Airflow organization optimization design system and method based on clean room tuyere

Abstract

The invention discloses an airflow organization optimization design system and method based on a clean room tuyere, and relates to the technical field of clean room tuyere design, the system comprises a digital twin model construction module, a real-time data acquisition module, a problem identification module, a parameter simulation optimization module and a pre-debugging scheme output module; according to the method, the digital twin model of the clean room is constructed through computer-aided design and linked with real-time operation data, conversion of airflow organization optimization from static design to dynamic response is achieved, the problem that in the prior art, an optimization scheme is disjointed with an actual operation scene is solved, and by automatically recognizing problematic tuyeres and simulating and optimizing a parameter adjustment scheme, the optimization efficiency is improved. Traditional blind debugging depending on manual experience is replaced, repeated trial and error are avoided, the debugging accuracy and efficiency are improved, stable maintenance of cleanliness indexes of the clean room is guaranteed, meanwhile, the influence of the debugging process on the production progress is reduced, and powerful support is provided for efficient and reliable operation of the clean room.

Description

Technical Field

[0001] This invention relates to the field of cleanroom air outlet design technology, specifically to a system and method for optimizing airflow organization based on cleanroom air outlets. Background Technology

[0002] Cleanrooms are specialized spaces designed to meet specific production or experimental requirements by controlling key parameters such as air pollutant concentration, temperature, humidity, and air pressure. They are indispensable functional buildings for the development of industries such as precision manufacturing, high-end scientific research, and biomedicine. Air vents, as the core components for airflow delivery and diffusion in cleanrooms, are crucial nodes connecting the air supply system to the interior space. Their structural design, installation location, and operating parameters directly determine the airflow diffusion path, coverage area, and uniformity of distribution. They are fundamental to ensuring cleanroom cleanliness standards are met and maintaining indoor environmental stability, and are of great significance for ensuring product quality and preventing contaminant interference during production.

[0003] Cleanroom airflow organization optimization design refers to the scientific configuration and adjustment of parameters such as air outlet type selection, layout planning, air volume distribution, and air supply angle, based on the cleanroom's usage scenario and process equipment layout requirements. This ensures that the indoor airflow forms a reasonable flow pattern, effectively removing particulate matter, microorganisms, and other pollutants from the air, avoiding problems such as airflow eddies and dead zones, and thus stably maintaining key parameters such as indoor cleanliness, temperature, and humidity. This design is not only a core element in ensuring that the cleanroom meets usage requirements but also an important means of reducing cleanroom operating energy consumption and improving operational efficiency. It plays an irreplaceable role in extending equipment lifespan and ensuring the continuity and stability of high-precision production processes.

[0004] However, existing cleanroom vent airflow organization optimization technologies still have certain shortcomings. The optimization process mostly remains at the static simulation level of the design stage, and can only perform airflow simulation analysis based on preset fixed operating conditions. It cannot respond to the dynamic changes in operating conditions such as the distribution of pollutants, equipment heat dissipation, and personnel flow in the cleanroom during actual operation. This leads to a disconnect between the optimization scheme in the design stage and the actual operating requirements, making it difficult to ensure the stability of cleanliness during actual operation. When problems such as local cleanliness failure or abnormal airflow occur in actual operation, there is a lack of effective dynamic monitoring and precise positioning methods. The debugging work requires repeated trial and error based on the experience of technical personnel, which is not only inefficient, but may also lead to the expansion of problems due to blind adjustments, affecting production progress and product quality. Therefore, it is of great significance to develop an airflow organization optimization design system and method based on cleanroom vents. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a system and method for optimizing airflow organization based on cleanroom vents. This system can construct a digital twin model of the cleanroom through computer-aided design and link it with real-time operational data, realizing the transformation of airflow organization optimization from static design to dynamic response. By automatically identifying problematic vents and simulating and optimizing parameter adjustment schemes, it replaces the traditional blind debugging that relies on manual experience, avoids repeated trial and error, improves the accuracy and efficiency of debugging, ensures the stable maintenance of cleanroom cleanliness indicators, and reduces the impact of the debugging process on production progress, providing strong support for the efficient and reliable operation of cleanrooms.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an airflow organization optimization design system based on cleanroom air outlets, the system comprising: a digital twin model construction module, a real-time data acquisition module, a problem identification module, a parameter simulation optimization module, and a pre-debugging scheme output module, wherein the modules communicate with each other through a preset data interaction protocol;

[0007] The digital twin model building module is based on computer-aided design technology to build a digital twin model that is completely mapped to the actual cleanroom space structure, equipment layout and air outlet configuration, and transmits the model data to the problem identification module and parameter simulation optimization module;

[0008] The real-time data acquisition module establishes a communication connection with the cleanroom sensors through Internet of Things (IoT) technology, collects real-time operating data on indoor particulate matter concentration, temperature, and humidity, and transmits the data to the problem identification module.

[0009] The problem identification module receives real-time operating data and digital twin model data, compares them with preset standard thresholds to identify airflow-related problems, and transmits the problems and corresponding air outlet location information to the parameter simulation optimization module.

[0010] The parameter simulation optimization module targets the problematic air outlet, simulates and adjusts the adjustable parameters of the air outlet in a digital twin model, and analyzes the airflow changes and cleanliness-related data through multiple sets of parameter combinations, and transmits the simulation results to the pre-debugging scheme output module.

[0011] The pre-debugging scheme output module selects parameter adjustment schemes based on simulation results, generates pre-debugging instructions, and pushes them to the on-site debugging terminal.

[0012] Furthermore, the digital twin model building module includes the following steps when building the model:

[0013] Collect the three-dimensional spatial dimensions of the cleanroom, the installation location of process equipment, the structural parameters of process equipment, the installation coordinates of air outlets, the structural form of air outlets, and the adjustable parameters of air outlets;

[0014] Computer-aided design tools are used to build a 3D geometric model based on the collected data.

[0015] The process equipment parameters, air outlet parameters and three-dimensional geometric model are linked and bound to form an initial model with parameter attributes;

[0016] Spatial collision detection is performed on the initial model to verify the matching accuracy between the model and the actual cleanroom scene, thus forming the final digital twin model.

[0017] Furthermore, the sensors in the real-time data acquisition module are evenly distributed according to the functional zones of the cleanroom, covering key process areas and areas prone to airflow abnormalities. The frequency of sensor data acquisition is set according to the cleanroom's operating conditions. The data transmission process uses an encryption protocol. Before transmission, the acquired data is standardized in format to unify the data's units, precision, and storage format, ensuring that the data can be directly accessed by the problem identification module.

[0018] Furthermore, the problem identification module includes the following steps when identifying airflow-related problems:

[0019] It receives standardized operational data transmitted from the real-time data acquisition module, performs noise reduction on the data using the moving average method, and removes data that exceeds the reasonable range through preset outlier judgment rules.

[0020] The processed data is continuously compared with a preset standard threshold in real time.

[0021] By combining the airflow coverage, operating parameters, and spatial relationships of the vents in the digital twin model, the corresponding vents that cause airflow problems are located, and a list of problematic vents is formed.

[0022] Furthermore, the parameter simulation optimization module includes the following steps when performing parameter simulation optimization:

[0023] Extract the current operating parameters, structural parameters, and corresponding local data of the digital twin model of the problematic areas from the list of problematic areas;

[0024] Based on the air outlet structure parameters and relevant parameters of the air supply system, the adjustment range of the adjustable parameters is set, and the adjustment step size is divided according to the principle of equality.

[0025] Simulations are started sequentially in the digital twin model according to the parameter combination order, and the simulation environment parameters are kept consistent for each group of simulations;

[0026] Simultaneously record the airflow trajectory coordinate data and cleanliness distribution matrix data corresponding to each set of parameter combinations to form multiple sets of simulation datasets.

[0027] Furthermore, the pre-debugging scheme output module includes the following steps when generating pre-debugging instructions:

[0028] Receive the simulation dataset transmitted by the parameter simulation optimization module, and extract the airflow diffusion data and cleanliness-related data of each scheme.

[0029] Multiple schemes are sorted according to preset filtering rules, target parameter combinations are selected, and the identification information of the problem hotspots, the specific adjustment values ​​of the target parameters, and the step-by-step adjustment operation steps are integrated.

[0030] The integrated information was converted into text and graphical formats, respectively. The graphical format included vent location markings and airflow distribution diagrams before and after parameter adjustments, and then pushed to the on-site commissioning terminal.

[0031] Furthermore, the digital twin model building module includes the following steps when updating the model:

[0032] The system interface or sensors are used to monitor whether the cleanroom has undergone space modifications, equipment additions or removals, or air vent position adjustments. When such changes occur, the three-dimensional spatial data, equipment parameters, or air vent parameters of the changed area are re-collected.

[0033] Modify the geometric structure and associated parameters of the corresponding regions based on the original digital twin model;

[0034] After the modifications are completed, collision detection and verification are performed, and the updated complete model data is synchronized to the problem identification module and the parameter simulation optimization module.

[0035] Furthermore, the parameter simulation optimization module has built-in parameter adjustment constraint rules. These constraint rules are set based on the air outlet structure design parameters, air supply system operating parameters, and cleanroom process requirements. During the simulation, only parameter combinations that conform to the constraint rules are generated. The constraint rules can be manually updated or automatically adapted according to changes in the cleanroom usage scenario.

[0036] An airflow organization optimization design method based on cleanroom air outlets, applicable to the aforementioned airflow organization optimization design system based on cleanroom air outlets, includes the following steps:

[0037] S1. Using computer-aided design technology, input relevant data on the three-dimensional spatial structure, equipment layout, and air outlet configuration of the cleanroom to construct a digital twin model that perfectly matches the actual cleanroom.

[0038] S2. Sensors connected to the Internet of Things continuously collect real-time operating data on particulate matter concentration, temperature and humidity in various areas of the cleanroom and upload the data to the system.

[0039] S3. Real-time operating data is synchronously mapped to a digital twin model and compared with preset standard thresholds to automatically identify airflow-related problems and locate the corresponding air outlets;

[0040] S4. In the digital twin model, simulate and adjust the adjustable parameters for the problematic air outlet. Through simulation analysis of multiple sets of parameters, compare the airflow diffusion data and cleanliness correlation of different schemes.

[0041] S5. Based on the simulation analysis results, filter the parameter adjustment schemes, generate pre-debugging instructions containing specific parameter adjustment values, and output them to the field debugging terminal.

[0042] Furthermore, in step S4, when simulating the adjustment of the adjustable parameters of the air outlet, the adjustment range of each adjustable parameter is first divided according to a preset step size. Then, multiple sets of parameter combinations are constructed using the control variable method. During the simulation of each set of parameter combinations, other parameters are fixed, only a single parameter is adjusted, and the corresponding airflow trajectory data and cleanliness distribution data are recorded. After all the combination simulations are completed, all data are classified and sorted according to cleanliness-related indicators to form a correlation dataset of parameters and airflow and cleanliness.

[0043] Compared with existing technologies, this airflow organization optimization design system and method based on cleanroom air outlets has the following advantages:

[0044] This invention constructs a digital twin model of a cleanroom using computer-aided design and links it with real-time operational data. This enables the transformation of airflow organization optimization from static design to dynamic response, solving the problem of existing optimization schemes being disconnected from actual operating scenarios. By automatically identifying problematic air vents and simulating and optimizing parameter adjustment schemes, it replaces the traditional blind debugging that relies on manual experience, avoiding repeated trial and error, improving the accuracy and efficiency of debugging, ensuring the stable maintenance of cleanroom cleanliness indicators, and reducing the impact of the debugging process on production progress. This provides strong support for the efficient and reliable operation of cleanrooms.

[0045] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0047] Figure 1 This is a schematic diagram of a system for optimizing airflow organization based on cleanroom air vents;

[0048] Figure 2A flowchart illustrating an optimized airflow organization design method based on cleanroom air vents;

[0049] Figure 3 This is a flowchart of a method for optimizing airflow organization based on cleanroom air vents. Detailed Implementation

[0050] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structure, features and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0051] This invention provides an airflow organization optimization design system and method based on cleanroom air vents. The core aim is to solve the problems of existing technologies where static design is disconnected from actual dynamic operating conditions and relies on manual adjustments, resulting in low efficiency. (See also...) Figure 1 , Figure 2 and Figure 3 The specific details are as follows:

[0052] The system comprises five modules, which communicate through a pre-defined data interaction protocol. The digital twin model construction module, based on computer-aided design technology, collects the three-dimensional spatial dimensions of the cleanroom, as well as relevant parameters of equipment and air vents. It then builds a three-dimensional geometric model and associates the parameters. After collision detection and verification, it forms a digital twin model that perfectly maps to the actual scene and supports model updates after space renovations or adjustments to equipment or air vents.

[0053] The real-time data acquisition module connects to sensors in the cleanroom via the Internet of Things. The sensors are evenly distributed according to functional zones, covering key areas and areas prone to airflow abnormalities. They collect real-time data such as particulate matter concentration, temperature, and humidity, which are then transmitted after being standardized and encrypted. The acquisition frequency is adapted to the operating conditions.

[0054] The problem identification module receives standardized data, processes it by reducing noise and removing outliers, and compares it with preset standard thresholds in real time. Combined with relevant information about hot spots in the digital twin model, it locates the hot spots and generates a list.

[0055] The parameter simulation optimization module extracts the parameters of the problem air outlet and local data of the model, sets the adjustable parameter adjustment range and step size, simulates in the model in the combination order, records data such as airflow trajectory and cleanliness distribution to form a dataset, has built-in constraint rules based on relevant parameter settings, and supports rule update adaptation.

[0056] The pre-debugging scheme output module receives the simulation dataset, extracts key data, sorts and filters target parameter combinations according to preset rules, integrates vent labels, adjustment values ​​and operation steps, generates text and graphical (including vent location labels and airflow distribution diagrams) pre-debugging instructions, and pushes them to the on-site debugging terminal.

[0057] The corresponding optimization design method first constructs a digital twin model, then collects real-time data through sensors and uploads it to the system. The data is mapped to the model and compared with standard thresholds to identify problems and locate air vents. Subsequently, the parameters of the problematic air vents are simulated and adjusted within the model. After multiple sets of combined simulation analyses, a final solution is selected to generate pre-debugging instructions. This solution realizes the transformation of airflow organization optimization from static to dynamic, improving the accuracy and efficiency of debugging and ensuring the stability of cleanroom cleanliness.

[0058] Example 1

[0059] This embodiment applies to a sterile production cleanroom in the biopharmaceutical industry. This cleanroom needs to maintain a strict cleanliness level to ensure that drug production is not affected by airborne particulate matter, microorganisms, or other contaminants. Its internal layout includes multiple functional zones such as a raw material pretreatment area, a formulation production area, and a finished product inspection area. The process equipment in each area is densely packed, and airflow is easily obstructed by the equipment, forming eddies and dead zones. Traditional airflow organization and adjustment methods, which rely on manual experience, often result in unstable cleanliness and long adjustment cycles, severely impacting production progress. See also... Figure 1 , Figure 2 and Figure 3 This embodiment achieves dynamic optimization and precise adjustment of airflow organization through a cleanroom vent-based airflow organization optimization design system and method, meeting the stringent requirements of biopharmaceutical production for a clean environment.

[0060] In the specific implementation process, the digital twin model construction module was first activated to carry out model building. The three-dimensional spatial dimensions of the biomedical cleanroom, the installation location and structural parameters of process equipment within each functional area were collected, and key data such as the installation coordinates, structural form, and adjustable parameters of each air vent were accurately recorded. Computer-aided design tools were then used to build a basic geometric model based on the collected three-dimensional spatial data. Subsequently, the process equipment parameters, air vent parameters, and the three-dimensional geometric model were linked and bound together to form an initial model with parameter attributes.

[0061] The initial model undergoes spatial collision detection to carefully verify its matching accuracy with the actual cleanroom scenario. This ensures the model perfectly maps the actual cleanroom's spatial structure, equipment layout, and air vent configuration, ultimately forming a complete digital twin model. The model data is then synchronously transmitted to the problem identification module and the parameter simulation optimization module. When subsequent changes occur in the cleanroom, such as spatial modifications, equipment additions or removals, or air vent position adjustments, relevant data for the changed areas are re-collected. The geometric structure and associated parameters of the corresponding areas are modified based on the original model. After collision detection verification, the updated model data is synchronized to the relevant modules to ensure the model remains consistent with the actual scenario.

[0062] Next, the real-time data acquisition module is activated, and sensors are evenly deployed according to the functional zones of the cleanroom, with a focus on covering key process areas such as the formulation production area and densely packed equipment areas prone to airflow anomalies. The sensor data acquisition frequency is set according to the cleanroom's operating conditions to ensure timely capture of dynamic changes in the indoor environment. After the sensors collect real-time operating data such as particulate matter concentration, temperature, and humidity, the data is first standardized in format, unifying the units, precision, and storage format to ensure that the data can be directly accessed by the problem identification module. The data transmission process uses an encryption protocol to ensure the security and integrity of data transmission, and then the standardized real-time data is continuously transmitted to the problem identification module.

[0063] After receiving standardized operational data from the real-time data acquisition module, the problem identification module uses a moving average method to denoise the data, removing interference noise. It then uses preset outlier judgment rules to eliminate abnormal data exceeding reasonable ranges, ensuring data reliability. The processed valid data is continuously compared in real-time with preset standard thresholds for cleanliness, temperature, and humidity to monitor whether indoor environmental parameters meet requirements. Combining the airflow coverage, operational parameters, and spatial relationships of the air vents in the digital twin model, the module accurately locates the corresponding air vents causing airflow problems, forming a list of problematic air vents.

[0064] In the specific implementation of this embodiment, it is also necessary to calculate the influence weight of each air outlet on the abnormal area, using the following formula: .in For the first The weight of the impact of each wind vent on abnormal areas For the first Real-time airflow at each vent For the first Distance attenuation coefficient from each air outlet to the abnormal area This represents the total number of wind tunnels participating in airflow regulation around the abnormal area. It is derived from the theory of cleanroom airflow dynamics and determined by fitting the measured data of airflow attenuation at different distances in similar biomedical cleanrooms. This formula can accurately determine the degree of influence of each air outlet on abnormal areas, providing a basis for subsequent parameter optimization.

[0065] After receiving the problem and corresponding air outlet location information transmitted by the problem identification module, the parameter simulation optimization module extracts the current operating parameters, structural parameters, and corresponding local data of the digital twin model of the problem air outlet from the list of problem air outlets. Based on the air outlet structural parameters and relevant parameters of the air supply system, the module sets the adjustment range of adjustable parameters such as air volume and air supply angle, and divides the adjustment step size according to the principle of equal distribution to ensure the comprehensiveness and rationality of parameter adjustment.

[0066] This module incorporates parameter adjustment constraint rules, which are set based on the air outlet structure design parameters, air supply system operating parameters, and cleanroom process requirements. During simulation, only parameter combinations that conform to these constraints are generated, avoiding invalid parameter combinations that exceed equipment capabilities or do not meet process requirements. Furthermore, the constraint rules support manual updates or automatic adaptation based on changes in the cleanroom's usage scenario. Simulations are initiated sequentially in the digital twin model according to the parameter combination order, maintaining consistent simulation environment parameters throughout each simulation to ensure the comparability of simulation results. Simultaneously, the airflow trajectory coordinate data and cleanliness distribution matrix data corresponding to each parameter combination are recorded, forming multiple simulation datasets.

[0067] In the specific implementation of this embodiment, the comprehensive evaluation index of each parameter combination is calculated using the following formula: .in For the first A comprehensive evaluation index for the combination of parameters. This is the quantified value of airflow uniformity under this parameter combination. This is the quantitative value for cleanliness compliance under this parameter combination. and These are the weighting coefficients. and Based on the determination of cleanroom process priorities, the weights recommended by relevant industry cleanliness level standards are combined with the specific process requirements of biomedical cleanrooms. This comprehensive evaluation index can fully assess the optimization effect of each parameter combination.

[0068] After receiving multiple sets of simulation datasets transmitted by the parameter simulation optimization module, the pre-debugging scheme output module extracts the airflow diffusion data and cleanliness-related data for each scheme. In the specific implementation of this embodiment, the multiple schemes are sorted according to preset screening rules, and the formula used to calculate the sorting score is as follows: .in For the first The ranking score of the group scheme, This represents the improvement in airflow uniformity of this scheme compared to the baseline scheme. This represents the improvement in cleanliness compared to the baseline solution. The difficulty of adjusting the parameters of this scheme is quantified. , , These are the weighting coefficients. , , The results were obtained by constructing a judgment matrix based on the priorities of airflow improvement, cleanliness requirements, and on-site operational feasibility using the analytic hierarchy process.

[0069] Based on the ranking scores, the optimal combination of target parameters with the best overall performance is selected. The identification information of the problematic air vents, the specific adjustment values ​​of the target parameters, and the step-by-step adjustment procedures are integrated. The integrated information is converted into both text and graphical formats. The graphical format clearly marks the location of the air vents and displays a schematic diagram of the airflow distribution before and after parameter adjustment, facilitating intuitive understanding by on-site personnel. Subsequently, both forms of pre-debugging instructions are pushed to the on-site commissioning terminal.

[0070] On-site staff can precisely debug the problematic air outlets based on the pre-debugging instructions received by the on-site debugging terminal, eliminating the need for repeated trial and error and significantly shortening the debugging time.

[0071] In summary, this embodiment, through the complete implementation process described above, realizes the transformation of cleanroom airflow organization optimization from static design to dynamic response, effectively solving the problem of the disconnect between optimization schemes and actual operating scenarios in traditional technologies. By linking digital twin models with real-time data, airflow problems can be identified promptly and problematic air vents can be accurately located. Combined with scientific formula calculations and multi-parameter simulation optimization, an optimal pre-debugging scheme is formulated, replacing the traditional blind debugging method relying on manual experience and avoiding the risk of problem escalation due to repeated trial and error. This not only significantly improves the accuracy and efficiency of debugging, ensuring the stable maintenance of cleanliness indicators in biopharmaceutical cleanrooms and meeting the stringent requirements of drug production, but also greatly reduces the impact of the debugging process on production schedules, lowers cleanroom operating energy consumption, and extends the service life of related equipment, providing strong support for the efficient and reliable operation of cleanrooms and possessing significant practical application value.

[0072] Example 2

[0073] This embodiment applies to a cleanroom for semiconductor chip manufacturing. This cleanroom has extremely high requirements for air cleanliness, temperature and humidity stability, and airflow uniformity. Core processes such as chip lithography and etching must avoid any contamination from minute particles or airflow disturbances; otherwise, chip yield will drop significantly. The cleanroom is filled with large pieces of equipment such as high-precision lithography and etching machines. The dense layout of the equipment generates significant heat, and airflow is easily obstructed by the equipment, creating localized eddies. Simultaneously, personnel movement and process start-ups and shutdowns cause dynamic changes in the distribution of pollutants within the cleanroom. Traditional static optimization designs are ill-suited to these dynamic conditions, necessitating a real-time, precisely optimized airflow organization scheme.

[0074] This embodiment, based on the aforementioned Embodiment 1, incorporates targeted optimizations for the process characteristics of semiconductor cleanrooms. (See [link to previous embodiment]). Figure 1 , Figure 2 and Figure 3 The specific details are as follows:

[0075] First, following the core construction process of the digital twin model building module, additional data such as heat dissipation parameters and airflow obstruction coefficients of semiconductor process equipment are collected. This data is then linked and bound to the original model parameters to improve the model's simulation accuracy of the impact of equipment heat dissipation on airflow. After the model is built, a dedicated model update monitoring mechanism is established to monitor in real time whether the cleanroom experiences equipment additions or removals, air vent adjustments, or spatial layout modifications due to process upgrades. Once a change occurs, a data re-collection and model modification process is immediately triggered. After collision detection verification, the model is synchronously updated to the relevant modules to ensure that it accurately reflects the dynamic changes of the semiconductor cleanroom.

[0076] Regarding the real-time data acquisition module, in addition to evenly distributing sensors according to functional zones, the sensor density in core process areas such as the photolithography and etching zones is significantly increased. Dedicated sensors are also added around equipment heat dissipation vents and airflow convergence areas to achieve high-density, high-precision acquisition of environmental parameters in key areas. The sensor acquisition frequency is dynamically adjusted according to the operating conditions of different process stages in the semiconductor cleanroom, with the acquisition frequency appropriately increased during process start-up and shutdown to ensure the capture of sudden changes in environmental parameters during condition switching. In the data processing stage, based on format standardization and encrypted transmission, a data timeliness verification step is added to eliminate data with excessive transmission delays, ensuring that the data transmitted to the problem identification module is both accurate and real-time.

[0077] After receiving real-time data, the problem identification module first uses a moving average method to reduce noise and remove outliers, and then continuously compares it with the cleanliness standard threshold specific to the semiconductor industry. Combining the relative positional relationship between the air vents and core process equipment, the airflow coverage range, and operating parameters in the digital twin model, the module accurately locates the problematic air vents and generates a list.

[0078] In the specific implementation of this embodiment, the formula is used. The influence weight of each air outlet on the abnormal area is calculated. This weight clarifies the contribution of each air outlet to the airflow anomaly in the core process area, providing a precise target for subsequent parameter optimization.

[0079] The parameter simulation optimization module, based on the aforementioned embodiments, adds a parameter adjustment sensitivity analysis step. After extracting the parameters of the problematic air outlet and local model data, in addition to setting the adjustment range and step size of the adjustable parameters, it analyzes the sensitivity of different parameter adjustments to airflow changes in conjunction with the dynamic response characteristics of the semiconductor cleanroom air supply system.

[0080] During the simulation, multiple parameter combinations were constructed using the controlled variable method. In each simulation, all other parameters were fixed while only one parameter was adjusted. Simultaneously, airflow trajectory coordinate data, cleanliness distribution matrix data, and the system response time after parameter adjustment were recorded. In the specific implementation of this embodiment, the formula... The comprehensive evaluation index for each parameter combination is calculated. Combined with sensitivity analysis results, parameter combinations that significantly improve airflow uniformity and cleanliness and exhibit rapid adjustment response are prioritized, forming multiple sets of high-efficiency simulation datasets. Simultaneously, the module's built-in parameter adjustment constraint rules supplement the limitations on airflow velocity imposed by semiconductor processes, ensuring that the generated parameter combinations comply with both equipment operating specifications and the specific requirements of core processes.

[0081] After receiving the simulation dataset, the pre-debugging scheme output module extracts airflow diffusion data, cleanliness-related data, and system response time data. In the specific implementation of this embodiment, the formula is used. The calculation scheme is ranked and scored, and the system response time is included in the calculation of the adjustment difficulty quantification value. Priority is given to the target parameter combination with low adjustment difficulty, good improvement effect and fast response.

[0082] After integrating the problematic air vent identification information, target parameter adjustment values, and step-by-step operation procedures, text and graphical pre-debugging instructions are generated. The graphical instructions include new labels indicating the relative positions of the core process equipment and the air vents, as well as a schematic diagram of the airflow velocity distribution after parameter adjustment. This facilitates on-site personnel to quickly understand and execute the debugging operations. Subsequently, the instructions are pushed to the on-site debugging terminal. After the on-site personnel complete the precise debugging according to the pre-debugging instructions, the system continuously monitors the environmental parameters of the core process area to ensure that the airflow organization is always in the optimal state.

[0083] In summary, this embodiment addresses the specific needs of cleanrooms in semiconductor chip manufacturing. Building upon the aforementioned embodiments, it features targeted optimizations. By supplementing with dedicated data acquisition, enhancing core area monitoring, incorporating parameter sensitivity analysis, and considering system response time, it further improves the accuracy and adaptability of airflow organization optimization. This effectively solves the airflow optimization challenges posed by the dense equipment, high heat dissipation, and dynamic changes in operating conditions in semiconductor cleanrooms. It avoids airflow disturbances and particulate contamination in core process steps, significantly improving chip manufacturing yield. Simultaneously, it maintains the core advantages of precise debugging and avoiding repeated trial and error, greatly shortening the debugging cycle, reducing the impact on production schedule, and lowering cleanroom operating energy consumption. This provides more targeted technical support for the efficient and stable operation of semiconductor cleanrooms, broadening the application scenarios and scope of this invention.

[0084] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A system for optimizing airflow organization based on cleanroom air vents, characterized in that, The system includes: a digital twin model construction module, a real-time data acquisition module, a problem identification module, a parameter simulation optimization module, and a pre-debugging scheme output module; The digital twin model building module is based on computer-aided design technology to build a digital twin model that is completely mapped to the actual cleanroom space structure, equipment layout and air outlet configuration, and transmits the model data to the problem identification module and parameter simulation optimization module; The real-time data acquisition module establishes a communication connection with the cleanroom sensors through Internet of Things (IoT) technology, collects real-time operating data on indoor particulate matter concentration, temperature, and humidity, and transmits the data to the problem identification module. The problem identification module receives real-time operating data and digital twin model data, compares them with preset standard thresholds to identify airflow-related problems, and transmits the problems and corresponding air outlet location information to the parameter simulation optimization module. The parameter simulation optimization module targets the problematic air outlet, simulates and adjusts the adjustable parameters of the air outlet in a digital twin model, and analyzes the airflow changes and cleanliness-related data through multiple sets of parameter combinations, and transmits the simulation results to the pre-debugging scheme output module. The pre-debugging scheme output module filters parameter adjustment schemes based on simulation results and generates pre-debugging instructions.

2. The airflow organization optimization design system based on cleanroom air outlets according to claim 1, characterized in that, The digital twin model building module includes the following steps when building the model: Collect the three-dimensional spatial dimensions of the cleanroom, the installation location of process equipment, the structural parameters of process equipment, the installation coordinates of air outlets, the structural form of air outlets, and the adjustable parameters of air outlets; Computer-aided design tools are used to build a 3D geometric model based on the collected data. The process equipment parameters, air outlet parameters and three-dimensional geometric model are linked and bound to form an initial model with parameter attributes; Spatial collision detection is performed on the initial model to verify the matching accuracy between the model and the actual cleanroom scene, thus forming the final digital twin model.

3. The airflow organization optimization design system based on cleanroom air outlets according to claim 1, characterized in that, The sensors in the real-time data acquisition module are evenly distributed according to the functional zones of the cleanroom, covering key process areas and areas prone to airflow abnormalities. The frequency of sensor data acquisition is set according to the operating conditions of the cleanroom. The data transmission process adopts an encryption protocol. Before transmission, the acquired data is standardized to unify the units, accuracy, and storage format of the data.

4. The airflow organization optimization design system based on cleanroom air outlets according to claim 1, characterized in that, The problem identification module includes the following steps when identifying airflow-related problems: It receives standardized operational data transmitted from the real-time data acquisition module, performs noise reduction on the data using the moving average method, and removes data that exceeds the reasonable range through preset outlier judgment rules. The processed data is continuously compared with a preset standard threshold in real time. By combining the airflow coverage, operating parameters, and spatial relationships of the vents in the digital twin model, the corresponding vents that cause airflow problems are located, and a list of problematic vents is formed.

5. The airflow organization optimization design system based on cleanroom air outlets according to claim 1, characterized in that, The parameter simulation optimization module includes the following steps when performing parameter simulation optimization: Extract the current operating parameters, structural parameters, and corresponding local data of the digital twin model of the problematic areas from the list of problematic areas; Based on the air outlet structure parameters and relevant parameters of the air supply system, the adjustment range of the adjustable parameters is set, and the adjustment step size is divided according to the principle of equality. Simulations are started sequentially in the digital twin model according to the parameter combination order, and the simulation environment parameters are kept consistent for each group of simulations; Simultaneously record the airflow trajectory coordinate data and cleanliness distribution matrix data corresponding to each set of parameter combinations to form multiple sets of simulation datasets.

6. The airflow organization optimization design system based on cleanroom air outlets according to claim 1, characterized in that, The pre-debug scheme output module includes the following steps when generating pre-debug instructions: Receive the simulation dataset transmitted by the parameter simulation optimization module, and extract the airflow diffusion data and cleanliness-related data of each scheme. Multiple schemes are sorted according to preset filtering rules, target parameter combinations are selected, and the identification information of the problem hotspots, the specific adjustment values ​​of the target parameters, and the step-by-step adjustment operation steps are integrated. The integrated information was converted into text and graphical formats, respectively. The graphical format included vent location markings and airflow distribution diagrams before and after parameter adjustments, and then pushed to the on-site commissioning terminal.

7. The airflow organization optimization design system based on cleanroom air outlets according to claim 1, characterized in that, The digital twin model building module includes the following steps when updating the model: Monitor whether the cleanroom has undergone spatial modifications, equipment additions or removals, or air vent position adjustments. If any of these changes occur, re-collect the three-dimensional spatial data, equipment parameters, or air vent parameters of the changed area. Modify the geometric structure and associated parameters of the corresponding regions based on the original digital twin model; After the modifications are completed, collision detection and verification are performed, and the updated complete model data is synchronized to the problem identification module and the parameter simulation optimization module.

8. The airflow organization optimization design system based on cleanroom air outlets according to claim 1, characterized in that, The parameter simulation optimization module has built-in parameter adjustment constraint rules. The constraint rules are set based on the air outlet structure design parameters, air supply system operation parameters and cleanroom process requirements. During the simulation, only parameter combinations that meet the constraint rules are generated. The constraint rules can be manually updated or automatically adapted according to changes in the cleanroom usage scenario.

9. A method for optimizing airflow organization based on cleanroom air vents, applicable to the airflow organization optimization design system based on cleanroom air vents as described in any one of claims 1-8, characterized in that, The method includes the following steps: S1. Using computer-aided design technology, input relevant data on the three-dimensional spatial structure, equipment layout, and air outlet configuration of the cleanroom to construct a digital twin model that perfectly matches the actual cleanroom. S2. Sensors connected to the Internet of Things continuously collect real-time operating data on particulate matter concentration, temperature and humidity in various areas of the cleanroom and upload the data to the system. S3. Real-time operating data is synchronously mapped to a digital twin model and compared with preset standard thresholds to automatically identify airflow-related problems and locate the corresponding air outlets; S4. In the digital twin model, simulate and adjust the adjustable parameters for the problematic air outlet. Through simulation analysis of multiple sets of parameters, compare the airflow diffusion data and cleanliness correlation of different schemes. S5. Based on the simulation analysis results, filter the parameter adjustment schemes, generate pre-debugging instructions containing specific parameter adjustment values, and output them to the field debugging terminal.

10. The airflow organization optimization design method based on cleanroom air outlets according to claim 9, characterized in that, In step S4, when simulating the adjustment of the adjustable parameters of the air outlet, the adjustment range of each adjustable parameter is first divided according to the preset step size. Then, multiple sets of parameter combinations are constructed using the control variable method. During the simulation of each set of parameter combinations, other parameters are fixed, only a single parameter is adjusted, and the corresponding airflow trajectory data and cleanliness distribution data are recorded. After all the combination simulations are completed, all data are classified and sorted according to cleanliness-related indicators to form a correlation dataset of parameters and airflow and cleanliness.

Images