Environmental cleanliness measurement and control method and device, computing equipment and storage medium

Abstract

The embodiment of the invention provides an environment cleanliness measurement and control method and device, computing equipment and a storage medium, and the environment cleanliness measurement and control method comprises the steps that multiple parameters of a target environment are obtained, and the multiple parameters comprise at least two of the following parameters: the number of times of ventilation, the FFU full distribution rate, the temperature and humidity, the pressure gradient, the equipment heat dissipating capacity and the airflow organization mode; the multiple parameters are input into a cleanliness prediction model, regulation and control parameters output by the cleanliness prediction model are obtained, and the cleanliness prediction model calculates and obtains the regulation and control parameters based on the multiple parameters and the incidence relation between coupling information among the multiple parameters and the target cleanliness; and corresponding parameters in the multiple parameters are adjusted to target values based on the regulation and control parameters, so that the target environment cleanliness reaches the target cleanliness. According to the scheme, a single-factor influence rule is independently analyzed, a multi-factor coupling mechanism is researched on the basis of an orthogonal experimental method, and a complete cleanliness prediction model is constructed. And factors and coupling action mechanisms thereof can be determined more comprehensively.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of environmental regulation, and particularly relate to an environmental cleanliness measurement and control method and device, a computing device, and a storage medium. BACKGROUND

[0002] With the rapid development of global technology industry, the production environment requirements of high-tech industries such as semiconductor manufacturing, biomedicine, precision electronics, aerospace, etc. are increasingly stringent. The cleanliness of the internal environment of the clean room, as the core infrastructure of these industries, is directly related to the product yield, performance stability and production cost. Cleanliness, as a key indicator of the quality of the clean room environment, has a crucial role in ensuring the stability of the production process and the reliability of the product. For example, in semiconductor chip manufacturing, micron-level or even nanometer-level dust particles can cause circuit short circuits or signal interference; in the biopharmaceutical field, microorganism control in the clean room is the key to ensuring the sterility of pharmaceutical products.

[0003] Currently, there are still deficiencies in the research process of the influencing factors of the cleanliness of the clean room: specifically, existing researches mainly focus on the linear influence of a single factor (such as the number of air changes, airflow organization method, etc.) on cleanliness, and lack of systematic analysis of the interaction of multiple factors.

[0004] Therefore, in-depth research on the influencing factors of the cleanliness of the clean room and their coupling mechanism is of great significance for improving the environmental control level of the clean room, optimizing the production process, and reducing production costs. SUMMARY

[0005] Therefore, embodiments of the present application provide an environmental cleanliness measurement and control method. One or more embodiments of the present application also relate to an environmental cleanliness measurement and control device, a computing device, and a computer-readable storage medium to solve the technical defects in the prior art.

[0006] In a first aspect, the embodiments of the present application provide an environmental cleanliness measurement and control method, comprising: obtaining a plurality of parameters of a target environment, the plurality of parameters comprising at least two of the following: the number of air changes, the full-face unit (FFU) coverage rate, temperature and humidity, pressure gradient, equipment heat dissipation, and airflow organization method; inputting the plurality of parameters into a cleanliness prediction model, and obtaining a control parameter output by the cleanliness prediction model, wherein the control parameter is calculated based on the association between the plurality of parameters and the coupling information and the target cleanliness; adjusting the corresponding parameter in the plurality of parameters to a target value based on the control parameter, so that the target environmental cleanliness reaches the target cleanliness.

[0007] In a possible implementation, before the plurality of parameters are input into the cleanliness prediction model, the method further comprises: pre-training the cleanliness prediction model; wherein the training process of the cleanliness prediction model comprises: obtaining a plurality of environmental parameters in different states in historical data, the plurality of environmental parameters comprising at least two of the following: air exchange frequency, FFU fullness rate, temperature and humidity, pressure gradient, equipment heat dissipation, and air flow organization mode; obtaining single-factor influence information of a single environmental parameter in the plurality of environmental parameters on cleanliness; obtaining coupling influence information of multi-factor mutual coupling between the plurality of environmental parameters on cleanliness; and training the model in combination with each single-factor influence information and the coupling influence information to obtain the cleanliness prediction model.

[0008] In a possible implementation, the obtaining of the single-factor influence information of a single environmental parameter in the plurality of environmental parameters on cleanliness comprises: obtaining single-factor influence data of a target single environmental parameter on cleanliness under different working conditions; and analyzing and comparing the single-factor influence data of the target single environmental parameter on cleanliness under different working conditions, and generating an influence curve of the target single environmental parameter on cleanliness.

[0009] In a possible implementation, the obtaining of the coupling influence information of multi-factor mutual coupling between the plurality of environmental parameters on cleanliness comprises: determining a range value and a variance value of each factor in the plurality of environmental parameters based on the single-factor influence information of each single environmental parameter on cleanliness; determining a significant factor based on the range value of each factor in the plurality of environmental parameters, wherein the range value of each factor in the plurality of environmental parameters is greater than a preset range value; verifying the significance of each significant factor based on the variance value of each factor in the plurality of environmental parameters; generating an interaction effect diagram based on the multi-factor interaction superposition effect, and analyzing the coupling effect between the plurality of factors based on the interaction effect diagram to determine the coupling influence information of multi-factor mutual coupling between the plurality of environmental parameters on cleanliness.

[0010] In a possible implementation, the training of the model in combination with each single-factor influence information and the coupling influence information to obtain the cleanliness prediction model comprises: constructing a cleanliness prediction model by using a response surface method in combination with each single-factor influence information and the coupling influence information, and determining a prediction formula: ; wherein Y is a particulate matter concentration, represents an intercept term, represents a regression coefficient of each factor, represents the i th factor, represents an additional interaction effect on cleanliness when the factors and vary together, represents a random error term; the cleanliness prediction model is trained based on training data until the mean square error and the determination coefficient corresponding to the trained cleanliness prediction model are both greater than corresponding threshold values.

[0011] In a second aspect, the embodiments of the present application provide an environmental cleanliness measurement and control device, comprising: a parameter acquisition module configured to acquire a plurality of parameters of a target environment, the plurality of parameters comprising at least two of the following: air exchange frequency, FFU fullness rate, temperature and humidity, pressure gradient, equipment heat dissipation, and air flow organization mode; a prediction module configured to input the plurality of parameters into a cleanliness prediction model and acquire a control parameter output by the cleanliness prediction model, wherein the control parameter is calculated by the cleanliness prediction model based on the association between the plurality of parameters and coupling information and the target cleanliness; and a control module configured to adjust corresponding environmental parameters of the target environment based on the control parameter output by the cleanliness prediction model.

[0012] In a possible implementation, the environmental cleanliness measurement and control device further comprises an association analysis module configured to train the cleanliness prediction model based on training data and obtain a trained cleanliness prediction model; wherein the training process comprises: acquiring a plurality of environmental parameters in different states in historical data, the plurality of environmental parameters comprising at least two of the following: air exchange frequency, FFU fullness rate, temperature and humidity, pressure gradient, equipment heat dissipation, and air flow organization mode; acquiring single-factor influence information of each of the plurality of environmental parameters on cleanliness; acquiring coupling influence information of multi-factor mutual coupling between the plurality of environmental parameters on cleanliness; and training the model in combination with each of the single-factor influence information and the coupling influence information to obtain the cleanliness prediction model.

[0013] In a possible implementation, training the model in combination with each of the single-factor influence information and the coupling influence information to obtain the cleanliness prediction model comprises: constructing a cleanliness prediction model using a response surface method in combination with each of the single-factor influence information and the coupling influence information, and determining a prediction formula: ; wherein Y is the concentration of particulate matter, represents an intercept term, represents a regression coefficient of each factor, represents the i-th factor, represents a factor and an additional interaction effect on cleanliness when they change together, represents a random error term; the cleanliness prediction model is trained based on training data until the mean square error and the determination coefficient corresponding to the trained cleanliness prediction model are both greater than corresponding threshold values.

[0014] In a third aspect, an embodiment of the present application provides a computing device, comprising: a memory and a processor; the memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions, and the computer executable instructions, when executed by the processor, implement the steps of the environment cleanliness measurement and control method.

[0015] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores computer executable instructions, and the instructions, when executed by a processor, implement the steps of the environment cleanliness measurement and control method.

[0016] An embodiment of the present application provides a technical solution that first analyzes the single factor influence law independently, and then studies the multi-factor coupling mechanism based on the orthogonal experiment method, to construct a complete cleanliness prediction model. Not only can the problem of fragmentation of existing single factor research be effectively solved, but also the influencing factors of clean room cleanliness and their coupling mechanism can be more comprehensively and systematically studied. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is an application scenario diagram provided by an embodiment of the present application; Figure 2 is a flowchart of an environment cleanliness measurement and control method provided by an embodiment of the present application; Figure 3 is a structural diagram of an environment cleanliness measurement and control device provided by an embodiment of the present application; Figure 4 is a structural block diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0018] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details presented herein. In other instances, well-known methods have not been described in detail in order to avoid obscuring the present application.

[0019] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of one or more embodiments of the present application. As used in one or more embodiments of the present application and the accompanying claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in one or more embodiments of the present application and the following claims, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0020] It should be understood that, although the terms first, second, etc. can be employed in describing various information in one or more embodiments of the present application, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another type of information. For example, a first can be termed a second, and, similarly, a second can be termed a first, without departing from the scope of one or more embodiments of the present application. The word "if" as used herein means "when" or "upon" or "in response to the determination" depending on the context.

[0021] For the convenience of understanding the technical solutions of the present application, the following explains the terms related to one or more embodiments of the present application.

[0022] 1. Cleanliness: refers to the amount of dust contained in the air environment. In general, it refers to the number of particles greater than or equal to a certain particle size per unit volume of air. The higher the dust content, the lower the cleanliness, and the lower the dust content, the higher the cleanliness.

[0023] 2. FFU fullness rate: refers to the percentage of the coverage area of the fan filter unit (FFU) to the grid area above the ceiling or work area of the clean room. It directly affects the uniformity of air distribution, cleanliness stability and energy efficiency.

[0024] 3. Air changes: refers to the ratio of the total volume of air sent into the clean room per hour to the volume of the clean room. It directly affects the dilution speed of pollutants, temperature and humidity stability and energy consumption level. Its physical meaning is the number of times (theoretical value) of clean air completely replacing indoor air per hour. For example: ACH = 60 indicates that the air is replaced 60 times per hour.

[0025] 4. Pressure gradient: refers to the air pressure difference formed inside the clean room or between different areas. This difference affects the direction of air flow and cleanliness.

[0026] 5. Airflow organization mode: refers to the flow mode and path of air in the clean room, such as designing three airflow modes for switching experiments: unidirectional flow, non-unidirectional flow, and mixed flow, and releasing NIST standard particles (0.5 μm) in the clean room to test the time required for the particle concentration to drop to the ISO standard under different airflow modes (such as 8 minutes for unidirectional flow and 12 minutes for mixed flow).

[0027] The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0028] Figure 1 The application scenario diagram provided by one embodiment of the present application is shown.

[0029] Referring to Figure 1 , in Figure 1The application scenario provided by the embodiment can be a target environment R1, for example, the target environment can be a clean room. The environment parameters of the target environment R1 (the clean room) are obtained. For example, the obtained environment parameters can include at least two of the following: air exchange frequency, FFU fullness rate, temperature and humidity, pressure gradient, device heat dissipation, and air flow organization mode. Further, the cleanliness of the target environment R1 can be predicted and controlled based on the environment parameters, so that the controlled cleanliness meets the user's use requirements.

[0030] In the related art, in the industries such as semiconductor manufacturing, biomedicine, precision electronics, aerospace, and other high-tech industries, the cleanliness of the application environment is increasingly stringent, especially in the biomedicine field, the microorganism control in the clean room is the key to ensure the sterility of the drug. However, some related clean room cleanliness control schemes mainly focus on determining the linear influence of a single factor (such as air exchange frequency, air flow organization mode, etc.) on cleanliness, lack of systematic analysis of the interaction of multiple factors, so that the cleanliness regulated by the above cleanliness control scheme is prone to deviation, and it is difficult to meet the demand for precise control of the cleanliness of the clean room in the fine field.

[0031] To at least overcome the above technical problems, the present application provides an environment cleanliness measurement and control method. The influence of each single parameter in the environment parameters of the target environment on the environment cleanliness is analyzed in advance, a complete cleanliness influence analysis model is constructed based on the coupling mechanism of at least two parameters in the environment parameters, and further, the complete cleanliness influence analysis model can be used to comprehensively analyze multiple factors and their interactions, accurately predict the cleanliness level of the clean room under different working conditions, provide theoretical support for dynamic optimization control and whole life cycle management of the clean room, and better provide scientific clean room environment control scheme for the service customers, so that the regulation accuracy of the cleanliness of the clean room can meet the application requirements of the users, and the user experience is improved.

[0032] Figure 2 A flowchart of an environment cleanliness measurement and control method provided by an embodiment of the present application is shown.

[0033] Referring to Figure 2 The environment cleanliness measurement and control method provided by the embodiment of the present application can specifically include the following steps: Step 201: obtaining a plurality of parameters of a target environment, the plurality of parameters including at least two of the following: air exchange frequency, FFU fullness rate, temperature and humidity, pressure gradient, device heat dissipation, and air flow organization mode.

[0034] In some embodiments, multiple environmental parameters in a target environment (e.g., a clean room) can be collected simultaneously, and the collection methods for each environmental parameter are described in detail below.

[0035] Regarding the number of air changes In the actual application of the target environment, the number of air changes can be collected by obtaining the number of air changes of the air conditioning system of the target environment. For example, the number of air changes of the target environment can be 5 times / hour, 15 times / hour, 25 times / hour, 35 times / hour, 45 times / hour, or 55 times / hour. When the current number of air changes set by the air conditioning system is 35 times / hour, the number of air changes of the target environment can be obtained as 35 times / hour.

[0036] Regarding the FFU fullness rate In some embodiments, the area value of the top of the target environment (e.g., the ceiling) can be obtained in advance, and the number of fan filter units arranged, so as to calculate the FFU fullness rate. For example, the ceiling of the clean room is divided into a 6x6 grid (with an area of 1 square meter per grid), if the number of fan filter units arranged is 18, the FFU fullness rate can be determined as 50%, if the number of fan filter units arranged is 27, the FFU fullness rate can be determined as 75%, and if the number of fan filter units arranged is 36, the FFU fullness rate can be determined as 100%.

[0037] Regarding temperature and humidity In some embodiments, a certain number of temperature and humidity monitoring devices can be arranged at corresponding target positions in the target environment. For example, the temperature and humidity information of the target environment can be collected by a temperature and humidity sensor.

[0038] Regarding the pressure gradient In some embodiments, the target area (e.g., the Dry Cooling Coil (DCC) outlet section) in the target environment can be arranged in a grid (e.g., divided into a 3x3 grid), and a pressure monitoring device can be configured at each grid, and the turbulent intensity in each grid can be recorded. In one implementation, 10 data can be continuously collected at the monitoring point of each grid to obtain the target turbulent intensity in the grid. Based on the obtained target turbulent intensity of each network, the pressure gradient of the target area can be further determined.

[0039] Regarding the heat dissipation of the equipment In some embodiments, the target environment can be configured with equipment that generates heat during operation, which can affect the cleanliness of the clean room. The heat dissipation, heat plume rising speed, and heat diffusion range of the equipment in operation can be recorded by arranging a heat source detection device (infrared thermal imager) around the equipment.

[0040] Airflow organization In some embodiments, the airflow organization in the cleanroom can be determined by monitoring the airflow pattern and path in the cleanroom. Specifically, the airflow organization can be determined by determining the airflow pattern as unidirectional flow, non-unidirectional flow, or mixed flow.

[0041] S202: inputting a plurality of parameters into the cleanliness prediction model, and obtaining a regulation parameter output by the cleanliness prediction model, wherein the regulation parameter is calculated by the cleanliness prediction model based on the association between the plurality of parameters, the coupling information therebetween, and the target cleanliness.

[0042] In some embodiments, the cleanliness prediction model can be trained by training data, and a trained cleanliness prediction model is obtained. In one embodiment, the training process includes: obtaining historical data as training data, wherein the historical data includes a plurality of environmental parameters in different states, which can include at least two of the following: different air exchange frequencies, different FFU fullness rates, different temperature and humidity, different pressure differences, and different equipment heat dissipation amounts.

[0043] Based on the single environmental parameter in the historical data, the influence of single factor on cleanliness is analyzed, and the coupling influence of multi-factor mutual coupling on cleanliness is analyzed. Thus, the cleanliness prediction model is constructed by combining the influence of single factor on cleanliness and the influence of multi-factor mutual coupling on cleanliness.

[0044] wherein the analysis process of each single factor on cleanliness includes: 1. Analysis of the influence of air exchange frequency on cleanliness The internationally recognized polystyrene (PSL) particles are designed as the standard dust source, and the particle size distribution is designed as 0.3 μm, 0.5 μm, and 1.0 μm to simulate common particulate matter in the cleanroom. The dust source generator is uniformly designed in the four corners and the center area of the cleanroom to ensure uniform initial distribution of particulate matter. The dust source release rate is designed as 10^4 particles per cubic meter per minute, and after 30 minutes of continuous release, it enters the steady-state test phase.

[0045] The experiment is designed with 6 groups of air exchange frequency gradients: 5 times / h (low), 15 times / h (regular), 25 times / h (high), 35 times / h (ultra-high efficiency), 45 times / h (extreme), and 55 times / h (theoretical value) to cover the common range and extreme working conditions in the industry. The variable frequency air conditioning system is designed to dynamically adjust the air exchange frequency, and the wind speed, temperature and humidity, and pressure difference are monitored synchronously to ensure the constancy of other variables.

[0046] The design adopts a high-precision laser particle counter to record the concentration of particulate matter in the entire room every 5 minutes, and focuses on monitoring the removal efficiency of particles larger than 0.5 μm. At the same time, the design combines CFD simulation to establish a three-dimensional clean room model to analyze the effect of air changes on air velocity field, turbulent kinetic energy (TKE), and particle deposition path. By analyzing and comparing the cleanliness data of the clean room under different air changes, the influence of air changes on the cleanliness of the clean room is revealed, and the influence curve of air changes on cleanliness is generated based on the data.

[0047] 2. Analysis of the influence of FFU fullness rate on cleanliness FFU (fan filter unit) fullness rate refers to the arrangement density of FFU in the clean room. The experimental design divides the top of the clean room into a 6x6 grid (with an area of 1 square meter per cell), and FFU (fan filter unit) is arranged according to fullness rates of 50% (18 units), 75% (27 units), and 100% (36 units). Two layout modes, random distribution and uniform distribution, are designed to compare the influence of different strategies on airflow uniformity.

[0048] Particle image velocimetry is used to capture the flow field distribution 0.5 m below the FFU to analyze the relationship between fullness rate and airflow velocity standard deviation (σ). Nine sampling points (3x3 grid) are arranged on the floor of the clean room to monitor the spatial differences in particulate matter concentration simultaneously using a portable particle counter. In addition, a personnel walking simulation device is introduced to test the response capability of the clean room to dynamic pollution sources under different fullness rates. By comparing the cleanliness data of the clean room under different FFU fullness rates, the influence of FFU fullness rate on the cleanliness of the clean room is revealed, and the influence curve of FFU fullness rate on cleanliness is generated based on the data.

[0049] 3. Analysis of the influence of temperature and humidity on cleanliness Four groups of temperature and humidity combination experiments are designed (low temperature and low humidity (18℃ / 30%RH), low temperature and high humidity (18℃ / 60%RH) high temperature and low humidity (28℃ / 30%RH), high temperature and high humidity (28℃ / 60%RH), constant temperature and humidity air conditioning is used to accurately control the parameters, with a fluctuation range of ±0.5℃ / ±2%RH.

[0050] The settling process of particles on the glass substrate was recorded by a high-speed camera to compare and analyze the influence of temperature and humidity on particle adhesion. The surface electrostatic potential under different humidity was measured by a surface electrometer, and a model of electrostatic adsorption and humidity correlation was established. In addition, the secondary dust raised by simulating personnel activities (such as equipment vibration and door opening) was tested to test the duration of secondary dust and the particle resuspension rate under different temperature and humidity. By comparing the cleanliness data of the cleanroom under different temperature and humidity conditions, the influence of temperature and humidity on the cleanliness of the cleanroom was revealed, and the influence curve of temperature and humidity single factor on the cleanliness was generated based on the data.

[0051] 4. Analysis of the influence of pressure gradient single factor on cleanliness A gas flow regulation system was constructed using DCC and FFU, and the airflow distribution characteristics of the experimental area under different FFU speeds were analyzed through multi-condition testing to provide data support for optimizing laboratory airflow organization.

[0052] At least 10 groups of FFU speed gradients (10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100% rated speed) were set. A 3x3 grid of measurement points was set at the outlet section of the DCC, and 10 consecutive data were collected at each measurement point to obtain the average value, and the turbulence intensity was recorded. Pressure gauges were arranged at the inlet and outlet of the DCC and the upstream and downstream of the FFU to record the dynamic pressure loss. A database correlating FFU group speed, pressure, and wind speed was established, and the influence curve of pressure gradient single factor on cleanliness was generated based on the data. The recommended value of DCC pressure compensation under non-uniform FFU load was proposed. The airflow organization cloud map of the experimental area was drawn, and the low-speed / vortex area was identified.

[0053] 5. Analysis of the influence of equipment heat dissipation amount single factor on cleanliness Equipment heat dissipation is one of the important heat sources in the cleanroom. The experimental design used an electric heating plate (power 500W, 1000W, 1500W) to simulate process equipment heat dissipation, and the surface temperature was controlled at 50℃~80℃. An infrared thermal imager (FLIRT1030sc) was used to record the rising speed and diffusion range of the thermal plume. At the same time, in order to track the migration path of particles, a particle generator was designed above the heat source to release fluorescent tracer particles, and PIV technology was used to track the motion trajectory of particles driven by thermal convection. Local exhaust hoods (air speed 0.3m / s~0.5m / s) were added around the heat source to test the suppression effect of thermal disturbance under different air speeds. By comparing the cleanliness data of the cleanroom under different process equipment heat dissipation amounts, the influence of process equipment heat dissipation on the cleanliness of the cleanroom was revealed, and the influence curve of equipment heat dissipation single factor on cleanliness was generated based on the data.

[0054] 6. Analysis of the influence of airflow organization method single factor on cleanliness Airflow organization refers to the flow and path of air in a cleanroom. Three airflow modes are designed for switching experiments: unidirectional flow, non-unidirectional flow, and mixed flow. NIST standard particles (0.5 μm) are released in the cleanroom to test the time required for the particle concentration to decrease to the ISO standard under different airflow modes (e.g., 8 minutes for unidirectional flow and 12 minutes for mixed flow). At the same time, smoke visualization is designed to compare the airflow dead angle area under different modes to measure the energy consumption of the fan under different modes. Combined with the cleanliness compliance rate, the energy consumption cost per cleanliness is calculated. By comparing the cleanliness data of the cleanroom under different airflow organization modes, the influence of airflow organization on the cleanliness of the cleanroom is revealed, and the influence curve of airflow organization on cleanliness is generated based on the data.

[0055] After obtaining the influence curve of each major factor on cleanliness, the single-factor experiment results are obtained.

[0056] In some embodiments, after obtaining the influence curve of each single factor on cleanliness, a regression equation can be created based on the least squares method, as follows: The relationship between cleanliness (dependent variable Y) and environmental parameters (independent variables X1, Z2, …, X n ) can be represented as: Y = β0 + β1X1 + β2X2 + … + β n X n + ϵ; Wherein, the independent variables X1, Z2, …, X n include at least two of the number of air changes, FFU fullness rate, temperature and humidity, pressure gradient (pressure difference), and device heat dissipation, β0 is the intercept term, β i is the regression coefficient of each parameter, and ϵ is the random error term.

[0057] Further, the least squares method can be used for optimization: The goal is to minimize the sum of squared residuals (SS~res~): min∑ i =1 m ( yi - y ^ i )2; Wherein, yi is the measured cleanliness, and y^i is the model prediction value.

[0058] The following will be described in detail through specific examples, which can be related to the particle concentration prediction example of the cleanroom: Wherein, the variables (environmental parameters) can include: the number of air changes (X1), the number of personnel (X2), and the device heat dissipation (X3).

[0059] The fitting equation can be obtained as follows: PM0.5= 2.1 + 0.05X1- 1.8X2+ 0.3X3- 0.02X1X2 The specific effects are: R2= 0.95 (interaction model) vs. 0.73 (no interaction model).

[0060] According to the equation, it can be determined that the interaction of personnel activity and air exchange frequency significantly affects cleanliness.

[0061] Therefore, in the regulation optimization process, the parameter combination under the target cleanliness can be deduced by the regression equation, for example, when the target cleanliness: PM~0.5~≤ 5 mg / m3, the optimal air exchange frequency is 35 times / h and the upper limit of personnel is 2.

[0062] Among them, the analysis process of the influence of multi-factor coupling on cleanliness includes: In some embodiments, after obtaining the influence information of single factors on cleanliness, the influence size of each factor can be further analyzed. Specifically, in an implementation, the range of each factor can be calculated to determine the influence of each factor, and the factors can be sorted based on a preset order, for example, sorted in descending order of influence, wherein the greater the range, the more significant the influence of the factor on cleanliness. Therefore, the influence size of each single factor on cleanliness can be sorted according to the range, wherein the factor with a range greater than a preset range can be set as a significant factor or a first target factor.

[0063] In some embodiments, after obtaining the influence information of single factors on cleanliness, the non-significant factors among the multiple environmental factors can be further removed according to the significance of the influence of each factor on cleanliness. Specifically, in an implementation, the variance of each factor can be calculated to distinguish the factor influence and random error, and to verify the statistical significance. In some other implementations, the way of selecting significant factors can also include determining that the variance less than a preset variance is a non-significant factor, and then the significant factors can be screened by removing the non-significant factors.

[0064] In some embodiments, after the above range analysis and variance analysis, the ranking of the influence of each factor on the cleanliness and the proposal of non-significant factors are completed, and then multi-factor interaction effect analysis can be performed, which can specifically include analysis of non-additive effects when at least two factors jointly act. It should be noted that when two independent variables (for example, the number of air changes A and the temperature B) jointly act, their effect on the dependent variable (cleanliness Y) is not equal to the algebraic sum of their independent effects (i.e., A+B ≠ A∩B), then it is called that there is an interaction effect. In an implementation, the multi-factor interaction effect analysis can include generating an interaction effect diagram based on the multi-factor interaction additive effect, so that the coupling action between the multi-factors can be analyzed based on the generated interaction effect diagram.

[0065] For example, by using an orthogonal experiment method, a multi-factor coupling experiment can be designed, and four key factors that have a greater impact on the cleanliness of the clean room can be selected as experimental factors. Specifically, the specific data of the four key factors can include: the number of air changes (20 times / h, 25 times / h, 30 times / h), the FFU full coverage rate (70%, 85%, 100%), the temperature (20°C, 24°C, 28°C), and the pressure difference (5 Pa, 10 Pa, 15 Pa). Further, an L9(3^4) orthogonal table can be used to arrange the experimental combinations, each experimental group is repeated three times, a total of 27 experiments are performed to ensure data reliability, and experimental data of the 27 experiments are obtained. On the other hand, the variance analysis (ANOVA) can be used to calculate the significance of the main effect and the interaction effect of each factor, for example, the interaction effect significance threshold p < 0.05, in other words, the confidence level of the existence of the interaction effect needs to reach 95%, so as to obtain the orthogonal experimental data.

[0066] In some embodiments, after obtaining the single-factor experiment results and the orthogonal experimental data, a model training can be performed based on the single-factor experiment results and the orthogonal experimental data, and after the corresponding training, a cleanliness prediction model under the influence of multi-factor coupling can be constructed. In an implementation, the response surface method (RSM) can be used to construct the cleanliness prediction model to determine the prediction formula: ; Y is the particulate matter concentration, represents an intercept term, represents the regression coefficient of each factor, represents the i-th factor, represents the factor and the additional interaction effect on the cleanliness when the factors change together, represents a random error term, and the CFD simulation is further analyzed to analyze the three-dimensional flow field characteristics, and a variable frequency control algorithm is introduced to improve the system response speed, so that the model learns the interaction relationship and influence path between the factors, and the influence law of the cleanliness of the clean room when they jointly act.

[0067] In some embodiments, the obtained single-factor experiment results and corresponding orthogonal experiment data can be divided into a training set and a test set to respectively implement training and testing of the model. In an implementation, 70% of the data can be used as the training set, and the remaining 30% of the data can be used as the test set. Finally, a cleanliness prediction model with a performance evaluation index greater than a preset condition is obtained through training of the model. In an implementation, the preset condition can include that the mean square error (MSE) and the determination coefficient (R2) corresponding to the model are both greater than the corresponding threshold, so that the trained model meets the user's preset demand. )are both greater than the corresponding threshold, so that the trained model meets the user's preset demand.

[0068] In some embodiments, when the trained model fails to meet the above-mentioned preset condition, the trained model can be optimized accordingly until the optimized model meets the preset condition. In an implementation, the optimization method of the model can include removing non-significant variables through stepwise regression to improve the accuracy of the model.

[0069] The above is the training process of the cleanliness prediction model provided by the embodiments of the present application.

[0070] In some embodiments, after obtaining the trained cleanliness prediction model, the cleanliness of the target environment can be predicted and regulated based on the model. Specifically, after obtaining the current environmental parameters of the target environment based on S201, the obtained target parameters can be input into the cleanliness prediction model, and then the cleanliness prediction model can calculate the predicted cleanliness in a future period of time through a prediction formula, thereby completing the cleanliness prediction operation.

[0071] In some embodiments, when the predicted cleanliness output by the cleanliness prediction model exceeds the target cleanliness, the corresponding parameters in the target environment need to be regulated, so that the predicted cleanliness corresponding to the regulated parameters falls within the target cleanliness. In an implementation, when the predicted cleanliness output by the cleanliness prediction model exceeds the target cleanliness, the cleanliness prediction model is triggered to calculate the regulation parameters. Specifically, the cleanliness prediction model can calculate the regulation parameters based on the input multiple environmental parameters, the coupling information therebetween, and the association relationship between the target cleanliness. In an implementation, the target cleanliness can be substituted into the prediction formula to respectively inversely calculate the target values of the environmental parameters (Xi). Further, the regulation parameters of the environmental parameters can be determined based on the difference between the current values and the target values of the environmental parameters.

[0072] S203: Adjusting the corresponding environmental parameters of the target environment based on the regulation parameters output by the cleanliness prediction model.

[0073] In some embodiments, after obtaining the regulation parameter output by the cleanliness prediction model, the corresponding environmental parameter of the target environment (such as a clean room) can be adjusted based on the regulation parameter, so that the cleanliness of the target environment after parameter adjustment reaches the target cleanliness, meeting the user application requirements.

[0074] The above-mentioned environmental cleanliness measurement and control method provided by the embodiments of the present application first independently analyzes the single-factor influence law, and then studies the multi-factor coupling mechanism based on the orthogonal experiment method, and constructs a complete cleanliness prediction model. Not only can the problem of fragmentation of existing single-factor research be effectively solved, but also the influencing factors of the cleanliness of the clean room and the coupling mechanism thereof can be more comprehensively and systematically studied.

[0075] Corresponding to the above-mentioned method embodiments, the present application also provides environmental cleanliness measurement and control device embodiments.

[0076] Figure 3 A structural schematic diagram of an environmental cleanliness measurement and control device provided by an embodiment of the present application is shown in the figure.

[0077] Referring to Figure 3 The device comprises: The parameter acquisition module 301 is configured to acquire a plurality of parameters of a target environment, the plurality of parameters comprising at least two of the following: air exchange frequency, FFU fullness rate, temperature and humidity, pressure gradient, equipment heat dissipation, and air flow organization mode.

[0078] The prediction module 302 is configured to input the plurality of parameters into a cleanliness prediction model, and acquire a regulation parameter output by the cleanliness prediction model, wherein the regulation parameter is calculated by the cleanliness prediction model based on the association between the plurality of parameters, the coupling information therebetween, and the target cleanliness.

[0079] The regulation module 303 is configured to adjust the corresponding environmental parameter of the target environment based on the regulation parameter output by the cleanliness prediction model.

[0080] In some embodiments, Figure 3 The device shown in the figure further comprises an association analysis module 304 configured to train the cleanliness prediction model based on the training data, and obtain a trained cleanliness prediction model, and the training process is the same as that of the cleanliness prediction model shown in Figure 2 The training process of the cleanliness prediction model is the same as that of the cleanliness prediction model shown in the figure, and will not be described again.

[0081] The above is a schematic scheme of an environmental cleanliness measurement and control device according to the present embodiment. It should be noted that the technical scheme of the environmental cleanliness measurement and control device belongs to the same concept as the technical scheme of the above-mentioned environmental cleanliness measurement and control method, and the details of the technical scheme of the environmental cleanliness measurement and control device that are not described in detail can be referred to the description of the technical scheme of the above-mentioned environmental cleanliness measurement and control method.

[0082] Figure 4 A structural block diagram of a computing device 400 is shown, according to one embodiment of the present application. The components of the computing device 400 include, but are not limited to, a processor 410 and a memory 420. In other embodiments, the processor 410 and the memory 420 can also be connected by a bus, and configured with a database for storing data.

[0083] The computing device 400 also includes an access device that enables the computing device to communicate via one or more networks. Examples of such networks include the public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or combinations of such networks, such as the Internet. The access device 440 can include one or more of any type of network interface (for example, a network interface card (NIC)), wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC).

[0084] In one embodiment of the present application, the above-mentioned components of the computing device 400, as well as other components not shown in the figure, can be connected to each other, for example, through a bus. It should be understood that the computing device structure block diagram shown is for the purpose of example only, and is not a limitation on the scope of the present application. Other components can be added or replaced as needed by those skilled in the art. Figure 4 Figure 4 The computing device structure block diagram shown is for the purpose of example only, and is not a limitation on the scope of the present application. Other components can be added or replaced as needed by those skilled in the art.

[0085] ​The computing device 400 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 400 can also be a mobile or stationary server.

[0086] The processor 410 is configured to execute computer-executable instructions to perform the steps of the method for measuring and controlling the environmental cleanliness described above. The above is a schematic solution of the computing device according to an embodiment. It should be noted that the technical solution of the computing device and the technical solution of the method for measuring and controlling the environmental cleanliness belong to the same concept, and the details of the technical solution of the computing device not described in detail can be referred to the description of the technical solution of the method for measuring and controlling the environmental cleanliness.

[0087] An embodiment of the present application further provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method for measuring and controlling the environmental cleanliness.

[0088] The above is a schematic solution of the computer-readable storage medium according to an embodiment. It should be noted that the technical solution of the computer-readable storage medium and the technical solution of the method for measuring and controlling the environmental cleanliness belong to the same concept, and the details of the technical solution of the computer-readable storage medium not described in detail can be referred to the description of the technical solution of the method for measuring and controlling the environmental cleanliness.

[0089] An embodiment of the present application further provides a computer program, which, when executed in a computer, causes the computer to perform the steps of the method for measuring and controlling the environmental cleanliness.

[0090] The above is a schematic solution of the computer program according to an embodiment. It should be noted that the technical solution of the computer program and the technical solution of the method for measuring and controlling the environmental cleanliness belong to the same concept, and the details of the technical solution of the computer program not described in detail can be referred to the description of the technical solution of the method for measuring and controlling the environmental cleanliness.

[0091] The above-described embodiments of the application have several aspects, no single one of which is solely responsible for the application's desirable attributes. Without limiting the scope of the application as expressed by the claims which follow, some further embodiments make these aspects even more useful. Other embodiments can result in less desirable attributes.

[0092] The computer readable medium can include any entity or apparatus capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier signal, telecommunication signal, software distribution medium, etc. It should be noted that the computer readable medium can include appropriate contents according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0093] It should be noted that for the foregoing method embodiments, the acts described therein can be performed in a different order from the order described, and that certain acts can be performed in parallel or concurrently. In addition, certain acts can be omitted, and other acts can be added. Furthermore, the acts described in the specification can be implemented as a set of computer readable instructions, software, or code stored in a computer readable medium or memory of a computer (e.g., RAM, ROM, EEPROM, flash memory, or the like) such that the various aspects of the present application can be incorporated into a computer readable medium or memory of a computer.

[0094] In the above embodiments, the description of each embodiment is focused on different aspects, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0095] The preferred embodiments of the application disclosed above are only used to help explain the application. Alternative embodiments do not describe all the details, nor limit the application to only the specific embodiments described. Obviously, according to the content of the embodiments of the application, many modifications and changes can be made. The embodiments of the application are selected and described in order to better explain the principles and practical applications of the embodiments of the application, so that those skilled in the art can well understand and use the application. The application is limited by the claims and their full scope and equivalents.

Claims

1. An environmental cleanliness monitoring method, characterized by, The method comprises the following steps: acquiring a plurality of parameters of a target environment, the plurality of parameters comprising at least two of the following: air change rate, FFU fullness rate, temperature and humidity, pressure gradient, equipment heat dissipation, and air flow organization mode; inputting the plurality of parameters into a cleanliness prediction model, and acquiring a regulation parameter output by the cleanliness prediction model, wherein the regulation parameter is calculated based on the association between the plurality of parameters and the coupling information and the target cleanliness; adjusting a corresponding parameter in the plurality of parameters to a target value based on the regulation parameter, so that the cleanliness of the target environment reaches the target cleanliness.

2. The method of claim 1, wherein, Before the plurality of parameters are input into the cleanliness prediction model, the method further comprises pre-training the cleanliness prediction model; wherein the training process of the cleanliness prediction model comprises: acquiring a plurality of environmental parameters in different states in historical data, the plurality of environmental parameters comprising at least two of the following: air change rate, FFU fullness rate, temperature and humidity, pressure gradient, equipment heat dissipation, and air flow organization mode; respectively acquiring single-factor influence information of a single environmental parameter in the plurality of environmental parameters on cleanliness; acquiring coupling influence information of multi-factor mutual coupling between the plurality of environmental parameters on cleanliness; training the model in combination with each single-factor influence information and the coupling influence information to obtain the cleanliness prediction model.

3. The method of claim 2, wherein, The acquiring of the single-factor influence information of the single environmental parameter in the plurality of environmental parameters on cleanliness comprises: acquiring single-factor influence data of a target single environmental parameter on cleanliness under different working conditions; analyzing and comparing the single-factor influence data of the target single environmental parameter on cleanliness under different working conditions, and generating an influence curve of the target single environmental parameter on cleanliness.

4. The method of claim 2, wherein, The acquiring of the coupling influence information of multi-factor mutual coupling between the plurality of environmental parameters on cleanliness comprises: determining a range value and a variance value of each factor in the plurality of environmental parameters based on the single-factor influence information of each single environmental parameter on cleanliness; determining a significant factor based on the range value of each factor in the plurality of environmental parameters, wherein the range value of the significant factor is greater than a preset range value; verifying the significance of each significant factor based on the variance value of each factor in the plurality of environmental parameters; generating an interaction effect diagram based on the multi-factor interaction superposition effect, and analyzing the coupling effect between the plurality of factors based on the interaction effect diagram to determine the coupling influence information of the multi-factor mutual coupling on cleanliness.

5. The method according to any one of claims 2-4, characterized in that, The training of the model in combination with each single-factor influence information and the coupling influence information to obtain the cleanliness prediction model comprises: combining each single-factor influence information and the coupling influence information, and constructing a cleanliness prediction model by using a response surface method, and determining a prediction formula: where Y is the particulate matter concentration, represents the intercept term, represents the regression coefficient of each factor, represents the i-th factor, represents the factor and the additional interaction effect on cleanliness when the two change together, represents the random error term; training the cleanliness prediction model based on training data until the mean square error and the determination coefficient corresponding to the trained cleanliness prediction model are greater than corresponding threshold values.

6. An environmental cleanliness monitoring device, characterized by, The method comprises the following steps: The parameter acquisition module is configured to acquire a plurality of parameters of the target environment, the plurality of parameters including at least two of the following: air exchange rate, FFU fullness rate, temperature and humidity, pressure gradient, equipment heat dissipation, and airflow organization mode; The prediction module is configured to input the plurality of parameters into a cleanliness prediction model, and acquire a regulation parameter output by the cleanliness prediction model, wherein the regulation parameter is calculated based on an association between the plurality of parameters and the coupling information and the target cleanliness. The regulation module is configured to adjust a corresponding environmental parameter of the target environment based on the regulation parameter output by the cleanliness prediction model.

7. The apparatus of claim 6, wherein, The environment cleanliness measurement and control device further includes: The association analysis module is configured to train the cleanliness prediction model based on the training data, and obtain a trained cleanliness prediction model. The training process includes: acquiring a plurality of environmental parameters in different states in the historical data, the plurality of environmental parameters including at least two of the following: air exchange rate, FFU fullness rate, temperature and humidity, pressure gradient, equipment heat dissipation, and airflow organization mode; acquiring single-factor influence information of each of the plurality of environmental parameters on cleanliness; acquiring coupling influence information of multi-factor mutual coupling between the plurality of environmental parameters on cleanliness; training the model in combination with each of the single-factor influence information and the coupling influence information to obtain the cleanliness prediction model.

8. The apparatus of claim 7, wherein, Training the model in combination with each of the single-factor influence information and the coupling influence information to obtain the cleanliness prediction model includes: constructing the cleanliness prediction model using a response surface method in combination with each of the single-factor influence information and the coupling influence information, and determining a prediction formula: where Y is the particulate matter concentration, represents the intercept term, represents the regression coefficient of each factor, represents the i-th factor, represents the factor and the additional interaction effect on cleanliness when the two change together, represents the random error term; training the cleanliness prediction model based on the training data until the mean square error and the determination coefficient corresponding to the trained cleanliness prediction model are greater than the corresponding threshold values.

9. A computing device, comprising: It includes: a memory and a processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions, which realize the steps of the environment cleanliness measurement and control method in any one of claims 1 to 7 when executed by the processor.

10. A computer readable storage medium storing computer executable instructions, which realize the steps of the environment cleanliness measurement and control method in any one of claims 1 to 7 when executed by a processor.

Images