Method, apparatus, medium, and product for managing cleanroom contaminant concentration

Abstract

The application provides a clean room pollutant concentration management method, device, medium and product, and relates to the technical field of clean room ventilation control. The method comprises the following steps: obtaining real-time gaseous molecular pollutant AMC concentrations of each region in the clean room according to a clean room space region division result; when the real-time AMC concentration of a target region is greater than or equal to a concentration threshold value, adjusting wind speed correction values of FFUs of each region according to distances between the remaining regions and the target region, with the target region as the center; wherein the closer the distance between the target region is, the greater the corresponding wind speed correction value is; when the AMC concentration of the target region is less than the concentration threshold value, obtaining current AMC concentration values of each region, and determining an FFU wind speed optimal solution according to the current AMC concentration values of each region, clean room FFU layout information and a particle swarm optimization algorithm (PSO); and adjusting the wind speed of each FFU according to the FFU wind speed optimal solution. The application can realize global monitoring, dynamic optimization and accurate regulation and control of pollutant concentration.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of clean room ventilation control, and in particular to a clean room pollutant concentration management and control method, device, medium and product. BACKGROUND

[0002] A clean room is an indispensable infrastructure in high-tech fields such as semiconductor manufacturing, biomedicine, aerospace, etc., and its environmental quality directly affects the production yield and performance stability of products. With the continuous progress of technology, the control requirements of clean rooms for airborne molecular contaminants (AMC) are becoming increasingly stringent. AMC includes a variety of chemical molecules (such as acidic gases, basic gases, and organic volatile compounds, etc.), even if a trace amount exists, it may cause corrosion, performance degradation or even scrap of sensitive components. Under this background, efficient and accurate AMC monitoring and control technology has become one of the core research topics in the clean room industry.

[0003] Currently, the monitoring and control of AMC in clean rooms mainly relies on the following technical means: first, fixed-position gas sensors are used to monitor the AMC concentration and notify personnel for manual adjustment in a pre-warning manner; second, the air flow rate is adjusted to dilute the pollutant concentration in combination with the air supply system and fan filter unit (FFU); third, high-efficiency filters or chemical adsorption equipment are used to reduce the AMC concentration under certain conditions. These methods can alleviate pollution problems to some extent, but their implementation usually depends on static rules or manual intervention, lacking dynamic adjustment capability. At the same time, many systems use air speed adjustment logic based on preset parameters, failing to fully utilize real-time data and intelligent optimization means for precise management and control.

[0004] Although traditional methods can meet some application requirements, their limitations gradually emerge in the face of frequent changes in pollution sources in clean rooms, dynamic complexity of pollution distribution, and increasingly stringent requirements for gas quality. First, the sensor monitoring range is limited and the response speed is slow, which cannot form a real-time pollution concentration map. Second, manual or fixed rule-based air speed adjustment cannot quickly and accurately respond to sudden pollution events, which easily leads to increased energy consumption or pollution diffusion. Finally, lacking global optimization and dynamic feedback mechanisms, it cannot balance the air flow of the entire clean room after local adjustment, which may cause secondary pollution in other areas.

[0005] Therefore, there is an urgent need for an AMC monitoring and control system based on real-time data and intelligent algorithms to achieve global monitoring, dynamic optimization, and precise regulation of pollution concentration, thereby making up for the shortcomings of existing technologies. SUMMARY

[0006] The embodiment of the present application provides a clean room pollutant concentration management and control method, equipment, medium and product, so as to solve the problems of global monitoring, dynamic optimization and accurate regulation of pollution concentration.

[0007] In a first aspect, the embodiment of the present application provides a clean room pollutant concentration management and control method, characterized in that the method comprises:

[0008] According to the clean room space area division result, the real-time AMC concentration of each area in the clean room is obtained;

[0009] When the real-time AMC concentration of the target area is greater than or equal to the concentration threshold value, the wind speed correction value of each area FFU is adjusted according to the distance between the remaining areas and the target area, taking the target area as the center; wherein the closer the distance between the target area is, the greater the corresponding wind speed correction value is;

[0010] When the AMC concentration of the target area is less than the concentration threshold value, the current AMC concentration value of each area is obtained, and the FFU wind speed optimal solution is determined according to the current AMC concentration value of each area, the clean room FFU layout information and the particle swarm optimization algorithm (Particle Swarm Optimization, PSO);

[0011] The wind speed of each FFU is adjusted according to the FFU wind speed optimal solution.

[0012] In a possible implementation manner, the FFU wind speed optimal solution is determined according to the current AMC concentration value of each area, the clean room FFU layout information and the PSO, comprising:

[0013] The number of FFUs is determined according to the clean room FFU layout information;

[0014] The particle swarm size, the position of the particle and the speed of the particle are initialized by taking the number of FFUs, the current wind speed of each FFU and the wind speed between the set wind speed interval;

[0015] In each round of iteration, the fitness function value of the particle corresponding to each FFU is calculated, and the individual optimal position is updated according to the wind speed corresponding to the optimal fitness of each FFU, and the global optimal position is updated according to the wind speed corresponding to the optimal fitness of all FFUs;

[0016] Based on the individual optimal position and the global optimal position, the speed and position of the particle corresponding to each FFU are updated, and when the convergence condition is met, the corresponding global optimal solution is taken as the FFU wind speed optimal solution.

[0017] In a possible implementation manner, the convergence condition comprises:

[0018] In the continuous multiple rounds of iterations, the fitness function value of the global optimal position changes less than a set threshold, or the number of iterations reaches a preset maximum number.

[0019] In a possible implementation, the acquiring of the real-time AMC concentration of each area in the clean room comprises:

[0020] Collecting AMC concentration monitoring data sent by all sensors distributed in the clean room space at a set period; wherein, the AMC concentration monitoring data comprises sensor number and AMC concentration detection value;

[0021] Determining the coordinates of the sensors according to the sensor number, and constructing a data matrix according to the coordinates of the sensors and the corresponding AMC concentration detection values;

[0022] Drawing a concentration map according to the data matrix, and determining the average value of the AMC concentration of all grid points in each area as the AMC concentration of each area according to the concentration map.

[0023] In a possible implementation, before the determining of the coordinates of the sensors according to the sensor number, the method further comprises:

[0024] Preprocessing the AMC concentration detection values of each sensor; wherein, the preprocessing comprises: outlier detection, missing value filling and denoising and smoothing processing.

[0025] In a possible implementation, the drawing of the concentration map according to the data matrix comprises:

[0026] Comparing each grid point with the coordinates of the sensors to determine a target grid point with a missing AMC concentration value;

[0027] Calculating the AMC concentration value of the target grid point according to the data matrix and an interpolation algorithm;

[0028] Drawing a concentration map based on the AMC concentration values of each grid point and the coordinates of the grid points.

[0029] In a possible implementation, the calculating of the AMC concentration value of the target grid point according to the data matrix and the interpolation algorithm comprises:

[0030] Wherein, the AMC concentration value of the target grid point is determined according to the inverse distance weighted method, and the formula is as follows:

[0031]

[0032] Wherein, is the AMC concentration value of the i th sensor; is the AMC concentration value of the i th sensor; ​​Euclidean distance of the following: ; is a distance weight index.

[0033] In a second aspect, an embodiment of the present application provides a clean room pollutant concentration management device, comprising:

[0034] An acquisition module is configured to acquire real-time AMC concentrations of each region in the clean room according to a clean room space region division result.

[0035] A wind speed correction module is configured to, when a real-time AMC concentration of a target region is greater than or equal to a concentration threshold, adjust wind speed correction values of FFUs of each region according to distances between the remaining regions and the target region, with the target region as a center; wherein the closer the distance between a region and the target region, the greater the corresponding wind speed correction value.

[0036] A wind speed optimization module is configured to, when the AMC concentration of the target region is less than the concentration threshold, acquire current AMC concentration values of each region, and determine an FFU wind speed optimal solution according to the current AMC concentration values of each region, clean room FFU layout information, and PSO.

[0037] An adjustment module is configured to adjust wind speeds of the FFUs according to the FFU wind speed optimal solution.

[0038] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the method in the first aspect or any possible implementation manner of the first aspect when executing the computer program.

[0039] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method in the first aspect or any possible implementation manner of the first aspect.

[0040] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement the method in the first aspect or any possible implementation manner of the first aspect.

[0041] In the embodiment of the present application, the clean room pollutant concentration control method can strengthen the airflow control near the pollution source and quickly suppress the spread of pollutants by monitoring the AMC concentration of each area in real time, adjusting the FFU airflow speed correction value according to the distance from the target area when the concentration of the target area exceeds the limit, and increasing the correction value as the distance decreases. When the concentration is below the threshold, the optimal airflow speed is determined by combining the current concentration, FFU layout and PSO algorithm, which ensures the stability of the overall concentration of the clean room and realizes the global optimization of the airflow speed through the intelligent algorithm, thereby balancing the efficiency of the system operation while quickly responding to the pollution risk. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 is an application scenario diagram of the clean room pollutant concentration control method provided by the embodiment of the present application;

[0043] Figure 2 is a flowchart of the implementation of the clean room pollutant concentration control method provided by the embodiment of the present application;

[0044] Figure 3 is a structural schematic diagram of the clean room pollutant concentration control device provided by the embodiment of the present application;

[0045] Figure 4 is a schematic diagram of an electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0046] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0047] Figure 1 is an application scenario diagram of the clean room pollutant concentration control method provided by the embodiment of the present application. As shown in Figure 1 , a plurality of concentration sensors and FFUs are distributed in the clean room, and a control terminal is used to remotely collect data sent by the concentration sensors and FFUs to comprehensively detect the concentration and the airflow speed of the FFUs, and to timely adjust the airflow speed of the FFUs according to the change of the clean room pollutant concentration, so as to meet the multi-directional requirements of global monitoring, dynamic optimization and accurate regulation of the pollution concentration. In the embodiments of the present application, the sensor refers to the concentration sensor arranged in the clean room for detecting the AMC concentration.

[0048] During the layout of the concentration sensor, it is ensured that the sensor can cover all key areas of the clean room, including possible pollution source positions, areas with weak air flow, and distribution centers of the FFUs.

[0049] Suppose the area of the clean room is , it is divided into unit cells, each unit cell has the same area. The center point of the sensor is defined as Each cell is arranged with at least one sensor. In areas where the pollutant diffusion speed is fast (such as clean room entrance, equipment operation position, etc.), the density of sensors should be appropriately increased. If the entire clean room is divided into high-risk areas and low-risk areas, the sensor spacing in high-risk areas is set smaller, and the spacing in low-risk areas may be appropriately increased.

[0050] The total number of sensors is calculated according to the sensor spacing :

[0051]

[0052] wherein, is the spacing of the sensors, depends on the cleanliness level of the clean room and the diffusion characteristics of the pollutants.

[0053] During the concentration collection process, each sensor periodically collects an AMC concentration value, denoted as , with units of ppb. The data collection frequency is set to every second or higher, depending on the speed of pollutant diffusion in the clean room and the regulation requirements. Wireless transmission (such as WiFi, ZigBee) or wired transmission (such as Ethernet) is used to transmit sensor data to the central control system.

[0054] The control terminal identifies each sensor with the sensor number and its coordinates as identifiers, saved as a real-time data matrix:

[0055]

[0056] To visualize the concentration distribution, the concentration matrix can be displayed in the form of a heat map, which is used to quickly identify pollution sources and abnormal areas. The color depth represents the level of AMC concentration. For example: green: low concentration, good air quality; yellow: concentration close to the threshold, need attention; red: concentration exceeds the standard, need to adjust immediately; the concentration map can be dynamically updated based on the constructed real-time data matrix.

[0057] To improve the real-time performance of pollutant concentration monitoring, the concentration map is updated at fixed time intervals (such as 1 second) according to the latest sensor data. The concentration map is updated in real time and will be used for subsequent FFU wind speed adjustment.

[0058] During the FFU layout process, ensure that the FFUs are evenly distributed within the cleanroom while covering all areas. Achieve uniform air flow within the cleanroom, avoiding dead zones and over-flow zones. Each FFU is responsible for air flow and purification within its covered area. Assume that the cleanroom is divided into a grid, with one FFU per grid point. According to the cleanroom air flow design requirements (such as unidirectional flow or non-unidirectional flow), determine the installation direction and arrangement density of the FFUs.

[0059] Further, in order to optimize the FFU layout, the high-cleanliness-level (such as ISO5 level) area needs to increase the FFU arrangement density, and the low-cleanliness-level area can appropriately reduce the arrangement density.

[0060] Assume that the area of the cleanroom is , which is divided into grid. The center point of each grid is the position of an FFU, and the coordinates are represented as , where . Each FFU is assigned a unique number , and its position is recorded.

[0061] In actual implementation, the area responsible for each FFU is defined as its grid and its surrounding part, called "FFU influence area". The area is defined as a certain range centered on , usually the size of the grid. Ensure that all grid points are covered by at least one FFU.

[0062] To ensure that all grid points are covered by at least one FFU, verify according to the following formula:

[0063] For each grid point , calculate the distance between it and the nearest FFU:

[0064]

[0065] If the distance from any grid point to the FFU exceeds the specified maximum range , the number of FFUs needs to be increased or the layout needs to be adjusted. Wherein, the specified maximum range is set according to the cleanliness level requirements of the cleanroom, to avoid the distance between FFUs being too far, resulting in insufficient cleanliness of the cleanroom.

[0066] The cleanroom pollutant concentration control method provided by the embodiments of the present application is based on the above, in order to make the specific process of the cleanroom pollutant concentration control method clearer, the following will be described in detail in combination with the drawings.

[0067] Referring to Figure 2It shows the implementation flowchart of the clean room pollutant concentration management method provided by the embodiment of the application, and includes the following steps:

[0068] S201, obtaining the real-time AMC concentration of each area in the clean room according to the clean room space area division result.

[0069] The execution subject of each embodiment of the application can be a server, a processor, a microprocessor, or other devices with data processing functions. In the actual implementation process, the specific implementation mode of the execution subject can be selected according to actual needs, and the application does not make special limitations on this. As long as it is a device with data processing function. In order to facilitate the understanding of the embodiments of the application, the control terminal shown in Figure 1 is taken as the execution subject for description.

[0070] As can be known from the foregoing description, the entire clean room is divided into a high-risk area and a low-risk area. In other embodiments, the clean room space can be divided into areas according to the functionality of different partitions of the clean room. In order to fully grasp the distribution of pollutants in the clean room, the real-time AMC concentration of each area in the clean room is obtained.

[0071] S202, when the real-time AMC concentration of the target area is greater than or equal to the concentration threshold value, adjusting the wind speed correction value of the FFU of each area according to the distance between the remaining areas and the target area, taking the target area as the center; wherein the closer the distance between the target area and the corresponding wind speed correction value is, the greater the wind speed correction value is.

[0072] Wherein, if the concentration value of a certain area exceeds the preset threshold value , it is considered that the area is a pollution source, and the system triggers a local optimization response to adjust the FFU wind speed of the target area and the surrounding area. Trigger condition formula:

[0073]

[0074] Adjusting the wind speed correction value of the FFU of each area according to the distance between the remaining areas and the target area aims to quickly respond to the sudden increase of the AMC concentration of a certain area in the clean room, and preferentially adjusts the FFU wind speed of the area and its adjacent areas to rapidly reduce the pollution concentration and prevent the spread of pollution. Ensure that the local pollution event is controlled within a short time, and lay a foundation for subsequent global optimization.

[0075] In different embodiments, when adjusting the wind speed correction value of the FFU of each area, the range of the adjusted FFU is different.

[0076] In one possible implementation, when the area of the clean room or the number of FFUs in the clean room is small, the wind speed of the FFU of all areas of the clean room can be adjusted.

[0077] In other possible implementations, when the cleanroom space is large, the number of FFUs is large, or the AMC concentration increase is small, the airflow speed of the FFUs within a set distance around the target area can be adjusted to avoid concentration fluctuations in other distant areas.

[0078] For example: the location of the pollution source, i.e., the target area where the real-time AMC concentration is greater than or equal to the concentration threshold, is... The corresponding AMC concentration is The radius extends outwards from the pollution source to adjacent areas. All FFU points within the range are the adjustment targets.

[0079] FFU set in the neighboring area:

[0080]

[0081] in, The radius of extension outside the adjacent area; Target area To FFU Euclidean distance:

[0082]

[0083] Prioritize reducing the wind speed of FFUs in the target area to minimize their diffusion effect. For FFUs in adjacent areas, the wind speed of FFUs closer to the pollution source is increased more, while the wind speed of FFUs farther away is increased less, with weights decreasing according to distance. For the target area... The FFU reduces its wind speed. To reduce the spread of pollutants. The wind speed adjustment formula is:

[0084]

[0085] in, The wind speed of the FFU at the pollution source; This represents the minimum wind speed of the FFU. This represents the maximum value of the FFU wind speed; This represents the maximum possible value of AMC concentration. This is a preset threshold.

[0086] For FFUs in the vicinity, wind speed is increased according to a distance-weighted decreasing principle. The closer the distance, the greater the increase in wind speed. The wind speed adjustment formula is:

[0087]

[0088] in, The wind speed of the nearby FFU; The concentration value of the target area is responsible for the concentration value of the adjacent area FFU.

[0089] In the implementation process, when adjusting the wind speed correction value of each area FFU according to the distance between the target area and the remaining areas, a distance weighting factor is determined according to the distance between the target area and the remaining areas, and the wind speed correction value of each area FFU is adjusted according to the distance weighting factor.

[0090] Specifically, on the basis of the aforementioned wind speed adjustment formula, the wind speed of the adjacent area FFU is further multiplied by the distance weighting factor So that the FFU closer to the pollution source increases more wind speed:

[0091]

[0092] The comprehensive adjustment formula is:

[0093]

[0094] In this implementation mode, adjusting the wind speed of each area FFU according to the distance weighting factor can ensure that the pollutants are stably diffused under the guidance of the wind speed of each area FFU.

[0095] In other possible implementation modes, in order to improve the wind speed adjustment effect, the wind speed correction value set in advance can be determined according to the distance sorting result between the target area and the remaining areas, and the FFU wind speed is adjusted according to the set wind speed correction value.

[0096] S203, when the AMC concentration of the target area is less than the concentration threshold value, obtaining the current AMC concentration value of each area, and determining the FFU wind speed optimal solution according to the current AMC concentration value of each area, the clean room FFU layout information and the PSO.

[0097] The system monitors the concentration changes of the pollution source point and its adjacent area in real time, and the adjusted FFU wind speed takes effect immediately. The concentration distribution is recalculated after adjusting the FFU wind speed each time, and the concentration map is updated.

[0098] The foregoing embodiments introduce that in different embodiments, when adjusting the wind speed correction value of each area FFU, the range of the adjusted FFU is different. Therefore, after the local optimization is completed, the optimal wind speed distribution of the FFU in the entire clean room is calculated through global optimization. The AMC concentration of all areas in the clean room is maintained in the safe range, and the accurate control of the overall pollutants is realized. After the local optimization, the local concentration may decrease, but the local adjustment may disturb the air flow distribution of other areas. Therefore, after the local optimization is completed, the global optimization will reevaluate the concentration distribution of the entire clean room, and balance the concentration fluctuation caused by the local adjustment through the FFU wind speed distribution. The global optimization wind speed distribution The adjustment results of local optimization will be covered, and these disturbances will be balanced through overall calculation to ensure that the entire clean room returns to the optimal state.

[0099] After global optimization, the system continues to monitor the changes in the concentration distribution . If a new pollution event is detected, local optimization is triggered again, and after the completion of local optimization, a new round of global optimization is entered. Even if there is no local pollution event, regular optimization of the clean room can ensure overall operational efficiency and pollution control effectiveness.

[0100] PSO is a bionic optimization algorithm that searches for the optimal solution of a problem by simulating the cooperative behavior of a particle swarm. In FFU wind speed optimization, each particle represents an FFU wind speed distribution scheme, and the algorithm gradually approaches the optimal wind speed distribution through particle speed and position updates, enabling fast response to pollution risks while ensuring system efficiency.

[0101] S204, adjusting the wind speed of each FFU according to the optimal solution of the FFU wind speed.

[0102] After the control terminal determines the optimal solution of the FFU wind speed, the corresponding FFU wind speed is distributed according to the number of each FFU to achieve synchronous adjustment.

[0103] In this embodiment, the clean room pollutant concentration management method adjusts the FFU wind speed correction value by distance from the target area when the target area concentration exceeds the limit, with the closer the distance, the larger the correction value, which can effectively strengthen the airflow control near the pollution source and quickly suppress the spread of pollutants. When the concentration falls below the threshold, the optimal wind speed is determined based on the current concentration, FFU layout, and PSO algorithm, ensuring the stability of the overall concentration of the clean room and achieving global optimization of the wind speed through intelligent algorithms, thereby responding quickly to pollution risks while ensuring system efficiency.

[0104] In one possible implementation, the optimal solution of the FFU wind speed is determined based on the current AMC concentration of each area, the clean room FFU layout information, and PSO, including:

[0105] Determine the number of FFUs based on the clean room FFU layout information;

[0106] Initialize the size of the particle swarm, the position of the particle, and the speed of the particle based on the number of FFUs, the current wind speed of each FFU, and the wind speed within the set wind speed range;

[0107] In each iteration, calculate the fitness function value of each FFU corresponding particle, and update the individual optimal position based on the optimal wind speed of each FFU fitness, and update the global optimal position based on the optimal wind speed of all FFU fitness;

[0108] Based on the individual optimal position and the global optimal position, the velocity and position of the particles corresponding to each FFU are updated. When the convergence condition is met, the corresponding global optimal solution is used as the optimal solution for the FFU wind speed.

[0109] In actual implementation, each Corresponding to its area of ​​responsibility The AMC concentration value is the concentration of all grid points covering the region. Average value:

[0110]

[0111] in, Indicates the region The number of grid points within.

[0112] Construct a mapping matrix between FFU and regional concentration. :

[0113]

[0114] Each of them This indicates the concentration value of the area covered by the corresponding FFU.

[0115] FFU layout information, including: FFU number ,coordinate and each FFU's responsible area concentration value , is represented as:

[0116]

[0117] Each time the concentration map is updated, the concentration value for each FFU region is recalculated. And update the concentration distribution matrix of FFU. This provides support for subsequent FFU wind speed adjustment.

[0118] In the specific implementation process, the number of FFUs is determined based on the FFU layout information.

[0119] The fitness function is used to evaluate the effectiveness of wind speed distribution schemes and is defined as the deviation between the pollutant concentration and the target concentration. Fitness function formula:

[0120]

[0121] in, This represents the wind speed distribution across all FFUs; The concentration value is calculated based on the current wind speed distribution; The target concentration value (usually the set ideal concentration range in the clean room).

[0122] Adaptation FFU control, determine particle definition:

[0123] Position: the current position of the particle Indicates the wind speed distribution of the FFU.

[0124] Velocity: the velocity of the particle Decide the direction and amplitude of the wind speed adjustment of the next step of the particle.

[0125] Before determining the optimal solution of the FFU wind speed based on PSO, first initialize the particle swarm size , the position of the particle and the velocity of the particle:

[0126] Among them, the initial wind speed (i.e. the position of the particle) is distributed as:

[0127]

[0128] Its value range is limited between the minimum wind speed and the maximum wind speed allowed by the FFU device.

[0129] The initial velocity (i.e. the velocity of the particle) is:

[0130]

[0131] Among them, Indicates a random number generated in the interval . The initial velocity is randomly set within a reasonable range to ensure the diversity of optimization.

[0132] Calculate the fitness value of each particle , and update the following two items:

[0133] Individual optimal position : record the particle The historical optimal wind speed distribution, that is, the optimal wind speed scheme of the particle since iteration of each FFU;

[0134] Global optimal position : record the current optimal wind speed distribution in the entire particle swarm, that is, the optimal wind speed scheme of all FFUs in terms of fitness.

[0135] Update the velocity of each particle according to the following formula:

[0136]

[0137] Among them, is the inertia weight, which controls the degree of dependence of the particle on the historical velocity; , is the learning factor, which respectively controls the learning ability of the particle to the optimal solution and the global optimal solution. is the global optimal position for the jth iteration; is the individual optimal position of the jth particle for the jth iteration.

[0138] The position of each particle is updated according to the following formula:

[0139]

[0140] wherein, is the position of the jth particle at the moment t; is the position of the jth particle at the moment t; is the position of the jth particle at the moment t; is the velocity of the jth particle at the moment t. When the convergence condition is met, the iteration is terminated, and the wind speed distribution scheme corresponding to the global optimal position is assigned as the target wind speed of the FFU in each region (i.e., the global optimization value). The global optimal FFU wind speed distribution is: .

[0141] In the specific implementation process, this scheme is used as the wind speed benchmark during daily stable operation, and also as the target value for restoring the FFU wind speed of the pollution source region and the adjacent region after the local optimization response is completed, so as to achieve a balance between concentration control and energy consumption optimization in space.

[0142] In this embodiment, the method initializes the particle swarm parameters according to the number of FFUs, dynamically updates the individual and global optimal positions in the iteration process, and adjusts the particle velocity and position based on this, and finally takes the global optimal solution at the time of convergence as the optimal solution of the wind speed, so that the optimization process of the PSO algorithm is closely combined with the actual layout and operation parameters of the clean room FFU, ensuring that the found optimal solution of the wind speed can adapt to the specific environment of the clean room, improving the practicality and reliability of the optimal solution, and further enhancing the accuracy of concentration control.

[0143] In different embodiments, the convergence condition is different according to different control requirements.

[0144] In one possible implementation, the convergence condition includes that the change amount of the fitness function value corresponding to the global optimal position in continuous multiple iterations is less than a set threshold.

[0145] In the actual implementation process, the change of the fitness value is less than the threshold is expressed as:

[0146]

[0147]

[0148] wherein, is the latest fitness function value; ​​is the fitness value of the previous iteration; is a set threshold.

[0149] In this embodiment, the convergence condition is set as that the global optimal fitness changes less than the threshold in continuous multiple iterations, which indicates that the optimization tends to be stable and it is difficult to significantly improve the scheme performance by continuing iteration. Therefore, the iteration is stopped when the global optimal fitness changes less than the threshold, which can avoid the algorithm from falling into meaningless loop iteration and ensure timely termination when the optimization effect tends to be stable to save computing resources.

[0150] In another possible implementation, the convergence condition includes that the number of iterations reaches a preset maximum number.

[0151] In this embodiment, the convergence condition is set as that the maximum number of iterations is reached, which balances the optimization accuracy and efficiency by limiting the maximum number of iterations to ensure the algorithm outputs results within an acceptable time.

[0152] In other possible implementations, the convergence condition includes:

[0153] In continuous multiple iterations, the global optimal position corresponds to a fitness value change less than a set threshold, or the number of iterations reaches a preset maximum number.

[0154] In this embodiment, the convergence condition is set as that the global optimal fitness changes less than the threshold or the maximum number of iterations is reached in continuous multiple iterations, which can not only avoid the algorithm from falling into meaningless loop iteration and ensure timely termination when the optimization effect tends to be stable to save computing resources, but also balance the optimization accuracy and efficiency by limiting the maximum number of iterations to ensure the algorithm outputs results within an acceptable time, making the PSO algorithm more feasible in practical applications.

[0155] In a possible implementation, the real-time AMC concentration of each area in the clean room is obtained, including:

[0156] The AMC concentration monitoring data sent by all sensors distributed in the clean room space is collected at a set period; wherein the AMC concentration monitoring data includes a sensor number and an AMC concentration detection value;

[0157] The coordinates of the sensor are determined according to the sensor number, and a data matrix is constructed according to the coordinates of the sensor and the corresponding AMC concentration detection value;

[0158] The concentration map is drawn according to the data matrix, and the average value of the AMC concentration of all grid points in each area is determined as the AMC concentration of each area according to the concentration map.

[0159] The real-time AMC concentration of each area in the clean room is obtained, and the concentration map is determined based on the foregoing embodiments. In the specific implementation process, in order to improve the accuracy of the area concentration detection, a region covers multiple grid points. Therefore, when determining the AMC concentration of each region, the average value of the AMC concentrations of all grid points in each region is taken as the AMC concentration of the region, so as to avoid that the detection value of a single grid point concentration is too high or too low to affect the accuracy of the overall regional AMC concentration.

[0160] In the embodiment, by periodically collecting sensor data, combining sensor coordinates to construct a data matrix and drawing a concentration map, and finally taking the average value of the grid point concentration in the region as the concentration of the region, the fine perception of the AMC concentration of each area in the clean room is realized. This concentration acquisition method based on spatial distribution can more truly reflect the pollution status of different areas, provide accurate judgment basis for subsequent wind speed adjustment, and avoid improper control caused by single point data deviation.

[0161] In a possible implementation, before determining the coordinates of the sensor according to the sensor number, the method further includes:

[0162] The AMC concentration detection value of each sensor is preprocessed; wherein the preprocessing includes: outlier detection, missing value filling and denoising smoothing processing.

[0163] In the actual implementation process, the concentration values collected are detected, and if the concentration value is out of the design range of the sensor (such as lower than zero or higher than the upper limit of the equipment), it is marked as abnormal data. If some sensor data is missing, an interpolation method (such as linear interpolation or high-order interpolation) is used to complete the missing value filling.

[0164] In the denoising smoothing processing, the concentration values collected are subjected to moving average or Kalman filtering to reduce the influence of short-term fluctuations, and the formula is:

[0165]

[0166] wherein, is the size of the sliding window; is the index of a data point in the sliding window.

[0167] In the embodiment, the preprocessing of the sensor detection value, including outlier detection, missing value filling and denoising smoothing, can effectively eliminate invalid data, supplement incomplete information and reduce noise interference, so that the constructed data matrix is closer to the real concentration distribution. Based on the purified data, the concentration map is drawn and the regional concentration is calculated, which can reduce the influence of data error on concentration judgment and improve the reliability of concentration monitoring, thereby laying a foundation for the accurate implementation of subsequent control measures.

[0168] In a possible implementation, the concentration map is drawn according to the data matrix, comprising:

[0169] comparing each grid point with the coordinates of the sensors to determine a target grid point with a missing AMC concentration value;

[0170] calculating the AMC concentration value of the target grid point according to the data matrix and an interpolation algorithm;

[0171] drawing the concentration map based on the AMC concentration values of each grid point and the coordinates of the grid points.

[0172] Based on Figure 1 As can be known from the related content of the scene diagram, the density of the sensors should be appropriately increased in the area where the pollutants diffuse at a relatively high speed (for example, the entrance of the clean room and the position where the equipment is operated). Therefore, when the concentration map is constructed, the number of the AMC concentration values of the grid points in the area where the density of the sensors is relatively low is small, and therefore, the concentration values of the grid points with the missing AMC concentration values need to be filled according to the density of the surrounding sensors.

[0173] Among them, the sensors around the target grid point with the missing AMC concentration value are determined according to the data matrix and the interpolation algorithm, the concentration change trend is analyzed by comprehensively monitoring the concentration values of the surrounding sensors to obtain the AMC concentration value at the target grid point.

[0174] In this embodiment, by determining the target grid point with the missing concentration and calculating the concentration of the target grid point by using the interpolation algorithm, a complete concentration map is drawn, and the problem of spatial data blank caused by the discrete distribution of the sensors is solved. The application of the interpolation algorithm enables the concentration map to completely cover all areas of the clean room, clearly presents the spatial change trend of the concentration, helps the system to more comprehensively master the pollution distribution, and thus makes the subsequent area division and air speed adjustment more targeted.

[0175] In a possible implementation, the AMC concentration value of the target grid point is calculated according to the data matrix and the interpolation algorithm, comprising:

[0176] Among them, the AMC concentration value of the target grid point is determined according to the inverse distance weighted method, and the formula is as follows:

[0177]

[0178] Among them, is the AMC concentration value of the i th sensor; is the Euclidean distance from the grid point to the sensor; is the distance weight index, which is usually taken as 2.

[0179] ​​​​In the embodiment, the AMC concentration value of the target grid point is calculated by using the inverse distance weighting method, and by giving greater weight to the sensor data closer to the target grid point, the interpolation result is more in line with the actual concentration spatial distribution rule. This method fully considers the influence of spatial correlation on concentration, avoids the excessive interference of remote sensor data on the concentration of nearby grid points, improves the accuracy of the concentration calculation of the target grid point, and further improves the accuracy and reference value of the concentration map.

[0180] In other possible implementations, the AMC concentration value of the target grid point is calculated according to the data matrix and the interpolation algorithm, comprising:

[0181] A certain number of sensors around the target grid point with a missing AMC concentration value are determined;

[0182] The average value of the concentration values corresponding to the certain number of sensors is calculated as the AMC concentration value of the target grid point, denoted as .

[0183] In the specific implementation process, when calculating the AMC concentration value of the target grid point, a two-dimensional matrix is constructed based on the AMC concentration values of each grid point to generate a concentration map:

[0184]

[0185] Wherein, each element represents the concentration value of the corresponding grid point of the matrix.

[0186] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0187] The above embodiments mainly introduce the FFU air speed adjustment scheme when the real-time AMC concentration of the target area is greater than or equal to the concentration threshold, to realize the control of the clean room pollutant concentration. In the specific implementation process, it is necessary to dynamically feedback and adjust the clean room pollutant concentration in the clean room to realize real-time updating of the concentration data and dynamic adjustment of the FFU air speed, so as to ensure that the FFU air speed adjustment scheme can adapt to the real-time change of the pollution concentration in the clean room. Dynamically adjust the FFU air speed distribution in the clean room, timely respond to sudden pollution source events, and balance the adjustment results of local optimization and global optimization.

[0188] The dynamic feedback and adjustment process is as follows, and for details not described in detail, reference can be made to the corresponding method embodiments described above:

[0189] First, based on the concentration map The system continuously monitors the changes of AMC concentration in each area of the cleanroom. The core of the dynamic feedback mechanism is to detect new concentration anomalies through cyclic feedback and trigger the corresponding optimization strategy (local optimization or global optimization) in time.

[0190] Secondly, concentration anomaly detection is performed. For each area concentration Real-time monitoring is performed, and if the concentration value at any position exceeds the set threshold , local optimization is triggered. The FFU wind speed of the pollution source and its adjacent areas is quickly adjusted.

[0191] Trigger condition formula:

[0192]

[0193] During the local optimization stage, the FFU wind speed of the pollution source and its adjacent areas is dynamically adjusted to reduce pollution diffusion. After local optimization, the wind speed is updated in real time:

[0194]

[0195] where, is the minimum value of the FFU wind speed; is the maximum value of the FFU wind speed; is the maximum possible value of the AMC concentration; is the preset threshold; is the concentration of each area.

[0196] Then, when the concentration of the pollution source point returns to the normal level (i.e. ), local adjustment ends and enters the recovery stage. The system enters the global optimization stage to balance the impact of local adjustment on the overall cleanroom airflow. In the global optimization stage, the wind speed of all FFUs is redistributed according to the results of particle swarm optimization to ensure that all FFUs operate at the optimal wind speed and ensure the overall airflow stability of the cleanroom.

[0197] To avoid new concentration fluctuations caused by sudden changes in wind speed, the system gradually restores the FFU wind speed:

[0198]

[0199] where, is the current wind speed; is the wind speed value of global optimization; is the recovery rate coefficient, with a value range of [0, 1]. A larger indicates slower recovery speed, suitable for scenarios sensitive to airflow.

[0200] Finally, the system updates the concentration map at fixed time intervals (such as 1 second) , ensure that the FFU wind speed adjustment is always based on the latest data.

[0201] Update formula:

[0202]

[0203] Wherein, is the weighted factor of the sensor to the position Real-time concentration data is stored in the database for subsequent analysis of the pollution change law of the clean room, prediction of potential pollution source location or high pollution risk area through historical data analysis, and optimization of the layout of sensors and FFUs. At the same time, the real-time updated concentration map , so that the latest AMC concentration distribution of all areas in the clean room can be intuitively viewed.

[0204] Dynamic feedback updated FFU wind speed distribution :

[0205]

[0206] The dynamic feedback output wind speed distribution is directly applied to the FFU regulation to provide input for the next round of optimization.

[0207] In the above dynamic optimization process, even if no concentration anomaly occurs in the clean room, global optimization will be performed according to the preset period to ensure the stability of the long-term operation of the clean room.

[0208] The following is an embodiment of the device of the present application, and for details not described in detail, reference can be made to the corresponding method embodiments described above.

[0209] Figure 3 The structure of the clean room pollutant concentration management and control device provided by the embodiment of the present application is shown, only the parts related to the embodiment of the present application are shown for the convenience of description, and the details are as follows:

[0210] As Figure 3 shown, the clean room pollutant concentration management and control device 3 comprises:

[0211] The acquisition module 301 is configured to acquire the real-time AMC concentration of each area in the clean room according to the clean room space area division result.

[0212] The wind speed correction module 302 is configured to, when the real-time AMC concentration of the target area is greater than or equal to the concentration threshold, adjust the wind speed correction value of each area FFU according to the distance between the remaining areas and the target area, taking the target area as the center; wherein the closer the distance between the target area and the corresponding wind speed correction value is, the greater the wind speed correction value is.

[0213] The wind speed optimization module 303 is configured to acquire current AMC concentration values of the regions when the AMC concentration of the target region is less than the concentration threshold, and determine an FFU wind speed optimal solution according to the current AMC concentration values of the regions, clean room FFU layout information and PSO.

[0214] The adjustment module 304 is configured to adjust the wind speed of each FFU according to the FFU wind speed optimal solution.

[0215] In this embodiment, the clean room pollutant concentration management method can adjust the FFU wind speed correction value according to the distance when the concentration of the target region exceeds the limit, and the correction value is greater when the distance is closer, so that the airflow control near the pollution source can be targeted to be strengthened, and the pollutant diffusion can be quickly inhibited. When the concentration is below the threshold, the optimal wind speed is determined by combining the current concentration, FFU layout and PSO algorithm, so that the stability of the overall concentration of the clean room is ensured, and the global optimization of the wind speed is realized through the intelligent algorithm, so that the efficiency of the system operation is considered while the pollution risk is quickly responded.

[0216] Figure 4 is a schematic diagram of an electronic device provided by an embodiment of the present application. As shown in Figure 4 The electronic device 4 of this embodiment includes a processor 40 and a memory 41. The memory 41 stores a computer program 42. The processor 40 implements the steps in each of the method embodiments described above when executing the computer program 42. Alternatively, the processor 40 implements the functions of each module / unit in each of the device embodiments described above when executing the computer program 42.

[0217] For example, the computer program 42 can be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to complete the present application. The one or more modules / units can be a series of computer program instruction segments that can complete a specific function, which are used to describe the execution process of the computer program 42 in the electronic device 4.

[0218] The electronic device 4 can include, but is not limited to, the processor 40 and the memory 41. Those skilled in the art can understand that Figure 4 The electronic device 4 is only an example and does not constitute a limitation on the electronic device 4, and can include more or fewer components than the diagram, or combine certain components, or different components, for example, the electronic device 4 can also include an input / output device, a network access device, a bus, etc.

[0219] The processor 40 can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0220] The memory 41 can be an internal storage unit of the electronic device 4, such as a hard disk or a memory of the electronic device 4. The memory 41 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, or the like equipped on the electronic device 4. Further, the memory 41 can also include both the internal storage unit and the external storage device of the electronic device 4. The memory 41 is used to store the computer program 42 and other programs and data required by the electronic device 4. The memory 41 can also be used to temporarily store data that has been output or will be output.

[0221] For the convenience and brevity of description, only the division of the above functional modules / units is exemplified, and in actual application, the above functions can be completed by different functional modules / units according to needs. The above modules / units can be realized in the form of hardware, in the form of software, or in the form of combination of hardware and software.

[0222] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program. When the computer program is executed by a processor, the method in each method embodiment described above is implemented.

[0223] The embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method in each method embodiment described above is implemented.

[0224] The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), electric carrier wave signal, telecommunication signal and software distribution medium, etc.

[0225] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in a certain embodiment can be referred to the relevant description of other embodiments. If there is no special description and logical conflict, the terms and / or descriptions of different embodiments are consistent and can be mutually referred to, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0226] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method of managing the concentration of contaminants in a cleanroom, comprising: The method comprises the following steps: ​ According to the clean room space area division result, the real-time gaseous molecular contaminant AMC concentration of each area in the clean room is obtained; When the real-time AMC concentration of the target area is greater than or equal to the concentration threshold value, the wind speed correction value of the fan filter unit FFU of each area is adjusted according to the distance between the remaining areas and the target area, with the target area as the center; wherein the closer the distance between the target area is, the greater the corresponding wind speed correction value is; When the AMC concentration of the target area is less than the concentration threshold value, the current AMC concentration value of each area is obtained, and the FFU wind speed optimal solution is determined according to the current AMC concentration value of each area, the clean room FFU layout information and the particle swarm optimization algorithm PSO; The wind speed of each FFU is adjusted according to the FFU wind speed optimal solution; Wherein, the real-time AMC concentration of each area in the clean room is obtained, comprising: Collecting the AMC concentration monitoring data sent by all sensors distributed in the clean room space at a set period; wherein the AMC concentration monitoring data includes sensor number and AMC concentration detection value; Determine the coordinates of the sensors according to the sensor number, and construct a data matrix according to the coordinates of the sensors and the corresponding AMC concentration detection value; Draw a concentration map according to the data matrix, and determine the average value of the AMC concentration of all grid points in each area as the AMC concentration of each area according to the concentration map; Wherein, the concentration map is drawn according to the data matrix, comprising: Determine the target grid point with missing AMC concentration value by comparing each grid point with the coordinates of the sensors; Calculate the AMC concentration value of the target grid point according to the data matrix and the interpolation algorithm; Draw a concentration map based on the AMC concentration value of each grid point and the grid point coordinates; Wherein, the AMC concentration value of the target grid point is calculated according to the data matrix and the interpolation algorithm, comprising: Determine the AMC concentration value of the target grid point according to the inverse distance weighted method, the formula is as follows: wherein, is the AMC concentration value for the th sensor; is the Euclidean distance from the grid point to the sensor ; ; is the distance weight exponent.

2. The method of claim 1, wherein, According to the current AMC concentration value of each area, the clean room FFU layout information and the PSO, the FFU wind speed optimal solution is determined, comprising: Determine the number of FFUs according to the clean room FFU layout information; Initialize the particle swarm size, the position of the particles and the speed of the particles with the number of FFUs, the current wind speed of each FFU and the wind speed between the set wind speed interval; In each round of iteration, the fitness function value of each FFU corresponding particle is calculated, and the individual optimal position is updated according to the wind speed corresponding to the optimal fitness of each FFU, and the global optimal position is updated according to the wind speed corresponding to the optimal fitness of all FFUs; Based on the individual optimal position and the global optimal position, the speed and position of each FFU corresponding particle are updated, and when the convergence condition is met, the corresponding global optimal solution is taken as the FFU wind speed optimal solution.

3. The method of claim 2, wherein the step of determining the concentration of the contaminant in the clean room is performed by a method comprising: The convergence condition comprises: In continuous multiple rounds of iteration, the change amount of the fitness function value corresponding to the global optimal position is less than a set threshold value, or the number of iterations reaches a preset maximum number.

4. The method for controlling the concentration of contaminants in a cleanroom according to claim 1, characterized in that, Before the sensor coordinates are determined according to the sensor number, the method further comprises the following steps: The AMC concentration detection values of each sensor are preprocessed; wherein the preprocessing includes: outlier detection, missing value filling and denoising smoothing processing.

5. An electronic device, comprising: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method in any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method in any one of claims 1 to 4.

7. A computer program product, characterised in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method in any one of claims 1 to 4.

Images