A warning correction system for clean room particle monitoring

Abstract

The present application relates to a kind of early warning correction systems of clean space particle monitoring, including particle monitoring hardware device, environmental parameter monitoring hardware device, modeling module, data acquisition monitoring module, simulation module and digital twin module.Simulation module is based on particle information and environmental information and is combined with fluid model to construct the correlation model about particle information and environmental information, so as to obtain CAE simulation result, and digital twin module is constructed based on the result.Digital twin module carries out multilevel prediction according to the particle information and environmental information of real-time acquisition to the future particle information variation trend of clean space, and the prediction result is fed back to environmental control module to make it continue to run current scheme or run optimization scheme to adjust the particle state of clean space, realize particle monitoring trend prediction, clean space control intelligentization.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of environmental monitoring technology and environmental quality control technology, and particularly relates to a warning correction system for clean space particle monitoring. BACKGROUND

[0002] Clean environment monitoring technology and environmental quality control technology are deeply integrated with purification engineering technology, and are widely applied to high-tech industries such as electronics, nuclear energy, aerospace, biological engineering, pharmaceuticals, precision machinery, chemical industry, food, automobile manufacturing and modern science. Taking semiconductor equipment as an example, clean engineering technology plays an important role in microelectronic technology. The core of microelectronic technology is integrated circuit, and the production of integrated circuit puts forward strict requirements on its process environment and clean technology and equipment in the process. In the prior art, particle monitoring devices such as particle counters are often used to monitor particles in the environment gas. Generally, when the monitoring device detects that the monitored space object exceeds the monitoring standard, it will issue an alarm to prompt the abnormality of the monitored space object and remind the operator to record or repair the abnormality of the monitored space object.

[0003] However, generally, when the monitored space object is abnormal, it is usually accompanied by losses such as product yield reduction. The essential purpose of the monitored object monitored by the monitoring device is to hope that the monitored object is always in a normal state. Although the abnormal alarm of the monitoring device can reduce the loss, the technical personnel hope to develop a monitoring device that can give a warning before the abnormality occurs, so as to take correction measures before the abnormality occurs, avoid the occurrence of loss, and realize the trend prediction of particle monitoring and intelligent clean control. SUMMARY

[0004] In view of all or part of the deficiencies of the prior art described above, the purpose of the present application is to provide a warning correction system for clean space particle monitoring, which can realize particle monitoring trend prediction by multi-level prediction of the change trend of future particle information of the clean space. Another purpose of the present application is to feed back the prediction result to the environmental control module and make it continue to run the current scheme or run the optimized environmental control scheme to adjust the future particle state of the clean space, so as to realize intelligent control of the clean space.

[0005] In order to achieve the above-mentioned purposes, the present application provides the following technical solutions: a clean space particle monitoring early warning correction system, comprising: a particle monitoring hardware device, used for monitoring and outputting particle information in a clean space; an environmental parameter monitoring hardware device, used for monitoring and outputting environmental information affecting the cleanliness of the clean space; a data acquisition monitoring module, in signal connection with the particle monitoring hardware device and the environmental parameter monitoring hardware device, acquiring the particle information and the environmental information, and monitoring the acquired particle information in real time; and a digital twin module, acquiring the particle information output by the particle monitoring hardware device and the environmental information output by the environmental parameter monitoring hardware device from the data acquisition monitoring module, and performing multi-level prediction on the change trend of future particle information in the clean space based on a digital twin analysis model in the digital twin module according to the received particle information and environmental information. The technical solution has the beneficial effects that: by setting the particle monitoring hardware device and the environmental parameter monitoring hardware device in the monitored clean space to acquire particle information and environmental information in the monitored environmental space, real particle information and environmental information in the clean space under a real physical environment are acquired, providing real data support for subsequent construction of a correlation analysis model of environmental information-particle information; by setting the data acquisition monitoring module to receive particle information and environmental information in real time and monitor the received particle information in real time, the data twin module performs multi-level prediction on the change trend and result of future particle information in the monitored clean space according to real-time particle information and environmental information and based on a data twin analysis model, realizing prediction of the change trend of particle information in the clean space.

[0006] The data acquisition monitoring module comprises a real-time monitoring unit, a data acquisition unit, and an alarm unit; the data acquisition unit is in signal connection with the particle monitoring hardware device and the environmental parameter monitoring hardware device to acquire the particle information and the environmental information; the real-time monitoring unit is pre-set with a monitoring standard, which receives particle information output from the data acquisition unit in real time and outputs a monitoring result according to the monitoring standard, the monitoring result can trigger the alarm unit to alarm. The technical solution has the beneficial effects that: the data acquisition unit acquires particle information and environmental information in real time, and the above data is transmitted to the real-time monitoring unit in real time, the real-time monitoring unit compares the received particle information with the pre-set monitoring standard, and if there is an unqualified condition at present, the monitoring result is transmitted to the alarm unit to alarm and prompt the staff to repair the clean space in time, which can realize real-time acquisition and monitoring of whether the particle information of the clean space is qualified before the digital twin module starts to run, and alarm if it is unqualified; on the other hand, continuously acquiring particle information and environmental information also provides real data support for subsequent construction of a digital twin model.

[0007] The data acquisition monitoring module further comprises a data storage unit for data accumulation, which receives and stores particle information and environmental information output from the data acquisition unit and the monitoring results output from the real-time monitoring unit. The particle information, environmental information and monitoring results are stored and accumulated by the data storage unit.

[0008] The modeling module and the simulation module are further included; the modeling module constructs a BIM model based on the particle monitoring hardware device, the position parameters of the environmental parameter monitoring hardware device in the clean space and the geometric parameters of the clean space, and constructs a fluid model about the clean space based on the particle information, the environmental information and the BIM model; the simulation module obtains particle information and environmental information from the data acquisition monitoring module, and constructs a correlation analysis model about the environmental information as a boundary condition and the particle information as an analysis target based on the fluid model through CAE simulation technology. The technical scheme has the beneficial effects that the modeling module constructs a BIM model according to the geometric parameters of the clean space (referring to a physical space required to meet certain cleanliness requirements), such as the position information of the environmental adjustment device and the position information of the production device in the clean space, and the position parameters of the particle monitoring hardware device and the environmental parameter monitoring hardware device in the clean space, and constructs a fluid model about the clean space based on the particle information (including the number of particles, the particle size, the particle concentration and the distribution thereof in the physical space) and the environmental information (including environmental air pressure, air flow velocity and direction, environmental temperature and environmental humidity and other parameters capable of affecting the particle state of gas and particles in gas in the clean space) and the BIM model; the simulation module receives particle information and environmental information from the data acquisition monitoring module, and performs strong correlation analysis research based on the fluid model through CAE technology with the environmental information as a boundary condition and the particle information as an analysis target, constructs a correlation model about the particle information and the environmental information in the currently monitored clean space, and measures the close degree of correlation between the environmental information and the particle information.

[0009] The correlation analysis model outputs CAE simulation particle information as a CAE simulation result through CAE finite element virtual simulation according to the received environmental information, and constructs a digital twin analysis model based on the CAE simulation result. In the correlation model, the CAE simulation particle information corresponding to the received environmental information is output as a CAE simulation result through CAE finite element virtual simulation, and a digital twin analysis model is constructed based on the CAE simulation result, which provides model support for subsequent prediction of the change trend of particle information in the clean space in the future.

[0010] The simulation module receives in real time particle information from the particle monitoring hardware device and environmental information from the environmental parameter monitoring hardware device output from the data acquisition monitoring module, and obtains CAE simulation particle information corresponding to the environmental information; if the CAE simulation particle information does not match the real-time received particle information, the correlation analysis model is iterated until the CAE simulation particle information matches the real-time received particle information. The beneficial effect of this technical solution is that the simulation module receives in real time particle information and environmental information output from the data acquisition unit, and outputs CAE simulation particle information (virtual particle information) corresponding to the real-time received environmental information through the correlation model, and compares the real-time received particle information with the output CAE simulation particle information. If the output CAE simulation particle information and the real-time received particle information are inconsistent, the correlation analysis model is iterated until the CAE simulation particle information matches the real-time received particle information, thereby realizing iterative optimization of the correlation analysis model and further improving the accuracy of subsequent simulation results.

[0011] The digital twin module obtains in real time particle information output by the particle monitoring hardware device and environmental information output by the environmental parameter monitoring hardware device from the data acquisition monitoring module, and predicts the future particle information change trend of the clean room in combination with the previously received particle information and environmental information. The digital twin module predicts the future particle information change trend of the monitored clean room according to the previously received particle information and environmental information in combination with the real-time received particle information and environmental information.

[0012] A set of warning conditions X is preset in the digital twin module, Y is a set of prediction result values output by the digital twin analysis model; when Y triggers X, the digital twin module issues a warning. The beneficial effect of this technical solution is that when the future particle value in the monitored clean room obtained by the digital twin analysis model through multi-level prediction does not meet the standard, i.e., Y triggers X, a warning is issued, realizing particle monitoring trend warning.

[0013] When Y triggers X, the digital twin module truncates the current collected particle information and environmental information, and constructs a first model through a classification algorithm; the correction target value of future particle information in the clean space is set, the environmental information is used as a boundary condition optimization, and a regression algorithm is used to establish a second model to generate a topology optimization scheme for adjusting the environmental information to make the particle information reach the correction target value. The beneficial effects of this technical solution are that when Y triggers X, i.e., the prediction result is not up to standard, the digital twin analysis model receives an alarm and performs data truncation processing on the collected particle information and environmental information of the clean space (local or whole), and a first model such as a random forest or a decision tree is constructed using the truncated data through a classification algorithm. On the basis of the first model, a correction target value (a value lower than the warning value) is set, a regression algorithm is used to establish a second model to generate a topology optimization scheme for adjusting the environmental information (i.e., using the environmental information as a boundary condition optimization) to make the particle information reach the correction target value, and the correction of future particle information in the clean space is realized.

[0014] The early warning correction system further comprises an environmental control module for implementing the topology optimization scheme; when Y triggers X, the environmental control module receives the topology optimization scheme and executes it, and when Y does not trigger X, the environmental control module continues to execute the current running scheme. By setting the environmental control module in the early warning correction system, the digital twin analysis model feeds back the prediction result to the environmental control module and makes it continue to run the current scheme or run the topology optimization scheme to adjust the environmental parameters of the monitored clean space, thereby adjusting the particle state in the future clean space and realizing intelligent control of the clean space.

[0015] The early warning correction system further comprises a correction control module, which is signal-connected with the digital twin module and the environmental control module; it receives the topology optimization scheme output from the digital twin module and controls the environmental control module to execute the topology optimization scheme it receives. By adding the correction control module, the prediction and correction functions of the early warning correction system are separated, i.e., the digital twin module is used to complete the prediction of the future particle information change trend, and then the correction control module is used to complete the intelligent control of the clean space running scheme.

[0016] The multi-level prediction includes the prediction of the change of each parameter in the environmental information causing the change of each parameter in the particle information. The beneficial effects of this technical solution are that the multi-level prediction can predict the future particle information of the whole clean space, predict the future particle information of the local clean space, and predict the future particle information according to different cleanliness levels.

[0017] The early warning correction system further comprises a display module, and the information displayed by the display module includes the data obtained by the data acquisition module, the prediction result of the digital twin module and the operation scheme of the environment control module. By setting the display module, the above particle information, environment information and data after secondary processing thereof can be displayed in the display module according to the user demand, so as to be manually viewed.

[0018] The early warning correction system further comprises an input configuration unit, and the parameters input by the input configuration unit include monitoring standards. By setting the input configuration unit, the monitoring standards can be introduced into the early warning correction system, and the early warning values and other data required to be input according to the demand can also be input into the early warning correction system.

[0019] The particle information includes one or more of the particle number, particle size, particle concentration; and the environment information includes one or more of the environmental air pressure, airflow flow rate, airflow flow direction, environmental temperature and environmental humidity in the clean space. By obtaining the particle information such as the particle number, particle size and concentration, and the data of factors that can affect the particle distribution in the clean space such as the air pressure, airflow flow rate and flow direction, temperature and humidity, the fluid model, simulation module and digital twin module are constructed, and the multi-level prediction of the particle information in the clean space is realized.

[0020] Compared with the prior art, the present application has at least the following beneficial effects: the particle detection hardware device and the environmental parameter monitoring hardware device are arranged to collect particle information and environmental information in the clean space, the data acquisition and monitoring module collects, stores and monitors the above-mentioned data in real time, and the function of real-time monitoring and early warning of the particle information in the current space is realized; the above-mentioned data are stored to provide real data support for building a correlation analysis model, and the correlation analysis model is continuously self-learned and iterated to further iteratively optimize the digital twin model, thereby continuously improving the prediction accuracy; the BIM model is built based on the positions of the particle detection hardware device and the environmental parameter monitoring hardware device in the clean space and the geometric parameters of the monitored clean space, the fluid model about the monitored clean space is built based on the BIM model combined with the particle information and the environmental information, so that the prediction of the future particle change trend of the clean space by the digital twin module is more accurate; the simulation module is arranged, the correlation model about the particle information and the environmental information collected by the data acquisition and monitoring module is built based on the BIM model, and the virtual digital space is built; the digital twin analysis model is built based on the CAE simulation result, the future particle information of the clean space is predicted in multiple levels, and the particle monitoring trend prediction is realized; the digital twin analysis model can also compare the prediction result with the pre-set early warning condition, and control the environmental module to run the topology optimization scheme or continue to run the current scheme according to the comparison result, so as to realize the intelligent control of the clean space; the display module and the input configuration module are arranged, the monitoring data and the monitoring result can be displayed in real time, and the early warning condition or other information to be input can be input or changed through the input configuration module. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the specific embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings described below are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0022] Figure 1 The structure diagram of the early warning correction system in the first embodiment of the present application.

[0023] Figure 2 The example diagram of the multi-level monitoring in the first embodiment of the present application.

[0024] Figure 3 The structure diagram of the early warning correction system in the second embodiment of the present application.

[0025] Figure 4 The structure diagram of another early warning correction system in the second embodiment of the present application.

[0026] Figure numerals: 1-clean space; 2-environmental parameter monitoring hardware equipment; 3-particle monitoring hardware equipment; 4-data acquisition and monitoring module; 5-digital twin module; 6-environmental control module; 7-fluid model; 8-simulation module; 9-correction control module. DETAILED DESCRIPTION

[0027] The following is a clear and complete description of the technical solutions in the specific embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0028] Example 1

[0029] like Figure 1 As shown, in this embodiment, a clean space particle monitoring early warning and correction system is provided for implementing early warnings for an environmental monitoring system. The system comprises: environmental parameter monitoring hardware equipment 2, particle monitoring hardware equipment 3, a data acquisition and monitoring module 4, and a digital twin module 5. Specifically, the particle monitoring hardware equipment 3 is used to monitor and output particle information within the monitored clean space 1. The particle monitoring hardware equipment 3 can employ clean space monitoring devices such as particle counters as needed and can be deployed at multiple locations. The specific particle monitoring hardware equipment 3 is not specifically limited as long as it can capture particle information. Particle information includes, but is not limited to, one or more of the following: particle number, particle size, concentration, and distribution. These particles primarily refer to tiny solid particulate matter (micrometer-level, e.g., 0.1-25 microns) within the clean space 1, such as tiny dust particles. Furthermore, the clean space 1 referred to above refers to a physical space where the air cleanliness meets a certain cleanliness level.

[0030] The environmental parameter monitoring hardware device 2 is used to collect and monitor the environmental information in the clean space that can affect the particle information in the clean space 1, and the above-mentioned influencing factors include one or more of the parameters that can affect the state of the gas in the clean space and the particle state of the particulate matter in the gas, such as air pressure, air flow rate, air flow, air flow direction, temperature, humidity, etc. The specific environmental parameter monitoring hardware device 2 can be determined according to actual needs. The environmental parameter monitoring hardware device 2 can be monitored by a separate temperature sensor, humidity sensor, air flow sensor, air flow direction detector, air flow speed detector, etc. for realizing environmental information acquisition, and can be arranged at multiple points. The specific environmental parameter monitoring hardware device 2 is not specifically limited under the premise of being able to realize environmental information acquisition, and general temperature and humidity control equipment can also be used to obtain environmental temperature and humidity information. However, in the case of a relatively large physical space of the monitored clean space 1, if only temperature and humidity control or air pressure, air flow rate, etc. are used to obtain environmental information, it is not possible to better monitor the environment in some local spaces in the clean space 1. In the present embodiment, a separate temperature and humidity sensor and an air flow sensor are used as the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2, and the above-mentioned sensors are placed in different places in the clean space 1 to obtain more realistic environmental information. Of course, the number of particle monitoring hardware devices 3 and environmental parameter monitoring hardware devices 2 in the clean space 1 to be monitored can be set according to needs, which is not limited herein.

[0031] The data acquisition and monitoring module 4 is signal connected with the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2 placed in the clean space 1, acquires the particle information and the environmental information monitored by the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2, and monitors the particle information in real time.

[0032] The digital twin module 5 receives the particle information and the environmental information output by the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2 acquired by the data acquisition and monitoring module 4, and performs multi-level prediction on the future particle information change trend of the clean space 1 based on the digital twin analysis model in the digital twin module 5.

[0033] In this embodiment, the data acquisition and monitoring module 4 includes a real-time monitoring unit, a data acquisition unit, and an alarm unit. The data acquisition unit is signal-connected to the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2 to receive particle information and environmental information. The real-time monitoring unit is pre-set with monitoring standards and receives the particle information output from the data acquisition unit in real time and outputs monitoring results based on the monitoring standards. The monitoring results can trigger the alarm unit to issue an alarm. When the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2 are distributed at multiple points, the real-time monitoring unit obtains particle information and environmental information from each of the particle monitoring hardware devices 3 and the environmental parameter monitoring hardware devices 2 at each distribution point, and monitors each distribution point in real time. The information output by the data acquisition and monitoring module 4 after receiving the particle information and environmental information from the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2 is referred to as first particle information and first environmental information. The first particle information includes the particle information obtained from the particle monitoring hardware device 3, and the first environmental information includes the environmental information obtained from the environmental parameter monitoring hardware 2. Inside the data acquisition and monitoring module 4, the information output by the data acquisition unit after acquiring particle information from the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2 is called second particle information, and the information output after acquiring environmental information is called second environmental information.

[0034] Specifically, the value D is set in the real-time monitoring unit as the value set of the particle information received by the real-time monitoring unit, and A is the warning value set according to the particle capacity of the clean space 1 to be monitored to maintain cleanliness; when the value in D is greater than or equal to the corresponding warning value in A, the real-time monitoring unit transmits the monitoring result to the alarm unit, and the alarm unit issues an alarm to prompt the operator to make repairs in time. In this embodiment, if Figure 2 As shown, the cleanliness level of the monitored clean space is ISO 2, which requires that the concentration of particles with a diameter of 0.1µm in the clean space is less than 100 particles / m 3 The warning value of 0.1µm particles in warning value set A is set to 100 particles / m 3 , when the concentration of 0.1µm particles in the value set D of the particle information received by the real-time monitoring unit is greater than or equal to 100 particles / m 3 When the value of , for example, the concentration of 0.1µm particles in the value set D of the particle information received by the real-time monitoring unit is 101 particles / m 3 When the particle concentration in the clean space 1 exceeds the warning value, the real-time monitoring unit transmits the monitoring result to the alarm unit to issue an alarm, prompting the operator to perform repairs in time. Of course, in other embodiments, the warning value can be set according to requirements and is not limited here.

[0035] The data acquisition and monitoring module 4 further comprises a data storage unit for data accumulation, which receives and stores the second particle information and the second environment information collected from the data acquisition unit to form third particle information and third environment information. In this embodiment, the second particle information and the second environment information output from the data acquisition unit and stored in the data storage unit are referred to as the third particle information and the third environment information, and the third particle information and the third environment information are essentially data sets including the second particle information and the second environment information. Among them, the first particle information includes the second particle information and the third particle information, and the first environment information includes the second environment information and the third environment information. In addition, the second particle information and the second environment information can also include device identification information of the information source.

[0036] By setting the real-time monitoring unit, the data acquisition unit, the alarm unit and the data storage unit in the data acquisition and monitoring module 4, the real-time monitoring unit and the alarm unit can be used to monitor and alarm the second particle information and the second environment information in real time before the digital twin module 5 starts running, to determine whether the particle information in the clean room 1 at the moment is qualified or not. On the other hand, the data storage unit provides actual data support, i.e. the third particle information and the third environment information, for subsequent construction of a digital twin model, and the data acquisition unit can continuously output real-time second particle information and second environment information to the digital twin model after the digital twin model is constructed, for real-time multi-level prediction.

[0037] In addition to various particle monitoring hardware devices 3 distributed in the clean room 1, there are various production devices, various environment parameter monitoring hardware devices 2, and various devices in the environment control module in the monitored clean room 1. The existence of the above-mentioned devices will inevitably affect the particle distribution in the clean room. Therefore, in order to more accurately predict the future particle change trend, a modeling module is further included in this embodiment, which constructs a BIM model based on the geometric parameters of the monitored clean room 1 and the position parameters of the particle monitoring hardware devices 3, the environment parameter monitoring hardware devices 2, the production devices, etc. in the clean room 1, as well as the physical information of the clean room 1 itself, such as the area information of the clean room 1. On this basis, a fluid model 7 about the monitored clean room 1 is constructed based on the completed BIM model and combined with the particle information and environment information output by the particle monitoring hardware devices 3 and the environment parameter monitoring hardware devices 2, so that the prediction result is closer to the particle information in the real physical space, and the prediction accuracy is improved. Here, the particle information and environment information output by the particle monitoring hardware devices 3 and the environment parameter monitoring hardware devices 2 carry the information of the modeling module, such as the position parameters corresponding to the particle information and the position parameters corresponding to the environment information.

[0038] Furthermore, this embodiment also includes a simulation module 8. By configuring simulation module 8, in an initial state (i.e., before the digital twin model is operational), it receives third particle information and third environmental information. Based on the third particle information and third environmental information, a correlation analysis model for the particle information and environmental information currently being monitored in the clean space 1 is constructed using CAE technology based on the fluid model. Specifically, a correlation model is constructed using CAE simulation technology based on the fluid model, with the third environmental information as the boundary condition and the third particle information as the analysis target. Specifically, a strong correlation analysis is performed on the particle information and environmental information in the clean space 1 using the third environmental information as the input parameter and the third particle information as the output parameter, analyzing the closeness of the correlation between the particle information and the environmental information. After constructing the correlation analysis model using the third particle information and the third environmental information in the data storage unit, CAE finite element virtual simulation is used within the correlation analysis model to set environmental information. Based on the received environmental information, CAE simulated particle information is output as a CAE simulation result, and the digital twin analysis model is constructed based on the CAE simulation result. The environmental information herein can be either real or virtual environmental information.

[0039] Furthermore, simulation module 8, based on the continuously received second particle information and second environment information, derives CAE-simulated particle information corresponding to the second environment information. If the received second particle information does not match the CAE-simulated particle information, the correlation analysis model is iterated until the real-time received second particle information matches the CAE-simulated particle information. By iterating the correlation analysis model, the digital twin analysis model is iterated, continuously improving the accuracy of the prediction results.

[0040] The digital twin module 5 receives the second particle information and the second environment information output from the data acquisition and monitoring module 4 in real time, and predicts the future change trend of the particle information in the clean space 1 in combination with the second particle information and the second environment information received previously.

[0041] The digital twin module 5 comprises a digital twin analysis model, in which X is a set of preset warning conditions, and Y is a set of multi-level prediction results output by the digital twin analysis model. When Y triggers X, the digital twin module 5 issues a warning. Specifically, X is a set comprising a set of warning values X1 for the overall particle information and environmental information of the monitored clean room 1, a set of warning values X2 for the particle information and environmental information of the first local space of the monitored clean room 1, a set of warning values X3 for the particle information and environmental information of the second layout space of the monitored clean room 1, and the like. Correspondingly, Y is a set comprising a set of prediction results Y1 output by the digital analysis model for the future overall particle information of the monitored clean room 1, a set of prediction results Y2 for the future overall particle information of the first local space of the monitored clean room 1, a set of prediction results Y3 for the future overall particle information of the second layout space of the monitored clean room 1, and the like. When Y triggers X, the digital twin module 5 issues a warning.

[0042] The multi-level prediction includes prediction of changes in each parameter of the particle information caused by changes in each parameter of the environmental information. Specifically, the multi-level prediction refers to prediction of changes in the particle distribution of the overall and local particle information caused by changes in the overall and local environmental information of the clean room 1.

[0043] Specifically, in this embodiment, the digital twin analysis model is used to predict the future change trend of the overall particle information in the monitored clean room 1. The cleanliness required by the clean room 1 is ISO 1, and the concentration of particles with a diameter of 0.1 µm in the clean room 1 is required to be less than 10 / m 3 The data set of the particle information accumulated by the data storage unit is a1, and the data set of the environmental information is b1. The geometric parameter set of the clean room 1 is c1. The modeling module constructs the fluid model 7 of the monitored clean room 1 based on the data sets a1, b1, and c1, as well as the position parameters of the particle monitoring hardware device 3 and the environmental parameter monitoring hardware device 2 in the clean room 1. Thereafter, the simulation module constructs a correlation analysis model between the environmental information and the particle information in the clean room at the moment by using CAE simulation technology with b1 as the input parameter and a1 as the output parameter. Then, the CAE simulation particle information is obtained by using a large amount of environmental information as the input parameter, and the digital twin analysis model is constructed. In the digital twin analysis model, the concentration of particles with a diameter of 0.1 µm in the set of overall particle information warning values X1 of the monitored clean room 1 is set to 10 / m 3 The digital twin analysis model receives the latest second particle information a n and the second environmental information b n in real time, and processes the received a n and b nThe constant analysis and prediction is performed, and when it is found that if the current environment control scheme is continued, the predicted result Y1 of the future overall particle information of the monitored clean space 1 will be greater than or equal to X1, a warning is issued, and the future particle information change trend prediction in the clean space is realized.

[0044] Of course, in some embodiments, the future particle information change trend in a local space in the clean space 1 can also be predicted by using the completed digital twin model. Specifically, if the future particle information change trend prediction of a certain local space S1 in the clean space 1 is needed while monitoring the future particle information change trend of the overall clean space S, in addition to setting the warning value set X1 for the overall clean space S, a warning value set X2 for the space S1 is also set in the digital twin analysis model. The cleanliness requirement of the overall clean space S is IOS3 level, and the cleanliness requirement of the space S1 is IOS1 level. When the digital twin analysis model finds that if the current environment control scheme is continued, the particle information Y2 in the future space S1 will be greater than or equal to X2, a warning is issued, and the future particle information change trend prediction in the space S1 is realized, so that the multi-level prediction of the future particle information change trend in the clean space 1 is realized.

[0045] Embodiment Two

[0046] As shown in Figure 3 , this embodiment is based on embodiment one and adds an environment control module 6. In this embodiment, when the prediction result Y of the digital twin analysis model triggers X, that is, the future particle information in the monitored clean space 1 obtained by multi-level prediction reaches the warning condition, a warning is issued. The digital twin analysis model generates a topology optimization scheme, and sends the topology optimization scheme to the environment control module 6 and makes it run the newly generated environment running scheme; otherwise, the clean space continues to execute the current running scheme. The environment control module 6 is used to change the environment information in the clean space, for example, for a clean air conditioning system, it adjusts the temperature, humidity, wind speed, pressure in the clean space or inputs the degree of purification of the particle-containing gas in the clean space, etc. Here, no specific limitation is made, and in addition, the environment control module 6 and the environment parameter monitoring hardware device 2 can be an integrated device or an independent device.

[0047] Furthermore, when Y triggers X, it indicates that the prediction result does not meet the standard. The digital twin analysis model performs data truncation on the received particle information and environmental information of the clean space 1 (partial or overall), and uses the truncated data through a classification algorithm to construct a first model such as a random forest or decision tree. Based on the first model, a corrected target value (a value lower than the warning value) of the particle information in the clean space 1 in the future is set, and a regression algorithm is used to establish a second model (such as a neural network) to generate a topology optimization solution that adjusts the environmental information (i.e., optimizes with the environmental information as the boundary condition) so that the particle information reaches the corrected target value.

[0048] When the digital twin analysis model predicts that the particle value in the clean space 1 will exceed the warning value in the future if the current operation plan continues to run, the digital twin analysis model generates a topology optimization plan and sends it to the environmental control module 6, which accepts and executes the topology optimization plan.

[0049] Of course, in some embodiments, such as Figure 4 As shown, when the digital twin analysis model predicts that the particle value in the clean space 1 will exceed the warning value in the future if the current operation plan continues to run, the digital twin analysis model generates a topology optimization plan and sends it to the correction control module 9. The correction control module 9 accepts the topology optimization plan and controls the environmental control module 6 to execute the plan.

[0050] Through the mutual cooperation between the above modules, the early warning correction system disclosed in the present invention can realize data accumulation and model optimization, and gradually realize the prediction of the state change trend of particulate matter in the clean space by combining the real-time monitoring data and generate a topology optimization plan to prevent the development trend of unqualified particle information in the clean space, promote the two-way interactive mapping between physical clean space and virtual clean space, and realize the optimization of clean space early warning control.

[0051] Furthermore, in this embodiment, the early warning and correction system includes a display module that displays monitored particle information, environmental information, secondary processing data, and prediction results based on user needs. Furthermore, the early warning and correction system also includes an input configuration module that allows for input of user-required information, such as warning values ​​for different cleanliness levels and BIM models.

[0052] The above embodiments are only intended to help understand the method and core concept of the present invention. It should be noted that, without departing from the principles of the present invention, a number of improvements and modifications may be made to the present invention by those skilled in the art, and such improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A clean space particle monitoring early warning correction system, characterized by, The application relates to a particle monitoring system for a clean room. The system comprises: a particle monitoring hardware device for monitoring and outputting particle information in a clean room; an environmental parameter monitoring hardware device for monitoring and outputting environmental information affecting the cleanliness of the clean room; a data acquisition and monitoring module connected to the particle monitoring hardware device and the environmental parameter monitoring hardware device, which acquires the particle information and the environmental information and monitors the acquired particle information in real time; a digital twin module which acquires the particle information output by the particle monitoring hardware device and the environmental information output by the environmental parameter monitoring hardware device from the data acquisition and monitoring module, and predicts the change trend of future particle information in the clean room based on a digital twin analysis model in the digital twin module according to the received particle information and environmental information. The system further comprises a modeling module and a simulation module. The modeling module constructs a BIM model based on the position parameters of the particle monitoring hardware device and the environmental parameter monitoring hardware device in the clean room and the geometric parameters of the clean room, and constructs a fluid model about the clean room based on the particle information, the environmental information and the BIM model; The simulation module acquires the particle information and the environmental information from the data acquisition and monitoring module, and constructs a correlation analysis model about the environmental information as a boundary condition and the particle information as an analysis target based on the fluid model through CAE simulation; 2. The clean space particle monitoring pre-alarm correction system of claim 1, wherein, The simulation module receives the particle information and the environmental information output by the data acquisition and monitoring module in real time, and obtains CAE simulation particle information corresponding to the environmental information; if the CAE simulation particle information does not match the real-time received particle information, the correlation analysis model is iterated until the CAE simulation particle information matches the real-time received particle information. The data acquisition and monitoring module comprises a real-time monitoring unit, a data acquisition unit and an alarm unit; the data acquisition unit is connected to the particle monitoring hardware device and the environmental parameter monitoring hardware device to receive the particle information and the environmental information; 3. The clean space particle monitoring pre-alarm correction system of claim 2, wherein, The real-time monitoring unit is pre-set with a monitoring standard, which receives the particle information output by the data acquisition unit in real time and outputs a monitoring result according to the monitoring standard; the monitoring result can trigger the alarm unit to alarm.

4. The clean space particle monitoring pre-alarm correction system of claim 1, wherein, The data acquisition and monitoring module further comprises a data storage unit for data accumulation, which receives and stores the particle information and the environmental information output by the data acquisition unit and the monitoring result output by the real-time monitoring unit.

5. The clean space particle monitoring pre-alarm correction system of claim 4, wherein, The correlation analysis model is constructed through CAE finite element virtual simulation, acquires CAE simulation particle information as a CAE simulation result according to the environmental information, and constructs a digital twin analysis model based on the CAE simulation result. The digital twin module receives the particle information and the environmental information output by the data acquisition and monitoring module in real time, and predicts the change trend of future particle information in the clean room in combination with the previously received particle information and environmental information.

6. The clean space particle monitoring pre-alarm correction system of claim 5, wherein, The digital twin module is preset with a set of early warning conditions X, and Y is a set of predicted result values output by the digital twin analysis model; when Y triggers X, the digital twin module issues a warning.

7. The clean space particle monitoring pre-alarm correction system of claim 6, wherein, When Y triggers X, the digital twin module performs data truncation on the currently collected particle information and environmental information, and constructs a first model through a classification algorithm; sets a correction target value of future particle information in the clean space, and establishes a second model through a regression algorithm to generate a topology optimization scheme for adjusting the environmental information to make the particle information reach the correction target value.

8. The clean space particle monitoring pre-alarm correction system of claim 7, wherein, The early warning correction system further comprises an environmental control module for implementing the topology optimization scheme; when Y triggers X, the environmental control module receives the topology optimization scheme and executes it, and when Y does not trigger X, the environmental control module continues to execute the current operation scheme.

9. The clean space particle monitoring pre-alarm correction system of claim 8, wherein, The early warning correction system further comprises a correction control module, which is signal connected with the digital twin module and the environmental control module, receives the topology optimization scheme output from the digital twin module, and controls the environmental control module to execute the topology optimization scheme received thereby.

10. The clean space particle monitoring pre-alarm correction system of claim 1, wherein, The multi-level prediction includes prediction of changes in each parameter in the particle information caused by changes in each parameter in the environmental information.

11. The clean room particle monitoring pre-alarm correction system of claim 1, wherein, The particle information includes one or more of the number of particles, the particle size, and the particle concentration; and the environmental information includes one or more of the environmental air pressure, the airflow velocity, the airflow direction, the environmental temperature, and the environmental humidity in the clean space.

Images