold that meets the diffusion distance constraint is obtained. For dust of different particle size segments, the target control concentration threshold is determined. Combined with the dust settling rate and diffusion distance data of different particle sizes, the intersection of the air-conditioning fan rotation frequency thresholds that simultaneously meet the constraints of settling uniformity and diffusion distance is obtained, and the optimal control range of the air-conditioning fan rotation frequency is determined. The settling rate of dust of different particle sizes is combined with the turbulence intensity distribution under different air-conditioning fan rotation frequencies, and the dust movement trajectory under different rotation frequencies is simulated by using computational fluid dynamics software, and the standard deviation of dust settling distribution is calculated to determine whether the settling uniformity requirement is met, and the air-conditioning fan rotation frequency threshold that meets the settling uniformity requirement is obtained, including: A particle motion equation is established based on the relationship between the settling velocity and the particle size, and the particle motion equation is used to describe the movement trajectory of the dust particles; Construct a 3D model of the workshop and obtain the geometric structure and surface characteristics of the workshop by setting mesh parameters and boundary conditions; Simulate the airflow field distribution under different rotation frequencies; Combining the airflow field distribution with the particle motion equation to obtain the motion trajectory of the dust particles; Performing statistical analysis on the movement trajectories of the dust particles, and using the statistical analysis results to calculate the spatial coordinates of dust settling distribution; Setting a standard deviation threshold for sedimentation uniformity requirements, if the calculated standard deviation is less than the standard deviation threshold, it is determined that the sedimentation uniformity at the rotation frequency meets the requirements; If the calculated standard deviation is greater than the standard deviation threshold, it is determined that the sedimentation uniformity at the rotation frequency does not meet the requirement; The method combines the dust diffusion distances of different particle sizes with the air flow velocity data under different air conditioner fan rotation frequencies, simulates the dust diffusion process under different air conditioner fan rotation frequencies, and determines whether the diffusion distance exceeds the safety range threshold according to different cleanliness grade standards, and obtains the air conditioner fan rotation frequency threshold that meets the diffusion distance constraint, including: A dust diffusion model is established using a convection diffusion equation, which describes the change of dust concentration over time and space; Numerically solving the dust diffusion model to obtain dust concentration distribution; Constructing a three-dimensional model of a workshop, wherein the three-dimensional model of the workshop sets meshing parameters and sets initial conditions and boundary conditions in combination with the internal structure of the workshop and the location of the dust source; Simulate the airflow field distribution under different rotation frequencies to obtain airflow field data; The airflow field data is combined with the dust concentration distribution to calculate the diffusion process of dust particles with different particle sizes at each rotation frequency; Calculate the diffusion distance when the dust concentration reaches half of the initial value, and establish a functional relationship between the diffusion distance and the rotation frequency of the air conditioner fan; The functional relationship is fitted using the least square method to obtain a trend curve of diffusion distance changing with frequency; According to the standards of different cleanliness levels, set the safe range threshold of the diffusion distance; The frequency range boundary that meets the requirements is searched by binary search, and the air conditioner fan rotation frequency threshold that meets the diffusion distance constraint is obtained; The target control concentration threshold for dust of different particle size segments is determined, and the settling rate and diffusion distance data of dust of different particle sizes are combined to obtain the intersection of the air-conditioning fan rotation frequency thresholds that simultaneously meet the constraints of settling uniformity and diffusion distance, and determine the optimal control range of the air-conditioning fan rotation frequency, including: Obtain the target control concentration data of clean rooms of different cleanliness levels, perform interpolation calculations on particle size segments that are not clearly specified, and obtain a complete particle size-target concentration correspondence data set; According to the complete particle size-target concentration correspondence data set, a target concentration threshold table including key particle size segments is established; An optimization model is constructed, wherein the settlement uniformity and diffusion distance constraints are used as optimization objective functions, and the air conditioning fan rotation frequency is used as a decision variable; Screen the Pareto optimal solution set, set the standard deviation threshold required for sedimentation uniformity, and select solutions that meet the standard deviation less than the standard deviation threshold and the diffusion distance less than the preset safety distance; Analyze the frequency intersection intervals corresponding to the dust in each particle size segment to obtain a frequency set containing multiple sub-intervals; For the frequency sets of the multiple sub-intervals, combined with the target control concentration threshold table, frequency-concentration response analysis is performed; The mapping relationship between frequency and dust concentration was established by using polynomial regression method, and the relationship between concentration C and frequency f was obtained by fitting. Using the relationship, calculate the dust concentration at different frequencies; Select the frequency interval that makes the dust concentration in each particle size segment closest to the target control concentration, and determine the optimal frequency range.

2. The method according to claim 1, wherein: The dust of different particle size segments in workshops with different cleanliness levels is sampled and analyzed to obtain the mass fraction data and particle size distribution characteristic parameters of the dust in each particle size segment, and obtain the dust particle size distribution data, including: A laser particle size analyzer is used to sample the air in workshops of different cleanliness levels to obtain the scattered light intensity signal of dust particles; The particle size of the dust particles is calculated based on the corresponding relationship between the scattered light intensity and the particle size, and the dust quantity distribution in each particle size segment is obtained by classification and statistics based on the preset particle size segments; The quantity distribution is converted into a mass distribution, and the mass fraction data of dust in each particle size segment is obtained by integral calculation to obtain the mass concentration distribution characteristics of dust in each workshop; A comparative analysis was conducted on the dust particle size distribution data of workshops with different cleanliness levels, and the difference coefficient of dust mass fraction in each particle size segment was calculated. The K-means clustering algorithm was used to classify the dust characteristics of different workshops, and the cumulative distribution diagram and frequency distribution diagram of dust particle size were drawn.

3. The method according to claim 1, wherein: The tracking of the movement trajectory of dust in different particle size segments and obtaining the dust settling rate and diffusion distance data include recording the dust falling process to calculate the settling rate, obtaining the dust concentration at different locations to determine the diffusion distance, and combining the dust particle size distribution data to establish the relationship between the dust particle size and the settling rate and diffusion distance under different cleanliness levels, including: Obtain a dust falling video sequence, and use the OpenCV image processing library to perform frame analysis on the video sequence; According to the frame analysis results, dust particles of different particle size segments are identified, and position changes of the dust particles between adjacent frames are extracted; According to the position change and the time interval between adjacent frames, the instantaneous velocity of the dust particles is calculated, and multiple dust particles in the same particle size segment are counted to obtain the average settling rate of the dust in the particle size segment; In workshops with different cleanliness levels, multiple samplers are arranged in the vertical and horizontal directions, and the dust concentration at different positions is measured by the samplers to calculate the spatial distribution of the dust concentration; According to the dust concentration spatial distribution and dust particle size distribution data, a relationship model between dust particle size and sedimentation rate is established, and the relationship model is used to obtain the quantitative relationship between dust particle size and sedimentation rate under different cleanliness levels; In combination with the dust particle size distribution data, a relationship model between dust particle size and diffusion distance is established to determine the quantitative relationship between dust particle size and diffusion distance under different cleanliness levels.

4. The method according to claim 1, wherein: The method comprises: arranging multiple hot-bulb anemometers in workshops of different cleanliness levels to synchronously measure the air flow velocity and turbulence intensity in different areas, analyzing the mapping relationship between the air conditioning fan rotation frequency and the air flow velocity and turbulence intensity, and obtaining the air flow velocity and turbulence intensity distribution under different air conditioning fan rotation frequencies, including: Obtaining the positions of key measurement points in the workshop, where the positions of key measurement points are obtained by pre-simulating airflow distribution using computational fluid dynamics software; Arrange multiple hot-bulb anemometers according to the key measurement point positions to synchronously collect airflow velocity and turbulence intensity data of the hot-bulb anemometers; Obtain air flow velocity and turbulence intensity data at different air conditioning fan rotation frequencies; A support vector regression algorithm is used to establish a mapping relationship between the rotation frequency of the air conditioner fan and the air flow velocity and turbulence intensity, wherein the mapping relationship selects the fan rotation frequency as an independent variable and the air flow velocity and turbulence intensity as dependent variables; The airflow velocity and turbulence intensity values ​​of each measuring point under multiple air-conditioning fan rotation frequencies are calculated. Combined with the spatial coordinate information of each measuring point, the entire workshop space is interpolated to obtain the spatial distribution data of airflow velocity and turbulence intensity under different air-conditioning fan rotation frequencies.

Images