A dust-free workshop automation control method and system based on data collection and analysis

Abstract

The present invention discloses a dust-free workshop automation control method and system based on data collection and analysis, which relates to the technical field of workshop dust removal. The method collects and pre-processes environmental data of the dust-free workshop, constructs a three-dimensional model of the dust-free workshop and divides the dust-free workshop into grids, obtains the air quality score AQS of each grid area for initial evaluation, constructs a first pollution source candidate grid area set SL, and performs secondary evaluation on the grid areas within the first pollution source candidate grid area set SL, constructs a second pollution source candidate grid area set SP, and obtains the particle rise rate UPR of each grid area within the second pollution source candidate grid area set SP, determines the location of the pollution source, triggers an automatic response mechanism, removes the pollution source, solves the problem of difficulty in locating the pollution source, optimizes the air quality management of the dust-free workshop, enhances the safety of the dust-free workshop, and ensures the continuous cleanliness and stable operation of the dust-free workshop environment.

Description

Technical Field

[0001] The present invention relates to the technical field of dust removal in workshops, and in particular to a method and system for automatic control of dust-free workshops based on data collection and analysis. Background Art

[0002] In industrial production, clean rooms, as key facilities in fields such as precision manufacturing and biopharmaceuticals, have very strict requirements on air quality. To ensure that the production environment of clean rooms meets quality standards, real-time monitoring and precise management of air quality are required. In current monitoring technologies, common practices include static monitoring and regular inspections. Although they can provide certain environmental data, they have obvious limitations.

[0003] In the Chinese invention patent with application publication number CN118482465A, a graded ventilation method and system for a dust-free workshop is disclosed. The compensation method obtains dust-free workshop image data and workshop level; divides the dust-free workshop into dust-free workshop areas and area levels based on the dust-free workshop image data and workshop level; determines whether the environmental parameters of the dust-free workshop meet the ventilation standards corresponding to the workshop level; when the environmental parameters of the dust-free workshop do not meet the ventilation standards corresponding to the workshop level, determines whether the environmental parameters of the dust-free workshop areas meet the ventilation standards corresponding to the area level one by one; when the environmental parameters of the dust-free workshop area do not meet the ventilation standards corresponding to the area level, adjusts the environmental parameters of the current dust-free workshop area based on the ventilation standards corresponding to the area level.

[0004] The above compensation method realizes graded ventilation in different areas of the dust-free workshop through hierarchical management to meet the different requirements of air quality in different areas and reduce energy consumption at the same time.

[0005] However, existing technologies mainly rely on the monitoring of workshop environmental data and usually lack the ability to track and accurately locate pollution sources in real time. In addition, many existing pollution source detection methods rely on manual judgment or experience-based threshold setting, which makes it difficult to effectively realize automated pollution source precision positioning and removal operations.

[0006] To this end, the present invention provides a dust-free workshop automation control method and system based on data collection and analysis. Summary of the invention

[0007] In view of the deficiencies in the prior art, the present invention provides a dust-free workshop automation control method and system based on data collection and analysis, which solves the problems in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: A dust-free workshop automation control method based on data collection and analysis, comprising the following steps:

[0009] S1. Set up data collection nodes in the dust-free workshop and deploy intelligent sensor groups at the data collection nodes to monitor the changes in the dust-free workshop environment status in real time and collect dust-free workshop environment data;

[0010] S2. Preprocess the collected clean room environment data, including denoising, filling missing values ​​and outlier processing, and construct a clean room environment data set based on the preprocessed clean room environment data;

[0011] S3, construct a three-dimensional model of the clean room, divide the clean room into n grid areas, preset an air quality threshold AQL, and obtain an air quality score AQS for each grid area in the clean room based on the clean room environment data in the clean room environment data set, and compare and analyze the air quality score AQS of each grid area with the air quality threshold AQL to construct a first pollution source candidate grid area set SL;

[0012] S4. According to the first pollution source candidate grid area set SL, obtain the particle concentration mean μ and particle concentration variance σ of each grid area in the first pollution source candidate grid area set SL, and evaluate each grid area in the first pollution source candidate grid area set SL according to the particle concentration mean μ and particle concentration variance σ, and construct a second pollution source candidate grid area set SP;

[0013] S5. Based on the second pollution source candidate grid area set SP, obtain the particle rise rate UPR in each grid area in the second pollution source candidate grid area set SP, and locate the grid area where the pollution source is located based on the particle rise rate UPR in each grid area, trigger the automatic response mechanism, and perform pollution source removal operations.

[0014] Preferably, the specific steps of S1 include:

[0015] S11. Setting a data collection node in the dust-free workshop, and deploying an intelligent sensor group at the data collection node, setting the data collection frequency of the intelligent sensor group to once per second, monitoring the environmental changes of the dust-free workshop in real time through the intelligent sensor group, recording the environmental data of the dust-free workshop, and transmitting the environmental data of the dust-free workshop to the central control system of the dust-free workshop through the wireless network equipment carried by the intelligent sensor;

[0016] The clean room environment data includes particle concentration, air flow rate and humidity data.

[0017] Preferably, the specific steps of S2 include:

[0018] S21, preprocessing the clean room environment data collected in S1, including denoising, filling missing values ​​and outlier processing;

[0019] The denoising is to apply Kalman filtering to denoise the collected clean room environment data;

[0020] The missing values ​​are filled by using the mean interpolation method to fill the missing data in the collected clean room environment data with the mean of the known data at the time point where the missing data is located;

[0021] The outlier processing is based on the collected cleanroom environment data, sorting the cleanroom environment data, recording the 25% quantile Q1 and the 75% quantile Q3 of the cleanroom environment data, and obtaining the interquartile range IQR based on the 25% quantile Q1 and the 75% quantile Q3 of the cleanroom environment data, and defining the range of outliers according to the interquartile range IQR as follows: , for those not in the range The data within are considered as outliers and removed;

[0022] S22: construct a clean room environment data set based on the pre-processed clean room environment data.

[0023] Preferably, the specific steps of S3 include:

[0024] S31, collecting three-dimensional data of the clean room, and using the three-dimensional modeling software Unity to construct a three-dimensional model of the clean room according to the collected three-dimensional data of the clean room, and dividing the three-dimensional model of the clean room into n grid areas to obtain a three-dimensional grid model of the clean room;

[0025] S32: for each grid area in the clean room, obtain the particle standard deviation coefficient SD and the average humidity change rate RH according to the clean room environment data in the pre-processed clean room environment data set. avg and the gas stagnation factor LVI;

[0026] The particle standard deviation coefficient SD is obtained as follows:

[0027] In the formula, C k is the particle concentration value collected by the kth data collection node, C avg is the mean of the particle concentration values ​​collected by all data collection nodes in each grid area, m is the total number of data collection nodes in each grid area, k=1, 2, 3, ..., m;

[0028] The average humidity change rate RH avg The method of obtaining is:

[0029] Where u is the total number of times each data collection node collects data, ΔRH k(t) is the humidity change rate of the kth data collection node at time t, t=1, 2, 3, ..., u;

[0030] The gas stagnation coefficient LVI is obtained in the following way:

[0031] Where, LVI j is the gas stagnation coefficient of the jth grid area, v j，max is the maximum air velocity in the jth grid area, v avg，j is the average air velocity in the jth grid area;

[0032] S33, based on the obtained particle standard deviation coefficient SD and average humidity change rate RH of each grid area avg and the gas stagnation coefficient LVI, to obtain the air quality score AQS of each grid area, wherein the air quality score AQS is obtained as follows:

[0033] Where α1, α2 and α3 are the particle standard deviation coefficient SD, the average humidity change rate RH, respectively. avg And the weight coefficient of the gas stagnation coefficient LVI.

[0034] Preferably, the specific step S3 further includes:

[0035] S34, preset air quality threshold AQL, compare and analyze the air quality score AQS obtained in S33 with the preset air quality threshold AQL to evaluate the pollution status of each grid area in the clean room. The specific evaluation steps are as follows:

[0036] S341. If the air quality score AQS is less than the preset air quality threshold AQL, that is, AQS<AQL, it is determined that the air quality in the grid area is within a safe range, that is, the clean room is not polluted, and the clean room environmental data is recorded and the clean room is continuously monitored;

[0037] S342. If the air quality score AQS is greater than or equal to the preset air quality threshold AQL, that is, AQS ≥ AQL, it is determined that the air quality in the grid area is within the dangerous range, that is, the clean room is polluted, and the alarm mechanism is triggered, and the alarm is continuously issued to notify the clean room operators and managers until the clean room operators and managers respond;

[0038] S35. Construct a first pollution source candidate grid area set SL based on the grid areas corresponding to the air quality scores AQS obtained in S342 being greater than or equal to the preset air quality threshold AQL, where SL=(SL1, SL2, SL3, ..., SL g ).

[0039] Preferably, the specific step of S4 includes:

[0040] S41. For each grid area in the first pollution source candidate grid area set SL, according to the clean room environment data in the clean room environment data set constructed in S22, set the time sliding window to x seconds and the step size to 1s, and in the current time sliding window, based on the clean room environment data, obtain the particle concentration mean μ and particle concentration variance σ in the current time sliding window.

[0041] Preferably, the specific step S4 further includes:

[0042] S42: Perform a secondary assessment of the pollution status of the clean room based on the obtained particle concentration mean μ and particle concentration variance σ, and analyze the pollution status of each grid area in the first pollution source candidate grid area set SL. The specific assessment steps are as follows:

[0043] S421, if the mean particle concentration μ of the sliding window at the current time c Greater than or equal to the mean particle concentration μ of the sliding window at the previous time p Add twice the particle concentration variance σ p , that is, μ c ≥μ p +2×σ p When , the current time sliding window is judged to be in an abnormal state, and the grid area where the time sliding window is in an abnormal state is marked as an abnormal area;

[0044] S422, if the mean particle concentration μ of the sliding window at the current time c Less than the mean particle concentration μ of the sliding window at the previous time p Add twice the particle concentration variance σ p , that is, μ c <μ p +2×σ p When the current time sliding window is judged to be in a normal state, the data is continuously monitored and recorded;

[0045] S43: Based on the grid areas marked as abnormal areas in S421, a second set of candidate grid areas for pollution sources SP is constructed, where SP = (SP1, SP2, SP3, ..., SP h ).

[0046] Preferably, the specific step of S5 includes:

[0047] S51, based on each grid area in the second pollution source candidate grid area set SP constructed in S43, obtain the particle rise rate UPR in each grid area, and the particle rise rate UPR is obtained in the following manner:

[0048] Where C(t) is the particle concentration value in the grid area at time t, △t is the data collection time interval, and C(t-△t) is the particle concentration value at the previous time point that is △t apart from the current time t.

[0049] S52. Based on the obtained particle rise rate UPR in each grid area, select the grid area with the largest particle rise rate UPR as the grid area where the pollution source is located.

[0050] Preferably, the specific step S5 further includes:

[0051] S53, triggering an automatic response mechanism according to the grid area where the pollution source is located obtained in S52, the specific content of the automatic response mechanism is as follows:

[0052] For the grid area where the pollution source is located, the local ventilation system is automatically turned on, the ventilation system power is increased, powerful fans and dust collectors are used to increase the air flow in the polluted area, and the automatic vacuum robot is started to clean the grid area where the pollution source is located. At the same time, the pollution source information is sent to the dust-free workshop technicians to eliminate the pollution source;

[0053] For grid areas outside the grid area where the pollution source is located, enhance the power of the overall ventilation system of the clean room, increase the ventilation frequency of the clean room, start the air purification equipment to remove pollution, and adjust the wind direction and wind speed of the clean room through the wind direction control system of the clean room to block the grid area of ​​the clean room where pollution occurs.

[0054] A dust-free workshop automation control system based on data collection and analysis, including a data collection module, a preprocessing module, a primary evaluation module, a secondary evaluation module and a pollution source positioning and automatic response module;

[0055] The data acquisition module is used to set up data acquisition nodes in the dust-free workshop, and deploy intelligent sensor groups at the data acquisition nodes to monitor the changes in the environmental state of the dust-free workshop in real time and collect the environmental data of the dust-free workshop;

[0056] The preprocessing module is used to preprocess the collected clean room environment data, including denoising, filling missing values ​​and outlier processing, and construct a clean room environment data set based on the preprocessed clean room environment data;

[0057] The initial assessment module is used to construct a three-dimensional model of a clean room, divide the clean room into n grid areas, preset an air quality threshold AQL, and obtain an air quality score AQS for each grid area in the clean room based on the clean room environment data in the clean room environment data set, and compare and analyze the air quality score AQS of each grid area with the air quality threshold AQL to construct a first pollution source candidate grid area set SL;

[0058] The secondary evaluation module is used to obtain the particle concentration mean µ and particle concentration variance σ of each grid area in the first pollution source candidate grid area set SL according to the first pollution source candidate grid area set SL, and evaluate each grid area in the first pollution source candidate grid area set SL according to the particle concentration mean µ and particle concentration variance σ, and construct the second pollution source candidate grid area set SP;

[0059] The pollution source positioning and automatic response module is used to obtain the particle rise rate UPR in each grid area in the second pollution source candidate grid area set SP based on the second pollution source candidate grid area set SP, and locate the grid area where the pollution source is located based on the particle rise rate UPR in each grid area, trigger the automatic response mechanism, and perform pollution source removal operations.

[0060] The present invention provides a dust-free workshop automation control method based on data collection and analysis, which has the following beneficial effects:

[0061] (1) The present invention can effectively improve the accuracy of pollution source positioning through real-time monitoring and data acquisition technology combined with the three-dimensional grid model of the dust-free workshop. In the S1 stage, an intelligent sensor group is deployed to continuously collect environmental data in the dust-free workshop to ensure the timeliness and coverage of the data. Through the three-dimensional grid model constructed in the S3 stage, the air quality score AQS of each grid area can reflect the pollution status of the area in real time. Combined with the air quality threshold AQL for preliminary screening, the grid areas with higher pollution risks can be quickly identified, thereby providing an accurate basis for the subsequent particle concentration analysis. On this basis, the particle concentration mean μ and particle concentration variance σ can help judge the intensity and volatility of the pollution source, providing data support for the accurate positioning of the pollution source.

[0062] (2) By combining the particle rise rate (UPR) and the automatic response mechanism, a rapid and automatic response to pollution sources can be made in the S5 stage. This process avoids manual intervention and delayed response, so that the removal of pollution sources can be started immediately when pollution first occurs, reducing the risk of pollution spread. In the S4 stage, through statistical analysis of particle concentration, candidate areas of pollution sources are further screened out, and combined with the calculation of the particle rise rate (UPR), the location of the pollution source is determined in the S5 stage, and the automatic removal operation is quickly triggered. This not only improves the pollution management efficiency of the clean room, but also ensures the continuous cleanliness of the clean room environment and reduces the potential impact of pollution sources on production quality.

[0063] (3) The present invention realizes a multi-dimensional and three-dimensional analysis of pollution sources and improves the management efficiency of dust-free workshops through an automated response mechanism. While reducing the impact of pollution sources on the production environment, it also reduces human operational errors and management costs, ensuring the safety and compliance of the production environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 The present invention is a schematic diagram of the steps of a dust-free workshop automation control method based on data collection and analysis.

[0065] Figure 2 This is a block diagram of a dust-free workshop automation control system based on data collection and analysis of the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Example 1, please refer to Figure 1 The present invention provides a dust-free workshop automation control method based on data collection and analysis, comprising the following steps:

[0068] S1. Set up data collection nodes in the dust-free workshop and deploy intelligent sensor groups at the data collection nodes to monitor the changes in the dust-free workshop environment status in real time and collect dust-free workshop environment data;

[0069] S2. Preprocess the collected clean room environment data, including denoising, filling missing values ​​and outlier processing, and construct a clean room environment data set based on the preprocessed clean room environment data;

[0070] S3, construct a three-dimensional model of the clean room, divide the clean room into n grid areas, preset an air quality threshold AQL, and obtain an air quality score AQS for each grid area in the clean room based on the clean room environment data in the clean room environment data set, and compare and analyze the air quality score AQS of each grid area with the air quality threshold AQL to construct a first pollution source candidate grid area set SL;

[0071] S4. According to the first pollution source candidate grid area set SL, obtain the particle concentration mean μ and particle concentration variance σ of each grid area in the first pollution source candidate grid area set SL, and evaluate each grid area in the first pollution source candidate grid area set SL according to the particle concentration mean μ and particle concentration variance σ, and construct a second pollution source candidate grid area set SP;

[0072] S5. Based on the second pollution source candidate grid area set SP, obtain the particle rise rate UPR in each grid area in the second pollution source candidate grid area set SP, and locate the grid area where the pollution source is located based on the particle rise rate UPR in each grid area, trigger the automatic response mechanism, and perform pollution source removal operations.

[0073] In this embodiment, through real-time data collection and intelligent analysis technology, the accuracy of locating pollution sources in the dust-free workshop is improved, ensuring that the pollution sources can be identified and dealt with in time at the initial stage. Through multi-layer screening of three-dimensional grid models and air quality scores AQS, the scheme realizes efficient and automated pollution source response, reduces the delay of human intervention, and improves the timeliness of pollution source removal. At the same time, the scheme's intelligent data preprocessing and particulate matter concentration analysis ensure the accuracy and completeness of the data, providing a solid foundation for locating pollution sources. Overall, the scheme optimizes the air quality management of the dust-free workshop, which not only improves management efficiency, but also enhances the safety and production quality control of the dust-free workshop, ensuring the continuous cleanliness and stable operation of the dust-free workshop environment.

[0074] Example 2, please refer to Figure 1 , specifically:

[0075] S1 specific steps include:

[0076] S11. Setting a data collection node in the dust-free workshop, and deploying an intelligent sensor group at the data collection node, setting the data collection frequency of the intelligent sensor group to once per second, monitoring the environmental changes of the dust-free workshop in real time through the intelligent sensor group, recording the environmental data of the dust-free workshop, and transmitting the environmental data of the dust-free workshop to the central control system of the dust-free workshop through the wireless network equipment carried by the intelligent sensor;

[0077] The clean room environment data includes particle concentration, air flow rate and humidity data;

[0078] The particle concentration is obtained by using an ionization particle sensor, which charges the particles in the air through an electric field, and then uses electrodes to collect the mass of the charged particles to measure the particle concentration;

[0079] The air flow rate data is obtained by using an air flow rate sensor;

[0080] The humidity data is obtained by using a humidity sensor.

[0081] In this embodiment, by setting up data collection nodes and deploying intelligent sensor groups in the dust-free workshop, real-time collection of environmental data can be achieved. The sensor group monitors the particle concentration, air flow rate and humidity, and provides multi-dimensional data support for the subsequent positioning of pollution sources. Setting the data collection frequency ensures continuous monitoring of the environment, timely reflects changes in the dust-free workshop environment, and quickly discovers potential pollution problems. The wireless network equipment transmits data to the central control system, improves the efficiency and flexibility of data transmission, and reduces wiring costs. This not only ensures efficient data collection and transmission, but also provides a reliable data basis for intelligent analysis, supporting accurate pollution source identification and control.

[0082] Example 3, please refer to Figure 1 , specifically:

[0083] The specific steps of S2 include:

[0084] S21, preprocessing the clean room environment data collected in S1, including denoising, filling missing values ​​and outlier processing;

[0085] The denoising is to apply Kalman filtering to denoise the collected clean room environment data to ensure the continuity and accuracy of the data;

[0086] The missing values ​​are filled by using a mean interpolation method to fill the missing data in the collected clean room environment data with the mean of the known data at the time point where the missing data is located;

[0087] The outlier processing is based on the collected cleanroom environment data, sorting the cleanroom environment data, recording the 25% quantile Q1 and the 75% quantile Q3 of the cleanroom environment data, and obtaining the interquartile range IQR based on the 25% quantile Q1 and the 75% quantile Q3 of the cleanroom environment data, and defining the range of outliers according to the interquartile range IQR as follows: , for those not in the range The data within are considered as outliers and removed;

[0088] S22: construct a clean room environment data set based on the pre-processed clean room environment data.

[0089] In this embodiment, denoising is performed through Kalman filtering to eliminate random noise in the data collected by the sensor, ensuring that the data reflects the actual environmental state rather than errors caused by external interference or measurement errors. The mean interpolation method fills the missing values, and can make reasonable predictions based on the surrounding data of the missing data points to avoid the impact of missing data on subsequent analysis and decision-making processes. Outlier processing uses the interquartile range (IQR) method to automatically identify and remove data that does not conform to the actual situation, ensuring the overall reliability of the data and avoiding misleading due to abnormal data. This series of preprocessing methods has laid a solid foundation for subsequent data analysis, improved the accuracy of pollution source identification and air quality assessment, and effectively supported the positioning and control of pollution sources in dust-free workshops.

[0090] Example 4, please refer to Figure 1 , specifically:

[0091] The specific steps of S3 include:

[0092] S31, collecting three-dimensional data of the clean room, and using the three-dimensional modeling software Unity to construct a three-dimensional model of the clean room according to the collected three-dimensional data of the clean room, and dividing the three-dimensional model of the clean room into n grid areas to obtain a three-dimensional grid model of the clean room;

[0093] S32: for each grid area in the clean room, obtain the particle standard deviation coefficient SD and the average humidity change rate RH according to the clean room environment data in the pre-processed clean room environment data set. avg and the gas stagnation factor LVI;

[0094] The particle standard deviation coefficient SD is obtained as follows:

[0095] In the formula, C k is the particle concentration value collected by the kth data collection node, C avg is the mean of the particle concentration values ​​collected by all data collection nodes in each grid area, m is the total number of data collection nodes in each grid area, k=1, 2, 3, ..., m;

[0096] The average humidity change rate RH avg The method of obtaining is:

[0097] Where u is the total number of times each data collection node collects data, ΔRH k(t) is the humidity change rate of the kth data collection node at time t, t = 1, 2, 3, ..., u. The humidity change rate reflects the dynamics of humidity changes in the air. A higher humidity change rate means that there are fluctuations in moisture in some links in the workshop, which will affect the suspension and sedimentation of particles.

[0098] The gas stagnation coefficient LVI is obtained in the following way:

[0099] Where, LVI j is the gas stagnation coefficient of the jth grid area, v j，max is the maximum air velocity in the jth grid area, v avg，j is the average air velocity in the jth grid area. The gas stagnation coefficient LVI is used to evaluate the degree of air flow stagnation in each grid unit in the clean room. Air flow stagnation will cause the particles in the air to not be taken away smoothly, but will be deposited and accumulated in the area. Therefore, the stagnant area may become the source of particle pollution and the hot spot of pollution accumulation;

[0100] S33, based on the obtained particle standard deviation coefficient SD and average humidity change rate RH of each grid area avg and the gas stagnation coefficient LVI, to obtain the air quality score AQS of each grid area, wherein the air quality score AQS is obtained as follows:

[0101] Where α1, α2 and α3 are the particle standard deviation coefficient SD, the average humidity change rate RH, respectively. avg And the weight coefficient of the gas stagnation coefficient LVI, α1+α2+α3=1.

[0102] S34, preset air quality threshold AQL, compare and analyze the air quality score AQS obtained in S33 with the preset air quality threshold AQL to evaluate the pollution status of each grid area in the clean room. The specific evaluation steps are as follows:

[0103] S341. If the air quality score AQS is less than the preset air quality threshold AQL, that is, AQS<AQL, it is determined that the air quality in the grid area is within a safe range, that is, the clean room is not polluted, and the clean room environmental data is recorded and the clean room is continuously monitored;

[0104] S342. If the air quality score AQS is greater than or equal to the preset air quality threshold AQL, that is, AQS ≥ AQL, it is determined that the air quality in the grid area is within the dangerous range, that is, the clean room is polluted, and the alarm mechanism is triggered, and the alarm is continuously issued to notify the clean room operators and managers until the clean room operators and managers respond;

[0105] S35. Construct a first pollution source candidate grid area set SL based on the grid areas corresponding to the air quality scores AQS obtained in S342 being greater than or equal to the preset air quality threshold AQL, where SL=(SL1, SL2, SL3, ..., SL g ), where g refers to the total number of grid areas in the first pollution source candidate grid area set SL.

[0106] In this embodiment, by dividing the dust-free workshop into multiple grid areas and obtaining the air quality score AQS of each grid area, accurate monitoring of the dust-free workshop environment can be achieved. The air quality score AQS combines the standard deviation coefficient SD of particulate matter and the average humidity change rate RH. avg and the gas stagnation coefficient LVI, so that the pollution situation of each grid area can be clearly and carefully evaluated. This grid management method can identify the pollution risk area in a targeted manner, thereby improving the efficiency and accuracy of locating the pollution source. In addition, by comparing the air quality score AQS with the preset air quality threshold AQL, it can automatically determine which grid areas in the dust-free workshop have air quality that exceeds the safe range. When the AQS value is greater than or equal to the AQL, the system will automatically trigger the alarm mechanism to notify the operators and managers of the dust-free workshop to take timely measures to avoid further expansion of the pollution incident. This automated early warning mechanism can greatly reduce human negligence and delays and improve the response speed of dust-free workshop pollution response. In addition, by constructing the first pollution source candidate grid area set SL, further analyzing and screening the pollution source candidate areas, the scheme can effectively determine the potential location of the pollution source. This process accurately identifies the area where the pollution source may occur through a comprehensive analysis of the particle concentration, humidity change rate and gas stagnation coefficient, and triggers further pollution source removal operations based on these data. This targeted screening and operation mechanism can intervene in time when the pollution source first appears to prevent the spread of pollution and affect product quality.

[0107] Example 5, please refer to Figure 1 , specifically:

[0108] S4 specific steps include:

[0109] S41. For each grid area within the first pollution source candidate grid area set SL, according to the clean room environment data in the clean room environment data set constructed in step S22, set the time sliding window to x seconds and the step length to 1s, and within the current time sliding window, based on the clean room environment data, obtain the particle concentration mean μ and particle concentration variance σ within the current time sliding window.

[0110] S42: Perform a secondary assessment of the pollution status of the clean room based on the obtained particle concentration mean μ and particle concentration variance σ, and analyze the pollution status of each grid area in the first pollution source candidate grid area set SL. The specific assessment steps are as follows:

[0111] S421, if the mean particle concentration μ of the sliding window at the current time c Greater than or equal to the mean particle concentration μ of the previous sliding window p Add twice the particle concentration variance σ p , that is, μ c ≥μ p +2×σ p When , the current time sliding window is judged to be in an abnormal state, and the grid area where the time sliding window is in an abnormal state is marked as an abnormal area;

[0112] S422, if the mean particle concentration μ of the sliding window at the current time c Less than the mean particle concentration μ of the sliding window at the previous time p Add twice the particle concentration variance σ p , that is, μ c <μ p +2×σ p When the current time sliding window is judged to be in a normal state, the data is continuously monitored and recorded;

[0113] S43: Based on the grid areas marked as abnormal areas in S421, a second set of candidate grid areas for pollution sources SP is constructed, where SP = (SP1, SP2, SP3, ..., SP h ), where h refers to the total number of grid areas in the second pollution source candidate grid area set SP.

[0114] In this embodiment, by combining the particle concentration mean μ and the particle concentration variance σ, the pollution state of each grid area in the first pollution source candidate grid area set SL is secondary evaluated within the time sliding window, which can accurately capture the changing trend of pollutant concentration and help identify possible pollution source areas. In this process, using the particle concentration mean μ and the particle concentration variance σ as evaluation criteria, it is possible to effectively screen out areas with more significant pollution changes, thereby excluding normal areas and focusing on grid areas with pollution risks. This method effectively improves the positioning accuracy of pollution sources and avoids misjudgment due to data noise or other interference factors. In addition, by marking and further screening abnormal areas, a second pollution source candidate grid area set SP is constructed. This set includes all areas with obvious increases in particle concentration and is the focus of subsequent pollution source positioning and treatment. By constructing the second pollution source candidate grid area set SP, accurate identification of each pollution source area is ensured, thereby providing more reliable data support and decision-making basis for the next step of pollution source removal operations.

[0115] Example 6, please refer to Figure 1 , specifically:

[0116] S5 specific steps include:

[0117] S51, based on each grid area in the second pollution source candidate grid area set SP constructed in S43, obtain the particle rise rate UPR in each grid area, and the particle rise rate UPR is obtained in the following manner:

[0118] Where C(t) is the particle concentration value in the grid area at time t, △t is the data collection time interval, and C(t-△t) is the particle concentration value at the previous time point that is △t apart from the current time t.

[0119] S52. Based on the obtained particle rise rate UPR in each grid area, select the grid area with the largest particle rise rate UPR as the grid area where the pollution source is located.

[0120] S53, triggering an automatic response mechanism according to the grid area where the pollution source is located obtained in S52, the specific content of the automatic response mechanism is as follows:

[0121] For the grid area where the pollution source is located, the local ventilation system is automatically turned on, the ventilation system power is increased, powerful fans and dust collectors are used to increase the air flow in the polluted area, and the automatic vacuum robot is started to clean the grid area where the pollution source is located. At the same time, the pollution source information is sent to the dust-free workshop technicians to eliminate the pollution source;

[0122] For grid areas outside the grid area where the pollution source is located, enhance the power of the overall ventilation system of the clean room, increase the ventilation frequency of the clean room, start the air purification equipment to remove pollution, and adjust the wind direction and wind speed of the clean room through the wind direction control system of the clean room to block the grid area of ​​the clean room where pollution occurs.

[0123] In this embodiment, by calculating and evaluating the particle rise rate UPR, the diffusion path of pollutants in the dust-free workshop can be accurately tracked. The particle rise rate UPR reflects the rate of change of particle concentration over time, so it can respond at the early stage of the pollution source, especially in areas where particulate matter accumulation is more significant. Compared with traditional manual inspection or single concentration data analysis, UPR provides a more dynamic and scientific basis for the identification of pollution sources, avoiding the failure of pollution source identification due to errors and lags. The automated response system based on the particle rise rate UPR can sense the changes in pollution sources in real time and quickly execute cleaning measures, especially for the grid area where the pollution source is located. The local ventilation system is automatically turned on, and powerful fans and dust collectors are used to increase the air flow rate to quickly reduce the pollutant concentration and prevent further spread of pollution. At the same time, the automatic vacuum robot cleans the pollution source to ensure the effective removal of particulate matter. This mechanism greatly improves the removal efficiency and reduces the time delay of human intervention to ensure that the cleanliness of the dust-free workshop is restored in time.

[0124] Example 7, please refer to Figure 2 , specifically:

[0125] A dust-free workshop automation control system based on data collection and analysis, including a data collection module, a preprocessing module, a primary evaluation module, a secondary evaluation module and a pollution source positioning and automatic response module;

[0126] The data acquisition module is used to set up data acquisition nodes in the dust-free workshop, and deploy intelligent sensor groups at the data acquisition nodes to monitor the changes in the environmental state of the dust-free workshop in real time and collect the environmental data of the dust-free workshop;

[0127] The preprocessing module is used to preprocess the collected clean room environment data, including denoising, filling missing values ​​and outlier processing, and construct a clean room environment data set based on the preprocessed clean room environment data;

[0128] The initial assessment module is used to construct a three-dimensional model of a clean room, divide the clean room into n grid areas, preset an air quality threshold AQL, and obtain an air quality score AQS for each grid area in the clean room based on the clean room environment data in the clean room environment data set, and compare and analyze the air quality score AQS of each grid area with the air quality threshold AQL to construct a first pollution source candidate grid area set SL;

[0129] The secondary evaluation module is used to obtain the mean particle concentration μ and the particle concentration variance σ of each grid area in the first pollution source candidate grid area set SL according to the first pollution source candidate grid area set SL, and evaluate each grid area in the first pollution source candidate grid area set SL according to the mean particle concentration μ and the particle concentration variance σ, and construct the second pollution source candidate grid area set SP;

[0130] The pollution source positioning and automatic response module is used to obtain the particle rise rate UPR in each grid area in the second pollution source candidate grid area set SP based on the second pollution source candidate grid area set SP, and locate the grid area where the pollution source is located based on the particle rise rate UPR in each grid area, trigger the automatic response mechanism, and perform pollution source removal operations.

[0131] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A dust-free workshop automation control method based on data collection and analysis, characterized in that: The following steps are included: S1. Set up data collection nodes in the dust-free workshop and deploy intelligent sensor groups at the data collection nodes to monitor the changes in the dust-free workshop environment status in real time and collect dust-free workshop environment data; S2. Preprocess the collected clean room environment data, including denoising, filling missing values ​​and outlier processing, and construct a clean room environment data set based on the preprocessed clean room environment data; S3, construct a three-dimensional model of the clean room, divide the clean room into n grid areas, preset an air quality threshold AQL, and obtain an air quality score AQS for each grid area in the clean room based on the clean room environment data in the clean room environment data set, and compare and analyze the air quality score AQS of each grid area with the air quality threshold AQL to construct a first pollution source candidate grid area set SL; S4. According to the first pollution source candidate grid area set SL, obtain the particle concentration mean μ and particle concentration variance σ of each grid area in the first pollution source candidate grid area set SL, and evaluate each grid area in the first pollution source candidate grid area set SL according to the particle concentration mean μ and particle concentration variance σ, and construct a second pollution source candidate grid area set SP; S5. According to the second pollution source candidate grid area set SP, obtain the particle rise rate UPR in each grid area in the second pollution source candidate grid area set SP, and locate the grid area where the pollution source is located according to the particle rise rate UPR in each grid area, trigger the automatic response mechanism, and perform pollution source removal operation; The specific steps of S3 include: S31, collecting three-dimensional data of the clean room, and using the three-dimensional modeling software Unity to construct a three-dimensional model of the clean room according to the collected three-dimensional data of the clean room, and dividing the three-dimensional model of the clean room into n grid areas to obtain a three-dimensional grid model of the clean room; S32: for each grid area in the clean room, obtain the particle standard deviation coefficient SD and the average humidity change rate RH according to the clean room environment data in the pre-processed clean room environment data set. avg and the gas stagnation factor LVI; The particle standard deviation coefficient SD is obtained as follows: ; In the formula, C k is the particle concentration value collected by the kth data collection node, C avg is the mean of the particle concentration values ​​collected by all data collection nodes in each grid area, m is the total number of data collection nodes in each grid area, k=1, 2, 3, ..., m; The average humidity change rate RH avg The method of obtaining is: ; Where u is the total number of times each data collection node collects data, ΔRH k (t) is the humidity change rate of the kth data collection node at time t, t=1, 2, 3, ..., u; The gas stagnation coefficient LVI is obtained by: ; Where, LVI j is the gas stagnation coefficient of the jth grid area, v j，max is the maximum air velocity in the jth grid area, v avg，j is the average air velocity in the jth grid area; S33, based on the obtained particle standard deviation coefficient SD and average humidity change rate RH of each grid area avg and the gas stagnation coefficient LVI, to obtain the air quality score AQS of each grid area, wherein the air quality score AQS is obtained as follows: ; Where α1, α2 and α3 are the particle standard deviation coefficient SD, the average humidity change rate RH, respectively. avg And the weight coefficient of the gas stagnation coefficient LVI; S34, preset air quality threshold AQL, compare and analyze the air quality score AQS obtained in S33 with the preset air quality threshold AQL to evaluate the pollution status of each grid area in the clean room. The specific evaluation steps are as follows: S341. If the air quality score AQS is less than the preset air quality threshold AQL, that is, AQS<AQL, it is determined that the air quality in the grid area is within a safe range, that is, the clean room is not polluted, and the clean room environmental data is recorded and the clean room is continuously monitored; S342. If the air quality score AQS is greater than or equal to the preset air quality threshold AQL, that is, AQS ≥ AQL, it is determined that the air quality in the grid area is within the dangerous range, that is, the clean room is polluted, and the alarm mechanism is triggered, and the alarm is continuously issued to notify the clean room operators and managers until the clean room operators and managers respond; S35. Construct a first pollution source candidate grid area set SL based on the grid areas corresponding to the air quality score AQS obtained in S342 being greater than or equal to the air quality threshold AQL, where SL = (SL1, SL2, SL3, ..., SL g ); The specific steps of S4 include: S41. For each grid area in the first pollution source candidate grid area set SL, according to the clean room environment data in the clean room environment data set constructed in S22, set the time sliding window to x seconds and the step length to 1 second, and obtain the particle concentration mean μ and particle concentration variance σ in the current time sliding window based on the clean room environment data in the current time sliding window; S42: Perform a secondary assessment of the pollution status of the clean room based on the obtained particle concentration mean μ and particle concentration variance σ, and analyze the pollution status of each grid area in the first pollution source candidate grid area set SL. The specific assessment steps are as follows: S421, if the mean particle concentration μ of the sliding window at the current time c Greater than or equal to the mean particle concentration μ of the sliding window at the previous time p Add twice the variance of particle concentration σ p , that is, μ c ≥μ p +2×σ p When , the current time sliding window is judged to be in an abnormal state, and the grid area where the time sliding window is in an abnormal state is marked as an abnormal area; S422, if the mean particle concentration μ of the sliding window at the current time c Less than the mean particle concentration μ of the sliding window at the previous time p Add twice the variance of particle concentration σ p , that is, μ c <μ p +2×σ p When the current time sliding window is judged to be in a normal state, the data is continuously monitored and recorded; S43: Based on the grid areas marked as abnormal areas in S421, a second set of candidate grid areas for pollution sources SP is constructed, where SP = (SP1, SP2, SP3, ..., SP h ); S5 specific steps include: S51, based on each grid area in the second pollution source candidate grid area set SP constructed in S43, obtain the particle rise rate UPR in each grid area, and the particle rise rate UPR is obtained in the following manner: ; Where C(t) is the particle concentration value in the grid area at time t, △t is the data collection time interval, and C(t-△t) is the particle concentration value at the previous time point that is △t apart from the current time t. S52, according to the obtained particle rising rate UPR in each grid area, select the grid area with the largest particle rising rate UPR as the grid area where the pollution source is located; S53, triggering an automatic response mechanism according to the grid area where the pollution source is located obtained in S52, the specific content of the automatic response mechanism is as follows: For the grid area where the pollution source is located, the local ventilation system is automatically turned on, the ventilation system power is increased, powerful fans and dust collectors are used to increase the air flow in the polluted area, and the automatic vacuum robot is started to clean the grid area where the pollution source is located. At the same time, the pollution source information is sent to the dust-free workshop technicians to eliminate the pollution source; For grid areas outside the grid area where the pollution source is located, enhance the power of the overall ventilation system of the clean room, increase the ventilation frequency of the clean room, start the air purification equipment to remove pollution, and adjust the wind direction and wind speed of the clean room through the wind direction control system of the clean room to block the grid area of ​​the clean room where pollution occurs.

2. The method for automatic control of a clean room based on data collection and analysis according to claim 1, characterized in that: S1 specific steps include: S11. Setting a data collection node in the dust-free workshop, and deploying an intelligent sensor group at the data collection node, setting the data collection frequency of the intelligent sensor group to once per second, monitoring the environmental changes of the dust-free workshop in real time through the intelligent sensor group, recording the environmental data of the dust-free workshop, and transmitting the environmental data of the dust-free workshop to the central control system of the dust-free workshop through the wireless network equipment carried by the intelligent sensor; The clean room environment data includes particle concentration, air flow rate and humidity data.

3. The method for automatic control of a clean room based on data collection and analysis according to claim 2 is characterized in that: The specific steps of S2 include: S21, preprocessing the clean room environment data collected in S1, including denoising, filling missing values ​​and outlier processing; The denoising is to apply Kalman filtering to denoise the collected clean room environment data; The missing values ​​are filled by using a mean interpolation method to fill the missing data in the collected clean room environment data with the mean of the known data at the time point where the missing data is located; The outlier processing is based on the collected cleanroom environment data, sorting the cleanroom environment data, recording the 25% quantile Q1 and the 75% quantile Q3 of the cleanroom environment data, and obtaining the interquartile range IQR based on the 25% quantile Q1 and the 75% quantile Q3 of the cleanroom environment data, and defining the range of outliers according to the interquartile range IQR as follows: , for those not in the range The data within are considered as outliers and removed; S22: construct a clean room environment data set based on the pre-processed clean room environment data.

4. A clean room automation control system based on data collection and analysis, used to implement a clean room automation control method based on data collection and analysis as described in any one of claims 1 to 3, characterized in that: It includes data acquisition module, pre-processing module, primary assessment module, secondary assessment module and pollution source location and automatic response module; The data acquisition module is used to set up data acquisition nodes in the dust-free workshop, and deploy intelligent sensor groups at the data acquisition nodes to monitor the changes in the environmental state of the dust-free workshop in real time and collect the environmental data of the dust-free workshop; The preprocessing module is used to preprocess the collected clean room environment data, including denoising, filling missing values ​​and outlier processing, and construct a clean room environment data set based on the preprocessed clean room environment data; The initial assessment module is used to construct a three-dimensional model of a clean room, divide the clean room into n grid areas, preset an air quality threshold AQL, and obtain an air quality score AQS of each grid area in the clean room according to the clean room environment data in the clean room environment data set, and perform an initial assessment of the air quality score AQS and the air quality threshold AQL of each grid area to construct a first pollution source candidate grid area set SL; The secondary evaluation module is used to obtain the particle concentration mean μ and particle concentration variance σ of each grid area in the first pollution source candidate grid area set SL according to the first pollution source candidate grid area set SL, and evaluate each grid area in the first pollution source candidate grid area set SL according to the particle concentration mean μ and particle concentration variance σ, and construct the second pollution source candidate grid area set SP; The pollution source positioning and automatic response module is used to obtain the particle rise rate UPR in each grid area in the second pollution source candidate grid area set SP based on the second pollution source candidate grid area set SP, and locate the grid area where the pollution source is located based on the particle rise rate UPR in each grid area, trigger the automatic response mechanism, and perform pollution source removal operations.

Images