Method for dividing constant temperature and humidity dust-free workshop into different cleanliness grade areas

Abstract

The application discloses a method for dividing different cleanliness grade areas in a constant-temperature and constant-humidity dust-free workshop in the field of information technology, and comprises the following steps: preliminarily dividing different cleanliness grade areas according to production process requirements and functions of each area, obtaining a region division scheme by using a space division algorithm, including space position coordinates, area and cleanliness grade requirement parameters of each area; setting a buffer zone between different cleanliness grade areas, determining the spatial layout of the buffer zone, including position coordinates, area and shape parameters, by calculating the space position coordinates and area of each area, and determining the cleanliness grade transition requirements of the buffer zone according to the cleanliness grade difference between adjacent areas; analyzing airflow organization and pressure difference control conditions under the optimized region division scheme, obtaining airflow velocity, airflow direction and pressure difference parameters of each area, and obtaining an airflow organization and pressure difference control scheme meeting the cleanliness grade requirements of each area by using a gradient descent iterative optimization algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, in particular to a method for dividing different cleanliness level areas in a constant temperature and humidity dust-free workshop. BACKGROUND

[0002] In the reconstruction project of a constant temperature and humidity dust-free clean production factory workshop, reasonable division of different cleanliness level areas is a key technical problem. The core production area requires a higher cleanliness level, while the auxiliary area can be set to a lower cleanliness level.

[0003] However, the division of areas is not simply based on function, but also needs to consider factors such as material flow and personnel flow between areas to avoid cross-contamination. At the same time, buffer zones need to be set up between different cleanliness level areas for effective isolation and transition. In addition, when dividing areas, factors such as spatial layout, air flow organization, and pressure difference control of each area need to be considered to ensure that the cleanliness level of each area can be stably maintained. Unreasonable division of areas may result in substandard cleanliness level in local areas or a decrease in the cleanliness level of the entire workshop, affecting product quality. Therefore, how to meet the production process requirements while taking into account the reconstruction cost and operation efficiency to make reasonable division of areas is a complex technical problem that requires comprehensive consideration of many factors and balancing of multiple contradictions to obtain the optimal solution. SUMMARY

[0004] The present application provides a method for dividing different cleanliness level areas in a constant temperature and humidity dust-free workshop, which comprises the following steps:

[0005] Step S101, according to the production process requirements and the functions of each area, preliminarily divide different cleanliness level areas, use a spatial division algorithm to obtain a region division scheme, including the spatial position coordinates, area and cleanliness level requirement parameters of each area;

[0006] Step S102, obtain material flow and personnel flow data, analyze the data through a K-means clustering algorithm to obtain the movement trajectory and frequency of material flow and personnel flow, and determine whether there are material flow and personnel flow cross-regions in the preliminary region division scheme. If there are, adjust the region division scheme according to the characteristics of material flow and personnel flow;

[0007] Step S103, set up a buffer zone between different cleanliness level areas, determine the spatial layout of the buffer zone, including position coordinates, area and shape parameters, by calculating the spatial position coordinates and area of each area, and determine the cleanliness level transition requirements of the buffer zone according to the cleanliness level difference between adjacent areas;

[0008] Step S104, the finite element method is used to carry out the computational fluid dynamics simulation, the airflow organization and the pressure difference control situation under the regional division scheme after optimization are analyzed, the airflow velocity, the airflow direction and the pressure difference parameter of each region are obtained, the airflow organization and the pressure difference control scheme meeting the cleanliness grade requirement of each region are obtained through the gradient descent iterative optimization algorithm;

[0009] Step S105, according to the regional division scheme, the airflow organization and the pressure difference control scheme after optimization, the engineering quantity list method is used to carry out the engineering cost estimation, the construction cost and the operation cost of the clean room reconstruction project are measured, the cost benefit analysis is carried out through the net present value method, the reconstruction cost and the operation efficiency are weighed, the regional division scheme is further optimized;

[0010] Step S106, the regional division scheme, the airflow organization and the pressure difference control scheme and the buffer zone setting scheme after optimization are input into the AutoCAD design software, the three-dimensional model and the construction drawing are generated, the Monte Carlo simulation method is used to simulate the operation of the clean room, the overall cleanliness grade of the clean room is verified, whether the clean production requirement is met is judged, if not, the airflow organization and the pressure difference control step are returned for further optimization;

[0011] Step S107, based on the results of all the optimization steps, the final clean room regional division scheme, the airflow organization and the pressure difference control scheme, the buffer zone setting scheme and the construction drawing clean room design scheme file are output, used to guide the construction and operation of the clean room Maintenance, ensure that the cleanliness grade of each region is stable and controllable, avoid cross contamination, realize efficient clean production.

[0012] The technical scheme provided by the embodiment of the application can include the following beneficial effects:

[0013] The application discloses a method for dividing different cleanliness level areas in a constant temperature and humidity dust-free workshop. To solve the problem of cleanliness level requirement difference and cross contamination risk in different areas in clean production, the application firstly uses a space division algorithm to preliminarily plan the area layout, combines material flow and personnel flow data of K-means clustering analysis, and dynamically adjusts the layout to avoid cross contamination. For the transition between different cleanliness level areas, the spatial layout of the buffer zone and the cleanliness level transition requirement are determined by calculation. Then, the finite element method is used for computational fluid dynamics simulation, combined with gradient descent iterative optimization algorithm, to realize the airflow organization and pressure difference control that meet the cleanliness level requirements of each area. And through the bill of quantities method and net present value method, the cost benefit analysis is carried out to further optimize the area division scheme. Finally, the optimized scheme is input into AutoCAD for three-dimensional modeling and construction drawing generation, and the overall cleanliness is verified by Monte Carlo simulation to ensure that the production requirements are met. The application realizes the collaborative optimization of clean room area division, airflow control, buffer zone setting and cost control, and outputs a complete design scheme including area division, airflow organization, pressure difference control, buffer zone setting and construction drawing, which ensures the stable operation and efficient clean production of the clean room. BRIEF DESCRIPTION OF DRAWINGS

[0014] Fig. 1 The flowchart of the method for dividing different cleanliness level areas in a constant temperature and humidity dust-free workshop of the application.

[0015] Fig. 2 The schematic diagram of the method for dividing different cleanliness level areas in a constant temperature and humidity dust-free workshop of the application.

[0016] Fig. 3 The schematic diagram of the method for dividing different cleanliness level areas in a constant temperature and humidity dust-free workshop of the application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the application will be described clearly and detailedly below with reference to the drawings in the embodiments of the application. The described embodiments are only some of the embodiments of the application.

[0018] As Figs. 1-3 , the method for dividing different cleanliness level areas in a constant temperature and humidity dust-free workshop of the embodiment can specifically include:

[0019] Step S101, according to the production process requirement and the function of each area, preliminarily divide different cleanliness level areas, use a space division algorithm to obtain an area division scheme, including the spatial position coordinates, area and cleanliness level requirement parameters of each area.

[0020] The step S101 further comprises: constructing an initial region division model according to preset production process parameters and function requirements of each region, determining a preliminary function classification and a cleanliness level requirement of each region of the initial region division model; inputting the initial region division model by using a space division algorithm, calculating spatial position coordinates and area parameters of each region of the initial region division model, and obtaining a preliminary region division scheme; determining a region cleanliness level adjustment scheme in combination with the cleanliness level requirement parameters for the preliminary region division scheme, performing a region cleanliness level refinement adjustment on the preliminary region division scheme according to the region cleanliness level adjustment scheme, and obtaining a region division scheme after cleanliness adjustment; recalculating the spatial position coordinates and the area parameters of each region by using the region function classification of the preliminary region division scheme and the cleanliness level data of the region division scheme after cleanliness adjustment, and obtaining an optimized region division scheme; if there is a difference between adjacent region cleanliness levels in the optimized region division scheme that is greater than a preset cleanliness level difference threshold, a buffer zone setting method is used to increase a transition region, and a region division scheme containing the transition region is obtained, which is used to balance the difference between the cleanliness levels; whether the optimized region division scheme or the region division scheme containing the transition region meets the production process and function requirements is determined by using a preset region division scheme threshold, if not, the initial region division model is adjusted and recalculated, and if yes, a region division scheme that meets the production process and function requirements is determined, and the spatial position coordinates, the area parameters, and the cleanliness level requirements of each region of the region division scheme that meets the production process and function requirements are output, and a region division report is obtained.

[0021] Specifically, in constructing an efficient production environment, regional division is a crucial step. First, according to the preset production process parameters and the functional requirements of each region, an initial regional division model is constructed. For example, in a pharmaceutical factory, it may be necessary to divide raw material storage area, production area, quality inspection area, and finished product storage area. The raw material storage area requires a lower cleanliness level, while the production area and quality inspection area require a high cleanliness level to ensure product quality. Next, through a space division algorithm, the initial regional division model is input, and the spatial position coordinates and area parameters of each region are calculated to generate a preliminary regional division scheme. Assuming the total area of the factory is 10000 square meters, the algorithm preliminarily allocates 2000 square meters for the raw material storage area, 5000 square meters for the production area, 1500 square meters for the quality inspection area, and 1500 square meters for the finished product storage area. The algorithm also determines the specific location of each region, such as the production area located in the center of the factory to facilitate logistics and personnel flow. Based on the preliminary regional division scheme, combined with the cleanliness level requirement parameters, the regional cleanliness level is refined and adjusted. For example, the production area may need to be further divided into high-cleanliness operation area and general-cleanliness preparation area. The high-cleanliness operation area may need to reach ISO5 cleanliness, while the general-cleanliness preparation area reaches ISO7. Through adjustment, the cleanliness of each region is ensured to meet the production process standards. Using the data of regional function classification and cleanliness level adjustment, the spatial position coordinates and area parameters of each region are recalculated to optimize the regional division scheme. Assuming that air purification equipment needs to be added in the high-cleanliness operation area, an additional 200 square meters of space may be required, so the total area of the production area is adjusted to 5200 square meters, and the areas of other regions are adjusted accordingly to ensure that the total area remains unchanged. If the optimized regional division scheme has a large difference in cleanliness level between adjacent regions, a buffer zone setting method is used to increase the transition area to balance the cleanliness level difference. For example, a buffer zone (ISO6 level) with an area of about 200 square meters is set between the high-cleanliness operation area (ISO5 level) and the general-cleanliness preparation area (ISO7 level) to reduce the pollution risk caused by air flow. The optimized regional division scheme is judged by the preset threshold whether it meets the production process and functional requirements. Assuming that the preset threshold is that the cleanliness level difference between each region does not exceed 2 levels, if it does not meet the requirement, the initial regional division model is returned for recalculation and adjustment. For example, if it is found that the cleanliness difference between the finished product storage area and the production area exceeds 2 levels, the cleanliness level of the finished product storage area needs to be adjusted or a buffer zone needs to be added. Finally, the regional division scheme that meets the production process and functional requirements is determined, and the spatial position coordinates, area parameters, and cleanliness level requirements of each region are output to form a complete regional division report.For example, the report details that the raw material storage area is located in the northwest corner of the factory, covering an area of 2000 square meters, with a cleanliness level of ISO8; the production area is located in the center, covering an area of 5200 square meters, including ISO5 and ISO7 level areas; the quality inspection area is located in the southeast corner, covering an area of 1500 square meters, with a cleanliness level of ISO6; the finished product storage area is located in the southwest corner, covering an area of 1500 square meters, with a cleanliness level of ISO8, and a 200 square meter buffer area is set between the production area and the finished product storage area, with a cleanliness level of ISO7. This method of dividing the area not only ensures the functionality and cleanliness requirements of each area, but also reduces the risk of pollution and improves production efficiency through reasonable space layout and buffer zone setting. Through preset thresholds and multiple optimization adjustments, the feasibility and scientificity of the area division scheme are ensured, and the final area division report provides an important basis for factory construction and operation. In the specific implementation process, the space division algorithm may use an optimization method based on genetic algorithm or ant colony algorithm, which simulates natural selection or ant foraging behavior to find the optimal space layout scheme. The refinement of cleanliness levels may be based on air flow simulation results to ensure that high cleanliness areas are not affected by low cleanliness areas. The setting of the buffer zone is based on the principles of fluid mechanics, calculating the air flow path and speed to determine the best location and area of the buffer zone. Through these scientific methods and meticulous adjustments, the area division scheme not only meets the production process and functional requirements, but also improves the overall cleanliness and operational efficiency of the production environment, laying a solid foundation for the efficient operation of the enterprise. This systematic area division method has been widely verified in practical applications and can effectively improve the quality and management level of the production environment.

[0022] In step S102, material flow and personnel flow data are obtained, and K-means clustering algorithm is used for analysis to obtain the moving track and frequency of the material flow and personnel flow. It is judged whether there is a cross-region of material flow and personnel flow in the preliminary area division scheme. If there is, the area division scheme is optimized and adjusted according to the characteristics of the material flow and personnel flow.

[0023] The step S102 further comprises: obtaining a material flow regular data set and a personnel flow regular data set, and using a K-means clustering algorithm for clustering analysis to obtain a material flow clustering result and a personnel flow clustering result respectively; the clustering result contains track and frequency information; obtaining an area division data set, and performing spatial superposition analysis on the material flow clustering result and the personnel flow clustering result with the area division data set respectively to determine whether the material flow track and the personnel flow track fall into a preset area to obtain a correlation; if there are correlated material flow clusters and correlated personnel flow clusters in the same area, the area is determined as a cross-region, the material flow characteristics and personnel flow characteristics in the cross-region are analyzed, and if there is a safety conflict between the material flow characteristics and the personnel flow characteristics, the area division optimization adjustment is triggered.

[0024] Specifically, first, deploying a sensor network and a positioning system is the basis of the entire process. For example, in a large manufacturing workshop, multiple RFID tags and UWB positioning base stations are installed to collect real-time location information of materials and personnel. Assuming that the material flow raw data set contains timestamp, location coordinates (such as (x1, y1, z1)), and other information, the personnel flow raw data set also contains similar information. These data provide the basis for subsequent analysis. Next, the raw data set is preprocessed. Taking the material flow data as an example, assume that the data points in a certain period of time are (t1, x1, y1, z1), (t2, x2, y2, z2), etc. In the preprocessing process, first, data cleaning is performed to remove invalid data points caused by sensor failure, such as (t3, NaN, NaN, NaN). Then, data deduplication is performed to remove duplicate records, such as (t4, x4, y4, z4) and (t4, x4, y4, z4). For missing values, interpolation is used to fill in, such as missing z coordinates for a data point, then the z value is estimated based on the previous and subsequent data points. Finally, time series smoothing is performed to smooth the data using the moving average method to reduce noise. In the clustering analysis stage, the K-means algorithm is used to process the normalized data set. Assuming that the number of material flow clustering clusters k1 is 3, representing the raw material area, processing area, and finished product area respectively. By calculating the distance of each data point to the cluster center, the data points are assigned to the nearest cluster. For example, data point (t5, x5, y5, z5) is assigned to the raw material area cluster because it is closest to the cluster center. Iteratively update the cluster center until the center position is stable, and obtain the material flow trajectory and frequency information of each cluster. After generating the region division data set, spatial superposition analysis is performed. Assuming that the pre-set regions include A, B, C three regions, the boundary coordinates are (A1, A2, A3, …), (B1, B2, B3, …), etc. Superimpose the trajectory information of the material flow cluster with the region boundary information to determine whether the trajectory falls into a certain region. For example, the trajectory (x6, y6, z6) of a certain material flow cluster falls into region A, so mark the cluster associated with region A. When judging the cross-region, all pre-set regions are traversed. Assuming that region A is associated with both material flow cluster M1 and personnel flow cluster P1, then region A is determined as a cross-region. Generate a cross-region data set containing the identification information of region A and the associated cluster identification. For the cross-region, analyze the material flow and personnel flow characteristics. For example, the material flow characteristics in region A include raw material type, daily throughput of 100 tons, use of forklift transportation, and medium-level danger rating. The personnel flow characteristics include 50 people per day, 20 times per hour, assembly type, and low-level safety risk rating. If the material danger rating and personnel safety risk rating are both higher than the pre-set safety threshold, then a safety conflict is determined. When generating the region division optimization adjustment scheme, adjustments are made based on the characteristic analysis results.For example, adjust the boundaries of region A, and divide out part of the high-risk area; add physical barriers to reduce the risk of interaction between material flow and personnel flow; change the material transportation path to avoid personnel-intensive areas; adjust the personnel route to avoid high-risk material areas. The updated regional division scheme is fed back to the regional division dataset. Through these steps, the intersection area of material flow and personnel flow can be effectively identified and managed, reducing safety risks and improving production efficiency. For example, the optimized regional division scheme reduces congestion during material transportation, improving material flow efficiency; at the same time, through isolation measures, the risk of personnel injury is reduced, and the overall safety management level is improved. The entire process not only relies on data analysis and algorithm application, but also needs to be combined with actual production environment and safety management needs, and flexible adjustment and optimization. In this way, scientific division and efficient management of production areas can be achieved to ensure smooth production process and personnel safety.

[0025] In step S103, a buffer zone is set between different cleanliness level regions. The spatial layout of the buffer zone, including position coordinates, area, and shape parameters, is determined by calculating the spatial position coordinates and area of each region. The cleanliness level transition requirements of the buffer zone are determined according to the cleanliness level difference between adjacent regions.

[0026] The boundary coordinate data of each clean region is obtained, and the geometric center coordinates and area of each region are calculated according to the boundary coordinate data. The adjacency relationship of each region is determined according to the geometric center coordinates and area data. The shape and boundary of the buffer region are initially generated according to the adjacency relationship and the geometric shape of each region. If the buffer region overlaps with existing obstacles, the shape or position of the buffer region is adjusted to obtain the adjusted buffer region shape and boundary. The cleanliness level difference is determined according to the cleanliness level of adjacent regions. If the cleanliness level difference is greater than a preset level threshold, the cleanliness level of the buffer zone is increased to obtain the target cleanliness level of the buffer zone.

[0027] Specifically, the boundary coordinate data of each clean area is obtained, the geometric center coordinates of each area are calculated, and the area of each area is calculated according to the boundary coordinate data. For example, assuming that a clean room is divided into three areas A, B and C, the boundary coordinates of area A are {(0, 0), (10, 0), (10, 5), (0, 5)}, the geometric center coordinates are (5, 2.5) obtained by calculation, and the area is 50 square meters. Similarly, the geometric center and area of each area are calculated. According to these data, it can be preliminarily determined that A and B, and B and C are adjacent areas, so as to obtain the region adjacency relationship. According to the preset minimum width of the buffer area and the safety distance parameters, as well as the region adjacency relationship and the geometric shape of each region, the shape and boundary of the buffer area are preliminarily generated. Assuming that the minimum buffer area width is 1 meter, the safety distance is 0.5 meters, and A and B are adjacent, a buffer area with a width of 1.5 meters is generated between A and B. If the buffer area overlaps with existing obstacles such as pillars or equipment, the shape or position of the buffer area is adjusted to avoid the obstacles, and the adjusted buffer area shape and boundary are obtained. For example, the original buffer area boundary is {(10, 0),

[0028] (11.5,0), (11.5,5), (10,5)}, and the adjusted boundary can become {(10,0), (11.5,0), (11.5,4), (10,4)} due to the overlap with the column. According to the adjusted boundary, the area, center coordinates, and shape parameters of the buffer zone are calculated. The cleanliness level difference is determined according to the cleanliness levels of adjacent zones. Assuming that Zone A is ISO5 level and Zone B is ISO7 level, and the preset level threshold is 2 levels, the cleanliness level difference between A and B is greater than the threshold. According to the preset buffer zone cleanliness level promotion rule, the buffer zone cleanliness level is promoted to ISO6 level. If the difference is less than the threshold, the buffer zone cleanliness level remains consistent with the lower cleanliness level of the adjacent zone. The purpose of this is to ensure smooth cleanliness transition and prevent contamination spread. The boundary coordinates of each zone and the buffer zone are converted to a unified spatial coordinate system using a spatial coordinate conversion algorithm. For example, using the Cartesian coordinate system, all zone coordinates are uniformly converted to ensure accurate relative positions of each zone and buffer zone. According to the coordinates in the unified coordinate system, the overall spatial layout diagram containing each clean zone and buffer zone is generated, which visually displays the distribution of each zone and the setting of the buffer zone. According to the shape parameters of the buffer zone, the type of the buffer zone is determined. If the buffer zone is a regular shape such as a rectangle, its length, width, and other geometric parameters are directly extracted. For example, a rectangular buffer zone has a length of 1.5 meters and a width of 5 meters. If the buffer zone is irregular, a polygon approximation algorithm or curve fitting algorithm is used to calculate its approximate geometric parameters. For example, an irregular buffer zone is approximated by a polygon, resulting in an approximate rectangular parameter with a length of 1.4 meters and a width of 4.8 meters. The purpose of this is to simplify the subsequent design and analysis process. According to the location coordinates, area, shape parameters, and cleanliness level of the buffer zone, the attribute information of the buffer zone is generated. For example, the attribute information of a buffer zone may include: location coordinates {(10,0), (11.5,0),

[0029] (11.5,4), (10,4)} with an area of 6 square meters and shape parameters (rectangle, 1.5 meters long and 4 meters wide). The cleanliness level is ISO 6. These attribute information are associated with the attribute information of each clean area to form the topological relationship between the clean area and the buffer zone. Data visualization techniques are used to mark the cleanliness level of each area and the attribute information of the buffer zone on the overall space layout diagram to generate a visualized space layout diagram, which is convenient for designers and managers to intuitively understand the layout. It is determined whether there are areas that do not meet the cleanliness level transition requirements in the visualized space layout diagram. For example, if it is found that the cleanliness level of a buffer zone differs greatly from that of the adjacent area, it may increase the risk of contamination. At this time, the cleanliness level of the buffer zone needs to be adjusted or the buffer zone layout needs to be re-planned to ensure smooth transition of cleanliness. If all areas meet the requirements, the final visualized space layout diagram and buffer zone attribute information are output, the clean room buffer zone layout design is completed, and the final buffer zone layout scheme is obtained. The purpose of this is to ensure that the environmental control in the clean room meets the design standards and guarantees the stability of the production process and the quality of the product. Through the above steps, not only can the clean room buffer zone layout be scientifically and reasonably designed, but also the risk of contamination can be effectively reduced, and the production efficiency and product quality can be improved. Each step is closely linked to ensure the feasibility and practicality of the design scheme.

[0030] In step S104, a computational fluid dynamics simulation is performed using the finite element method to analyze the air flow organization and pressure difference control under the optimized regional division scheme, and to obtain the air flow velocity, air flow direction and pressure difference parameters of each region. Through the gradient descent iterative optimization algorithm, an air flow organization and pressure difference control scheme that meets the cleanliness level requirements of each region is obtained.

[0031] The step S104 further comprises: constructing the calculation domain of the computational fluid dynamics simulation by using the finite element method according to the three-dimensional model of the building and the area division scheme; setting the boundary conditions of the calculation domain, wherein the boundary conditions include physical parameters such as inlet wind speed, outlet pressure, and wall conditions; dividing the grid of the calculation domain, wherein the density of the grid is adaptively adjusted according to the intensity of airflow variation; and obtaining the grid data of the finite element calculation model. The parameters of the gradient descent iterative optimization algorithm are initialized, wherein the parameters include learning rate, iteration number, and convergence threshold; the objective function of the airflow organization and pressure difference control is set, wherein the objective function defines the expected value and the allowable deviation range of the airflow velocity, the airflow direction, and the pressure difference parameter of each area according to the cleanliness level requirement of each area; and the initial airflow organization and pressure difference control scheme is obtained as the starting point of iteration. Based on the current airflow organization and pressure difference control scheme, the computational fluid dynamics simulation is performed; the finite element model is solved to obtain the airflow velocity field, the airflow direction field, and the pressure field of each area; the airflow velocity, the airflow direction, and the pressure difference parameter of each area are extracted from the solving result, wherein the airflow velocity includes x, y, and z direction components, the airflow direction is represented by a vector, and the pressure difference parameter represents the pressure difference between different areas. The objective function value corresponding to the current airflow organization and pressure difference control scheme is calculated; if the objective function value meets the cleanliness level requirement of each area, or the convergence threshold is reached, or the iteration number reaches the upper limit, the iteration is stopped, and the current scheme is determined as the optimized scheme; if the stopping iteration condition is not met, the gradient of the objective function with respect to each parameter of the airflow organization and pressure difference control scheme is calculated according to the calculated airflow velocity, airflow direction, and pressure difference parameter, wherein the gradient represents the direction in which the objective function value changes fastest and is used to guide the adjustment of the parameters. The gradient descent iterative optimization algorithm updates each parameter of the airflow organization and pressure difference control scheme according to the gradient information to generate a new airflow organization and pressure difference control scheme; when updating the scheme, the variation amplitude of the parameters is adjusted according to the size of the learning rate, and the larger the learning rate, the faster the parameter changes; it is judged whether the updated airflow organization and pressure difference control scheme meets the physical constraint condition, wherein the physical constraint condition includes that the wind speed cannot exceed a certain maximum value and the pressure difference cannot exceed a certain maximum value; if the constraint condition is not met, the scheme is adjusted to meet the constraint condition; after the scheme adjustment is completed, the corrected airflow organization and pressure difference control scheme is obtained; and if the scheme meets the constraint condition.Output the airflow organization and pressure difference control scheme meeting the cleanliness level requirement, the scheme including the air speed, air direction of each area air supply port, the setting position of the return air port, and the pressure difference control strategy between areas; generate control instructions for controlling the air conditioning system and ventilation equipment according to the optimized airflow organization and pressure difference control scheme, to realize accurate control of the clean room environment; after the scheme is output, visual display is performed.

[0032] Specifically, in clean room design, it is crucial to ensure that the cleanliness levels of each zone meet the requirements. Here is a detailed analysis and example of the above steps to help understand the implementation method and principles of each step. First, based on the three-dimensional model of the building and the zoning plan, the computational domain for computational fluid dynamics (CFD) simulation is constructed using the finite element method. Assume that a clean room is divided into three zones: high-cleanliness zone, medium-cleanliness zone, and low-cleanliness zone. By using the finite element method, the entire clean room is divided into tens of thousands of grid cells, and the size of each cell is adaptively adjusted according to the severity of airflow changes. For example, near the high-cleanliness zone, the grid density is higher to ensure simulation accuracy; while in the low-cleanliness zone, the grid density is relatively lower to reduce the amount of calculation. Next, the parameters of the gradient descent iterative optimization algorithm are initialized. Set the learning rate to 0.01, the upper limit of the number of iterations to 100, and the convergence threshold to 0.001. The objective function is defined as the sum of the deviations between the expected values and the actual values of the airflow velocity, airflow direction, and pressure difference parameters in each zone. For example, the expected wind speed in the high-cleanliness zone is 0.3 m / s, with a tolerance of ±0.05 m / s; the pressure difference expected value is 10 Pa, with a tolerance of ±2 Pa. In the initial scheme generation stage, according to experience or pre-set rules, set the wind speed and direction of each air supply outlet. For example, set the air supply outlet wind speed in the high-cleanliness zone to 0.4 m / s, and the wind direction towards the workbench; set the air supply outlet wind speed in the medium-cleanliness zone to 0.2 m / s, and the wind direction parallel to the ground. Perform CFD simulation, solve the finite element model, and obtain the airflow velocity field, airflow direction field, and pressure field in each zone. Assume that the simulation results show that the actual wind speed in the high-cleanliness zone is 0.25 m / s, and the pressure difference is 8 Pa; the actual wind speed in the medium-cleanliness zone is 0.15 m / s, and the pressure difference is 5 Pa. Extract the airflow velocity, airflow direction, and pressure difference parameters from these data for subsequent target function calculations. Calculate the target function value of the current scheme. If the target function value is greater than the pre-set convergence threshold, or the number of iterations has not reached the upper limit, continue iterating. For example, the current target function value is 0.02, which is greater than the convergence threshold of 0.001, so further optimization is needed. According to the calculated airflow parameters, calculate the gradient of the target function with respect to each parameter. Assume that the gradient results show that the air supply outlet wind speed in the high-cleanliness zone needs to be increased by 0.05 m / s, and the pressure difference needs to be increased by 2 Pa to reduce the target function value. The gradient descent algorithm updates the airflow organization and pressure difference control scheme based on the gradient information and learning rate. For example, adjust the air supply outlet wind speed in the high-cleanliness zone to 0.45 m / s, and the pressure difference to 12 Pa. Determine whether the updated scheme meets the physical constraint conditions. Assume that the maximum wind speed limit is 0.5 m / s, and the maximum pressure difference limit is 15 Pa. If the wind speed in a certain zone exceeds 0.5 m / s in the updated scheme, adjust it to 0.5 m / s; if the pressure difference exceeds 15 Pa, adjust it to 15 Pa. Ensure that the scheme is physically feasible. Finally, output the airflow organization and pressure difference control scheme that meets the cleanliness level requirements.For example, the air supply outlet in the high-clean zone has a wind speed of 0.45 m / s, the wind direction is towards the workbench, and the pressure difference is 12 Pa. The air supply outlet in the medium-clean zone has a wind speed of 0.2 m / s, the wind direction is parallel to the ground, and the pressure difference is 7 Pa. According to this scheme, control instructions are generated to accurately control the air conditioning system and ventilation equipment. Through the above steps, not only is the cleanliness level of each area ensured to meet the requirements, but also the optimization of airflow organization and pressure difference control is achieved. For example, the adjustment of wind speed and pressure difference in the high-clean zone effectively prevents the intrusion of pollutants; the setting of wind speed and wind direction in the medium-clean zone ensures the uniform distribution of airflow. This optimization method not only improves the operating efficiency of the clean room, but also reduces energy consumption and prolongs the service life of the equipment. In addition, the visual display makes the scheme more intuitive, making it easier for users to view and analyze. For example, through the three-dimensional model and color coding, users can clearly see the airflow speed and pressure difference distribution in each area, and timely discover potential problems and make adjustments. In summary, through the finite element method and gradient descent iterative optimization algorithm, combined with CFD simulation and physical constraint conditions, an airflow organization and pressure difference control scheme that meets the cleanliness level requirements can be efficiently designed, providing scientific basis and technical support for the environmental control of the clean room.

[0033] In step S105, according to the optimized regional division scheme, airflow organization and pressure difference control scheme, the engineering quantity list method is used to estimate the engineering cost, and the construction cost and operation cost of the clean room renovation project are measured. Through the net present value method, the cost-benefit analysis is carried out, the construction cost and operation efficiency are weighed, and the regional division scheme is further optimized.

[0034] The step S105 further comprises: generating an initial regional division scheme according to the architectural structure diagram and functional requirements of the clean room; the scheme divides different cleanliness level areas, and determines the boundaries and areas of each area. According to the regional division scheme, the computational fluid dynamics simulation is used to analyze the airflow speed, flow direction and cleanliness distribution under different airflow organization forms; the best airflow organization scheme is selected, which makes the cleanliness of each key position in the clean room meet the design requirements, and the airflow organization scheme includes the position and number parameters of the air supply outlet and return air outlet. According to the regional division scheme, the airflow organization scheme and the pressure difference control scheme, an engineering quantity list is generated; the estimated value of construction cost and the estimated value of operation cost of each sub-item in the list are calculated, the estimated value of construction cost includes building material cost and equipment cost, and the estimated value of operation cost includes electricity cost and consumable cost.

[0035] Step S106, input the optimized area division scheme, air flow organization and pressure difference control scheme and buffer zone setting scheme into AutoCAD design software to generate a three-dimensional model and construction drawings. The Monte Carlo simulation method is used to simulate the operation of the clean room, verify the overall cleanliness level of the clean room, and determine whether it meets the clean production requirements. If not, return to the air flow organization and pressure difference control step for further optimization.

[0036] The step S106 further comprises: obtaining area division scheme data, air flow organization scheme data, pressure difference control scheme data and buffer zone setting scheme data according to AutoCAD software, generating first three-dimensional model data and the first construction drawing data. Using the Monte Carlo simulation method, input the first three-dimensional model data, the first construction drawing data, the air flow organization scheme data and the pressure difference control scheme data, simulate the operating condition of the clean room, generate the first simulation data set, and determine whether the cleanliness level E is less than or equal to the preset cleanliness level threshold E0; if the cleanliness level E is greater than the preset cleanliness level threshold E0, adjust the air flow organization scheme or the pressure difference control scheme according to the first simulation data set.

[0037] Step S107, based on the results of all the optimization steps, output the final clean room area division scheme, air flow organization and pressure difference control scheme, buffer zone setting scheme and construction drawing clean room design scheme file, which is used to guide the construction and operation of the clean room. Maintenance, ensure the stability and controllability of the cleanliness level of each area, avoid cross contamination, and achieve efficient clean production.

[0038] According to the preset production process and the requirement of product to cleanliness, the clean room functional area is automatically divided by computer aided design software to obtain the clean area layout scheme; the clean area layout scheme is obtained, the air flow distribution in the clean room is simulated based on the computational fluid dynamics model, the air supply port and the return air port position are adjusted, the air flow organization is optimized, and the air flow organization scheme is obtained; according to the air flow organization scheme, the pressure difference distribution of each area of the clean room is calculated by using the finite element analysis method, the pressure difference value between each area is set, and if the pressure difference value meets the design requirement, the pressure difference control scheme is obtained.

[0039] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation using the content of the specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for dividing different cleanliness grade areas in a constant temperature and humidity dust-free workshop, characterized in that, The method comprises the following steps: Step S101, according to the production process requirements and the functions of each region, preliminarily divide the regions of different cleanliness levels, obtain the region division scheme by using a space division algorithm, including the spatial position coordinates, area and cleanliness level requirement parameters of each region; Step S102, obtain the material flow and personnel flow data, analyze the data by using a K-means clustering algorithm, obtain the moving track and frequency of the material flow and personnel flow, judge whether there is a material flow and personnel flow cross region in the preliminary region division scheme, if there is, adjust and optimize the region division scheme according to the characteristics of the material flow and personnel flow; Step S103, set a buffer zone between the regions of different cleanliness levels, determine the spatial layout of the buffer zone by calculating the spatial position coordinates and area of each region, including the position coordinates, area and shape parameters, determine the cleanliness level transition requirements of the buffer zone according to the cleanliness level difference between adjacent regions; Step S104, perform a computational fluid dynamics simulation by using a finite element method, analyze the airflow organization and pressure difference control under the optimized region division scheme, obtain the airflow velocity, airflow direction and pressure difference parameters of each region, obtain the airflow organization and pressure difference control scheme meeting the cleanliness level requirements of each region by using a gradient descent iterative optimization algorithm; Step S105, according to the optimized region division scheme, airflow organization and pressure difference control scheme, perform an engineering cost estimation by using a bill of quantities method, measure the construction cost and operation cost of the clean room reconstruction project, perform a cost-benefit analysis by using a net present value method, balance the reconstruction cost and operation efficiency, and further optimize the region division scheme; Step S106, input the optimized region division scheme, airflow organization and pressure difference control scheme and buffer zone setting scheme into AutoCAD design software, generate a three-dimensional model and construction drawings, simulate the operation of the clean room by using a Monte Carlo simulation method, verify the overall cleanliness level of the clean room, judge whether it meets the clean production requirements, if not, return to the airflow organization and pressure difference control step for further optimization; Step S107, based on the results of all the optimization steps, output the final clean room region division scheme, airflow organization and pressure difference control scheme, buffer zone setting scheme and construction drawing clean room design scheme file, which is used to guide the construction and operation and maintenance of the clean room, ensure the stability and controllability of the cleanliness level of each region, avoid cross contamination, and realize efficient clean production.

2. The method of claim 1, wherein the different cleanliness level zones are divided by the constant temperature and humidity clean room. The step S101 further comprises: According to the preset production process parameters and the function requirements of each region, construct an initial region division model, determine the preliminary function classification and cleanliness level requirements of each region of the initial region division model; Input the initial region division model by using a space division algorithm, calculate the spatial position coordinates and area parameters of each region of the initial region division model, and obtain a preliminary region division scheme; According to the preliminary regional division scheme, a regional cleanliness level adjustment scheme is determined in combination with the cleanliness level requirement parameters, and the preliminary regional division scheme is adjusted in detail in the regional cleanliness level according to the regional cleanliness level adjustment scheme, to obtain a regional division scheme after cleanliness adjustment; The spatial position coordinates and area parameters of each region are recalculated using the regional function classification of the preliminary regional division scheme and the cleanliness level data of the regional division scheme after cleanliness adjustment, to obtain an optimized regional division scheme; If there is a large difference in the cleanliness level between adjacent regions in the optimized regional division scheme, a buffer zone setting method is used to increase a transition region, to obtain a regional division scheme containing a transition region, which is used to balance the difference in cleanliness level; Whether the optimized regional division scheme or the regional division scheme containing a transition region meets the production process and functional requirements is determined by a preset regional division scheme threshold, and if not, the initial regional division model is adjusted and recalculated and adjusted again using the adjusted initial regional division model; If it is satisfied, a regional division scheme that meets the production process and functional requirements is determined, and the spatial position coordinates, area parameters and cleanliness level requirements of each region of the regional division scheme that meets the production process and functional requirements are output, to obtain a regional division report.

3. The method of claim 1, wherein the different cleanliness level zones are divided by the constant temperature and humidity clean room. The step S102 further comprises: Obtain material flow regularity data set and personnel flow regularity data set, and perform clustering analysis using K-means clustering algorithm to obtain material flow clustering results and personnel flow clustering results respectively; The clustering results contain trajectory and frequency information; Obtain regional division data set, and perform spatial superposition analysis on the material flow clustering results and the personnel flow clustering results respectively with the regional division data set, to determine whether the material flow trajectory and the personnel flow trajectory fall into a preset region, to obtain an association relationship; If there are associated material flow clusters and associated personnel flow clusters in the same region, the region is determined as a cross region, and the material flow characteristics and personnel flow characteristics in the cross region are analyzed, and if there is a safety conflict between the material flow characteristics and the personnel flow characteristics, the regional division optimization adjustment is triggered.

4. The method of claim 1, wherein the different cleanliness level zones are divided by the constant temperature and humidity clean room. The step S103 further comprises: Obtain boundary coordinate data of each clean region, and calculate the geometric center coordinates and area of each region according to the boundary coordinate data; Determine the adjacency relationship of each region according to the geometric center coordinates and area data; According to the adjacency relationship and the geometric shape of each region, the shape and boundary of the buffer region are initially generated; If the buffer region overlaps with an existing obstacle, the shape or position of the buffer region is adjusted to obtain an adjusted buffer region shape and boundary; According to the cleanliness level of adjacent regions, the cleanliness level difference is determined; If the cleanliness level difference is greater than a preset level threshold, the cleanliness level of the buffer zone is increased to obtain a target cleanliness level of the buffer zone.

5. The method for dividing different cleanliness grade areas in the constant temperature and humidity dust-free workshop according to any one of claims 1-4, characterized in that, The step S104 further comprises: According to the three-dimensional model of the building and the area division scheme, a computational fluid dynamics simulation calculation domain is constructed by using a finite element method; Boundary conditions of the calculation domain are set, including physical parameters such as inlet wind speed, outlet pressure and wall conditions; The calculation domain grid is divided, and the density of the grid is adaptively adjusted according to the intensity of airflow variation; Grid data of the finite element calculation model is obtained; Parameters of a gradient descent iterative optimization algorithm are initialized, including learning rate, iteration number and convergence threshold; A target function of airflow organization and pressure difference control is set, which defines the expected value and allowable deviation range of airflow velocity, airflow direction and pressure difference parameters of each area according to the cleanliness level requirements of each area; An initial airflow organization and pressure difference control scheme is obtained as the starting point of iteration; Based on the current airflow organization and pressure difference control scheme, a computational fluid dynamics simulation is performed; The finite element model is solved to obtain the airflow velocity field, airflow direction field and pressure field of each area; From the solving results, the airflow velocity, airflow direction and pressure difference parameters of each area are extracted, the airflow velocity includes x, y and z direction components, the airflow direction is represented by a vector, and the pressure difference parameter represents the pressure difference between different areas; The target function value corresponding to the current airflow organization and pressure difference control scheme is calculated; If the target function value meets the cleanliness level requirements of each area, or reaches the preset convergence threshold, or the iteration number reaches the upper limit, the iteration is stopped, and the current scheme is determined as the optimized scheme; If the iteration stopping condition is not met, the gradient of the target function with respect to the airflow organization and pressure difference control scheme parameters is calculated according to the calculated airflow velocity, airflow direction and pressure difference parameters, the gradient represents the direction in which the target function value changes fastest, and is used to guide the adjustment of parameters; The gradient descent iterative optimization algorithm updates the parameters of the airflow organization and pressure difference control scheme according to the gradient information to generate a new airflow organization and pressure difference control scheme; When updating the scheme, the parameter change amplitude is adjusted according to the learning rate, the larger the learning rate, the faster the parameter change; It is judged whether the updated airflow organization and pressure difference control scheme meets the physical constraint conditions, including that the wind speed cannot exceed a certain maximum value and the pressure difference cannot exceed a certain maximum value; If the constraint condition is not met, the scheme is adjusted to meet the constraint condition; After the scheme adjustment is completed, the corrected airflow organization and pressure difference control scheme is obtained; If the scheme meets the constraint condition; The airflow organization and pressure difference control scheme that meets the cleanliness level requirements is output, which includes the wind speed, wind direction of each area air supply outlet, the setting position of the return air outlet and the pressure difference control strategy between areas; According to the optimized airflow organization and pressure difference control scheme, control instructions for controlling air conditioning systems and ventilation equipment are generated to realize accurate control of the clean room environment; After the scheme is output, it is visualized and displayed.

6. The method of dividing different cleanliness grade areas in a constant temperature and humidity dust-free workshop according to any one of claims 1-4, characterized in that, The step S105 further includes: Generate an initial area division plan based on the clean room's architectural structure and functional requirements; The scheme divides areas of different cleanliness levels and determines the boundaries and areas of each area; Based on the area division scheme, the airflow velocity, flow direction and cleanliness distribution under different airflow organization forms are analyzed through computational fluid dynamics simulation; Selecting the optimal airflow organization scheme so that the cleanliness of each key position in the clean room meets the design requirements, the airflow organization scheme including the position and quantity parameters of the supply and return air vents; Generate a bill of quantities based on the area division plan, the airflow organization plan, and the pressure difference control plan; Calculate the estimated construction cost and the estimated operating cost of each sub-item in the list, wherein the estimated construction cost includes the cost of building materials and equipment, and the estimated operating cost includes the cost of electricity and consumables.

7. The method of dividing different cleanliness grade areas in a constant temperature and humidity dust-free workshop according to any one of claims 1-4, characterized in that, The step S106 further includes: Acquire area division scheme data, airflow organization scheme data, pressure difference control scheme data, and buffer zone setting scheme data using AutoCAD software to generate first three-dimensional model data and first construction drawing data; Using a Monte Carlo simulation method, inputting the first three-dimensional model data, the first construction drawing data, the airflow organization scheme data, and the pressure difference control scheme data, simulating the clean room operating conditions, generating a first simulation data set, and determining whether the cleanliness level E is less than or equal to a preset cleanliness level threshold E0; If the cleanliness level E is greater than the preset cleanliness level threshold E0, the airflow organization scheme or the pressure difference control scheme is adjusted according to the first simulation data set.

Images