Intelligent constant humidity control FFU integrated system and dust-free workshop application method thereof

Abstract

The invention discloses an intelligent constant-humidity control FFU integrated system and a dust-free workshop application method thereof, and relates to the technical field of workshop application, and the method comprises an environment sensing unit which is used for collecting internal and external environment parameters and equipment operation states of a workshop in real time; the prediction unit is used for training and generating a model with a time sequence prediction capability by fusing historical production data and real-time environment fluctuation characteristics so as to output a temperature and humidity change trend and a key adjustment time window of each partition of the workshop in a future time period; the decision-making unit is used for dynamically selecting a target operation strategy from a process priority mode, an energy-saving mode and an emergency mode based on real-time production process requirements, an environment sensitivity threshold value and an energy consumption constraint condition; the parameter optimization unit is used for performing wind field parameter adjustment on the FFU unit under the target operation strategy, and minimizing system energy consumption while maintaining uniform distribution of temperature and humidity; adjustment comprises fan rotating speed gradient configuration, air outlet angle cooperative control and airflow circulation path optimization.

Description

Technical Field

[0001] This application relates to the technical field of workshop applications, and more particularly, to an intelligent constant humidity control FFU integrated system and its application method in a clean workshop. Background Art

[0002] In modern industrial production, special environments such as clean workshops and constant temperature and humidity workshops have strict requirements for humidity control. For example, in places such as pharmaceutical clean workshops, food clean workshops, and medical device clean workshops, precise humidity control is of great significance for ensuring product quality and preventing the growth of microorganisms.

[0003] However, when existing humidity control technologies are applied to clean workshops, there are often some deficiencies. On the one hand, traditional humidity control equipment and FFU usually operate independently, lacking effective cooperation, resulting in less than ideal accuracy and stability of humidity control; on the other hand, the monitoring and adjustment of temperature and humidity often rely on manual control, which is not timely enough to meet the humidity requirements of some production environments.

[0004] For the above problems, no effective solutions have been proposed yet. Summary of the Invention

[0005] Embodiments of this application provide an intelligent constant humidity control FFU integrated system and its application method in a clean workshop to solve the above technical problems.

[0006] This application provides an intelligent constant humidity control FFU integrated system, including:

[0007] An environmental perception unit, composed of a temperature and humidity sensor, a production equipment heat load monitoring module, and a meteorological data interface, for real-time collection of internal and external environmental parameters of the workshop and the operating status of equipment;

[0008] A prediction unit, for training and generating a model with time series prediction ability by fusing historical production data and real-time environmental fluctuation characteristics, to output the temperature and humidity change trends and key adjustment time windows of each partition of the workshop in the future time period;

[0009] A decision-making unit, for dynamically selecting a target operation strategy from a process priority mode, an energy-saving mode, and an emergency mode based on real-time production process requirements, environmental sensitivity thresholds, and energy consumption constraint conditions;

[0010] A parameter optimization unit, for adjusting the wind field parameters of the FFU unit under the target operation strategy by using an optimization algorithm, while maintaining a uniform distribution of temperature and humidity, minimizing the system energy consumption; the adjustment includes fan speed gradient configuration, coordinated control of air outlet angles, and optimization of air flow circulation paths.

[0011] Furthermore, the intelligent constant humidity control FFU integrated system further includes:

[0012] A linkage unit that integrates a UWB personnel positioning system and an infrared heat source detection device, and is used to trigger a stepped attenuation of the FFU power in the partition and a dormancy instruction for the temperature and humidity adjustment equipment when it is detected that there is no personnel activity in the target partition for 10 consecutive minutes and the equipment heat radiation intensity is lower than the threshold;

[0013] A feedback unit that collects the environmental response data and equipment execution status of each partition, constructs a deviation traceability model through a Bayesian network, and directionally feeds it back to the prediction unit and the parameter optimization unit for model iteration and update.

[0014] Furthermore, the production equipment heat load monitoring module in the environmental perception unit is connected to the PLC controller through an industrial bus to parse the equipment start-stop signal, material processing volume, and heat load change curve in real time; the environmental perception unit includes:

[0015] A redundant check network that uses a hexagonal honeycomb topology to cover each partition of the workshop, and three groups of cross-checked temperature and humidity sensing nodes are configured for each partition;

[0016] A meteorological data cleaning module that performs wavelet transform denoising and dynamic time alignment on the temperature, humidity, and air pressure data of an external weather station;

[0017] A production equipment interface module that collects the equipment operation frequency, motor torque, and cooling water flow data in the PLC controller in real time through the OPCUA protocol.

[0018] Furthermore, the prediction unit includes:

[0019] A spatio-temporal hypergraph network module that models each partition of the workshop as a hypergraph node, connects partitions with environmental propagation relationships through hyperedges, and extracts the temperature and humidity diffusion characteristics across partitions using spatio-temporal convolution;

[0020] A time window predictor that analyzes the time interval of historical adjustment events and the Poisson distribution law of production activities based on the hidden Markov jump diffusion model, and predicts the trigger probability and duration of future critical adjustment time windows;

[0021] A feature fusion module that performs tensor fusion on the spatial correlation features output by the hypergraph network and the time window prediction results to generate a partition temperature and humidity evolution map with an uncertainty interval.

[0022] Furthermore, the prediction unit further includes:

[0023] A physical constraint network that generates physically reasonable extreme condition training data by fusing the thermodynamic equations of workshop equipment and the boundary conditions of fluid mechanics, including scenarios such as sudden equipment overload, ventilation duct blockage, and material moisture absorption mutation;

[0024] A search module that reconstructs the hypergraph network topology based on the partition layout change data collected in real time. Its search strategies include:

[0025] Dynamically optimize the hyperedge connection method through a differentiable architecture search algorithm to make the hypergraph network adapt to partition reorganization;

[0026] Introduce topological persistence constraints to ensure smooth transition of the network structure between adjacent production batches;

[0027] Introduce a meta-knowledge distillation engine to encode the laws of workshop environment response implicit in the historical prediction model into a lightweight rule set as the prior for model initialization in the new scenario.

[0028] Furthermore, the decision-making unit includes:

[0029] An attention-enhanced reinforcement learning framework that takes the hypergraph features and time window probabilities output by the prediction unit as state inputs and extracts environment sensitivity features at different spatial scales through a hierarchical attention mechanism;

[0030] A fuzzy differential game controller that constructs a game model between the production department and the equipment department and uses fuzzy differential equations to solve the Pareto optimal strategy;

[0031] A risk-constrained policy distillation module that extracts time window constraint rules from the event-driven model of the prediction unit, encodes them as action masks for reinforcement learning, and restricts the selection of high-risk strategies.

[0032] Furthermore, the decision-making unit also includes:

[0033] A heterogeneous policy fusion module for performing the following operations: extracting key propagation path features from the hypergraph network of the prediction unit to construct an environmental fluctuation propagation probability map; using a random walk graph attention mechanism to simulate the propagation influence of different decision-making strategies in the probability map to generate a risk heat map; fusing the risk heat map and real-time production yield data through a dual-stream gated network to dynamically adjust the utility function of the game model;

[0034] A causal reinforcement intervention module that executes a closed-loop operation chain: when the feedback unit detects an environmental deviation, uses a counterfactual causal reasoning algorithm to deduce the optimal intervention action set; constructs a virtual twin decision-making environment to pre-evaluate the long-term impact of intervention actions in the digital twin; reshapes the strategy through policy gradients and migrates the optimized strategy in the virtual environment to the physical system.

[0035] Furthermore, the linkage unit includes:

[0036] A radar micro-motion perception system that distinguishes between short-term personnel stays and long-term departures through Doppler feature analysis;

[0037] A thermal inertia prediction module constructs a thermal radiation attenuation model after the equipment stops based on the material heat conduction equation;

[0038] A power mapping module calculates the FFU power attenuation curve according to the thermal inertia model to control the temperature and humidity fluctuation not exceeding the process threshold.

[0039] Furthermore, the deviation traceability analysis module of the feedback unit includes:

[0040] A causal discovery engine uses the PC algorithm based on conditional independence to identify the key causal links of environmental control deviation;

[0041] An incremental knowledge distillation component converts the traceability result into a lightweight knowledge graph and injects it into the prediction model adjustment process.

[0042] The present application provides a method for applying an intelligent constant humidity control FFU integrated system in a dust-free workshop, including:

[0043] Deploy temperature and humidity sensors, a production equipment heat load monitoring module and a meteorological data interface inside the dust-free workshop to collect the internal and external environmental parameters and the equipment operation status in real time;

[0044] Fuse historical production data and real-time environmental fluctuation characteristics, and use the trained model with time series prediction ability to output the temperature and humidity change trends and key adjustment time windows of each partition in the workshop in the future time period;

[0045] Based on the real-time production process requirements, environmental sensitivity threshold and energy consumption constraint conditions, dynamically select the target operation strategy from the process priority mode, energy-saving mode and emergency mode;

[0046] Use an optimization algorithm to adjust the wind field parameters of the FFU unit under the target operation strategy, including fan speed gradient configuration, outlet angle collaborative control and air flow circulation path optimization, to minimize the energy consumption of the intelligent constant humidity control FFU integrated system while maintaining uniform temperature and humidity distribution.

[0047] Based on the embodiments provided by the present application, the coordination between the FFU unit and the temperature and humidity adjustment equipment is realized. A dynamic prediction model is constructed through multi-source data fusion (environmental parameters, production plan, meteorological data) to improve the humidity control accuracy and meet the requirements for humidity fluctuation in key areas in the GMP specification of the pharmaceutical workshop. The parameter optimization module is used to reduce the system energy consumption on the premise of ensuring temperature and humidity uniformity. Based on the production process requirements and environmental sensitivity threshold, the system can complete the operation strategy switching in time (such as switching from the process priority mode to the emergency mode), and the response speed is significantly improved compared with manual operation, effectively avoiding wafer oxidation defects caused by sudden humidity changes in the semiconductor workshop. Description of the Drawings

[0048] The accompanying drawings described herein are used to provide a further understanding of the embodiments of the present invention, and constitute a part of this application. The illustrative embodiments of this application and their descriptions are used to explain this application, and do not constitute an improper limitation to this application. In the drawings:

[0049] Figure 1 is a structural diagram of an optional intelligent constant humidity control FFU integrated system according to an embodiment of the present application;

[0050] Figure 2 is a flowchart of an optional causal reinforcement intervention module executing a closed-loop operation chain according to an embodiment of the present application;

[0051] Figure 3 is a flowchart of an optional method for applying an intelligent constant humidity control FFU integrated system in a dust-free workshop according to an embodiment of the present application.

[0052] The realization of the object of the present invention, functional features and advantages will be further described with reference to the embodiments and the accompanying drawings. Detailed Embodiments

[0053] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0054] Optionally, as Figure 1 shown, this application provides an intelligent constant humidity control FFU integrated system, including:

[0055] An environment perception unit 101, composed of a temperature and humidity sensor, a production equipment heat load monitoring module and a meteorological data interface, is used to collect the internal and external environment parameters of the workshop and the operation status of the equipment in real time;

[0056] Among them, there are multiple temperature and humidity sensors, which are evenly distributed at different positions in the workshop to achieve comprehensive monitoring of the temperature and humidity in the workshop;

[0057] A prediction unit 102 is used to train and generate a model with time series prediction ability by fusing historical production data and real-time environmental fluctuation characteristics, so as to output the temperature and humidity change trends and key adjustment time windows of each partition in the workshop in the future time period;

[0058] A decision-making unit 103 is used to dynamically select a target operation strategy from a process priority mode, an energy-saving mode and an emergency mode based on real-time production process requirements, environmental sensitivity thresholds and energy consumption constraint conditions;

[0059] The parameter optimization unit 104 is used to adjust the wind field parameters of the FFU unit under the target operation strategy by using an optimization algorithm, while maintaining the uniform distribution of temperature and humidity and minimizing the system energy consumption; the adjustment includes the configuration of the fan speed gradient, the coordinated control of the air outlet angle, and the optimization of the air flow circulation path.

[0060] Based on the embodiments provided in this application, the coordination between the FFU unit and the temperature and humidity adjustment equipment is realized. By constructing a dynamic prediction model through multi-source data fusion (environmental parameters, production plans, meteorological data), the accuracy of humidity control is improved, meeting the requirements for humidity fluctuations in key areas in the GMP specification of the pharmaceutical workshop. The parameter optimization module is used to reduce the system energy consumption on the premise of ensuring the uniformity of temperature and humidity. Based on the production process requirements and the environmental sensitivity threshold, the system can complete the operation strategy switch in a timely manner (such as switching from the process priority mode to the emergency mode), and the response speed is significantly improved compared with manual operation, effectively avoiding the wafer oxidation defects caused by sudden changes in humidity in the semiconductor workshop.

[0061] Furthermore, the intelligent constant humidity control FFU integrated system further includes:

[0062] The linkage unit integrates the UWB personnel positioning system and the infrared heat source detection device, and is used to trigger the stepped attenuation of the FFU power in the target partition and the dormancy instruction of the temperature and humidity adjustment equipment when it is detected that there is no personnel activity in the target partition for 10 consecutive minutes and the equipment heat radiation intensity is lower than the threshold;

[0063] Wherein, the threshold is a standard value used to judge whether the equipment is in a low heat load state. When the equipment heat radiation intensity is lower than this threshold, combined with the situation of no personnel activity, the system will consider that the heat load in this area is low and can trigger energy-saving measures.

[0064] For example, in an electronic manufacturing workshop, when the production equipment is in a standby or low-power operation state, its heat radiation intensity may be low. Assume that the equipment heat radiation intensity threshold is set to 50W / m 2 , when the infrared heat source detection device monitors that the equipment heat radiation intensity in the target partition is continuously lower than 50W / m 2 for 10 consecutive minutes, and at the same time the UWB personnel positioning system detects that there is no personnel activity in this area, the system will trigger the stepped attenuation of the FFU power in the target partition and the dormancy instruction of the temperature and humidity adjustment equipment to reduce energy consumption.

[0065] Based on the embodiments provided in this application, the collaborative judgment mechanism of UWB positioning and infrared heat source avoids mis-triggering by a single sensor (such as misjudgment caused by equipment waste heat), ensuring the reliability of the dormancy instruction.

[0066] In some high-precision instrument manufacturing workshops, the control of the environmental temperature and humidity is more stringent, and the equipment heat radiation intensity threshold may be set lower, such as 30W / m2 to ensure that the minute heat changes of the equipment will not affect the environment. In some workshops with relatively loose requirements for temperature and humidity, this threshold may be appropriately increased, such as 70W / m 2 to meet different production requirements.

[0067] The feedback unit collects the environmental response data and equipment execution status of each partition, constructs a deviation traceability model through a Bayesian network, and directionally feeds it back to the prediction unit and the parameter optimization unit for model iteration and update.

[0068] Furthermore, the production equipment heat load monitoring module in the environmental perception unit is connected to the PLC controller through an industrial bus to parse the equipment start-stop signal, material handling volume, and heat load change curve in real time; the environmental perception unit includes:

[0069] The redundancy check network uses a hexagonal honeycomb topology to cover each partition of the workshop, and three groups of cross-checked temperature and humidity sensing nodes are configured for each partition;

[0070] The meteorological data cleaning module performs wavelet transform denoising and dynamic time alignment on the temperature, humidity, and air pressure data of the external weather station;

[0071] The production equipment interface module collects the equipment operation frequency, motor torque, and cooling water flow data in the PLC controller in real time through the OPCUA protocol.

[0072] Based on the embodiments provided in this application, the hexagonal honeycomb topology network suppresses the influence of local interference (such as airflow disturbance caused by personnel movement) on data collection through three groups of sensor cross-checks.

[0073] Furthermore, the prediction unit includes:

[0074] The spatio-temporal hypergraph network module models each partition of the workshop as a hypergraph node, connects the partitions with environmental propagation relationships through hyperedges, and uses spatio-temporal convolution to extract the temperature and humidity diffusion characteristics across partitions;

[0075] The time window predictor analyzes the time interval of historical adjustment events and the Poisson distribution law of production activities based on the hidden Markov jump diffusion model to predict the triggering probability and duration of the future critical adjustment time window; among them, the time window predictor is event-driven;

[0076] The feature fusion module performs tensor fusion on the spatial correlation features output by the hypergraph network and the time window prediction results to generate a partition temperature and humidity evolution map with an uncertainty interval.

[0077] Based on the embodiments provided in this application, the spatio-temporal hypergraph network explicitly models the environmental propagation path between regions (such as the humidity impact of the clean area on the buffer area), improving the cross-regional prediction accuracy.

[0078] Furthermore, the prediction unit further includes:

[0079] A physical constraint network that generates physically reasonable extreme-condition training data by fusing the thermodynamic equations of workshop equipment and the hydrodynamic boundary conditions, including scenarios such as sudden equipment overload, vent blockage, and sudden material moisture absorption;

[0080] A search module that reconstructs the hypergraph network topology based on the partition layout change data collected in real time. Its search strategies include: The partition layout change data includes equipment relocation and partition adjustment;

[0081] Dynamically optimize the hyperedge connection method through a differentiable architecture search algorithm to make the hypergraph network adapt to partition recombination;

[0082] Introduce topological persistence constraints to ensure smooth transition of the network structure between adjacent production batches;

[0083] Introduce a meta-knowledge distillation engine to encode the implicit laws of workshop environment response in the historical prediction model into a lightweight rule set as the prior for model initialization in new scenarios.

[0084] In the embodiments of the present application, the dynamic hypergraph topology optimization equation is as follows:

[0085]

[0086] Among them, W is a hypergraph adjacency matrix, and the matrix element W kn represents the environmental propagation intensity between partition k and partition n (such as humidity diffusion coefficient, temperature gradient influence, etc.); if there are 3 partitions in the workshop (lithography area, etching area, packaging area), then W can be: Among them, W 12 = 0.8 indicates that the environmental propagation intensity from the lithography area to the etching area is relatively high; ⊙ represents the Hadamard Product, that is, element-wise multiplication of matrices; the role of W⊙M {(t-1)} is to retain the key connections of the historical topology; through the Hadamard Product, the system will preferentially retain the historical valid connections when optimizing the new topology W (t) ; W (t) is the edge weight tensor of the hypergraph network at the optimized time t; L task is the loss function for the temperature and humidity prediction task (including spatio-temporal convolution error); M {(t-1)} is the mask matrix (0 / 1 binary) of the historical topology structure at time t-1; p old and p new represent the probability distributions of the hypergraph structure at different times respectively; p old is the probability distribution representing the old hypergraph structure, while p newRepresents the probability distribution of the new hypergraph structure; λ3KL(p old ||p new ) is the KL divergence between the old and new hypergraph structure distributions; λ1, λ2, λ3 are dynamic balance coefficients;

[0087] In practical applications, the values of λ1, λ2, λ3 need to be adjusted according to specific application scenarios and optimization goals. For example, in a scenario with high requirements for the prediction accuracy of temperature and humidity, the value of λ1 can be set relatively large to pay more attention to the loss function of the prediction task. In a scenario that needs to balance the historical topological structure and the difference between the old and new structures, the weights of λ2 and λ3 can be appropriately increased.

[0088] Suppose in the temperature and humidity control application of a certain dust-free workshop, we hope to ensure the prediction accuracy of temperature and humidity while also considering the stability of the historical topological structure and the smooth transition between the old and new structures. We can set the values of λ1, λ2, λ3 to 0.5, 0.3, and 0.2 respectively. This means that we give a higher weight to the loss function of the temperature and humidity prediction task, and at the same time appropriately consider the mask matrix of the historical topological structure and the difference between the old and new hypergraph structure distributions. Such a setting helps to find a balance point in the optimization process, so that the edge weight tensor W(t) of the hypergraph network can maintain the stability and continuity of the structure while meeting the prediction accuracy requirements.

[0089] In the embodiment of this application,

[0090] H t =σ(W s ×Conv3D(F {t-k:t} )+W h ×G(E {ij} ))

[0091] Among them, H t ∈R N×d represents the fusion feature matrix of each partition at time t (N is the number of partitions, d is the feature dimension); F {t-k:t} is the original temperature and humidity data of the past k time slices; i, j represent the node indices in the hypergraph, which are used to define the node set connected by the hyperedge; E {ij} is the partition set connected by the hyperedge; G() is the hypergraph aggregation function, which calculates the environmental propagation intensity between partitions; W s , W h are learnable weight matrices (initialized by meta-knowledge distillation), which are used to perform linear transformations on the original temperature and humidity data and the partition set connected by the hyperedge; The size of W s is dxd, where d is the feature dimension (for example, 8), and it can be initialized to a small random value, such as W s= [[0.1, 0.2, ..., 0.8], [0.2, 0.3, ..., 0.9], ..., [0.8, 0.9, ..., 0.5]]; W h The size of W is dxd and is also initialized with random small values, such as W h = [[0.3, 0.4, ..., 0.6], [0.4, 0.5, ..., 0.7], ..., [0.6, 0.7, ..., 0.4]];

[0092] σ is an activation function. For example, in the temperature and humidity control application of this dust-free workshop, ReLU can be selected as the activation function, which is defined as σ(x) = max(0, x), that is, when the input value is greater than 0, the value is output, otherwise 0 is output;

[0093] Based on the embodiments provided in this application, differentiable architecture search realizes dynamic adjustment of hypergraph topology to adapt to the requirements of workshop reorganization (such as adding new isolation areas), avoiding the retraining cost of traditional fixed topology models; solving the problem that traditional models cannot capture the environmental propagation across partitions. For example, in a pharmaceutical workshop, the lag effect of humidity change in the filling area on the packaging area.

[0094] Furthermore, the decision-making unit includes:

[0095] An attention-based reinforcement learning framework that takes the hypergraph features and time window probabilities output by the prediction unit as state inputs, and extracts environmental sensitivity features at different spatial scales through a hierarchical attention mechanism; where different spatial scales include partition level, device level, and workshop level;

[0096] A fuzzy differential game controller that constructs a game model between the production department (pursuing maximum yield) and the equipment department (pursuing minimum energy consumption), and uses fuzzy differential equations to solve the Pareto optimal strategy;

[0097] A risk-constrained policy distillation module that extracts time window constraint rules from the event-driven model of the prediction unit, encodes them as action masks for reinforcement learning, and restricts the selection of high-risk policies.

[0098] Based on the embodiments provided in this application, the hierarchical attention mechanism focuses on key partitions (such as the aseptic filling area), reducing the control energy consumption in the edge area while ensuring the accuracy of the core area.

[0099] Furthermore, the decision-making unit also includes a heterogeneous policy fusion module and a causal reinforcement intervention module:

[0100] Among them, the heterogeneous policy fusion module is used to perform the following operations: extract key propagation path features from the hypergraph network of the prediction unit to construct an environmental fluctuation propagation probability map; adopt a random walk graph attention mechanism to simulate the propagation influence of different decision-making strategies in the probability map to generate a risk heat map; fuse the risk heat map with real-time production yield data through a dual-stream gating network to dynamically adjust the utility function of the game model;

[0101] As Figure 2 shown, the causal reinforcement intervention module is used to execute the following closed-loop operation chain:

[0102] S201, when the feedback unit detects an environmental deviation, use the counterfactual causal inference algorithm to deduce the optimal intervention action set;

[0103] S202, construct a virtual twin decision-making environment to preview the long-term impact of intervention actions in the digital twin;

[0104] S203, reshape the policy through policy gradients and transfer the optimized policy in the virtual environment to the physical system.

[0105]

[0106] Among them, Δθ is the policy network parameter update amount, which is output by the policy gradient reshaping; π virt is the virtual twin environment policy, which is generated by the digital twin; Q fact is the actual action value function, which represents the value of the actual executed action a in the state s; Q cf is the counterfactual action value function, which represents the value of executing the counterfactual action a cf in the state s; s is the state, which is composed of the hypergraph features output by the prediction unit and the time window probability; a is the action, which is output by the fuzzy differential game controller and is used to control the operation of the FFU unit; a cf is the counterfactual action, which is used for comparative analysis in the virtual environment; η is the transfer learning rate, which is adaptively adjusted according to the workshop stability and can take a value of 0.001; D is the deviation data set, which is provided by the causal discovery engine;

[0107] Based on the embodiments provided in this application, through the counterfactual causal inference algorithm, the system can deduce the optimal intervention action set, thereby improving the accuracy and reliability of decision-making. Compared with traditional experience-based decision-making methods, this method can more scientifically predict the consequences of different decisions and avoid the risks brought by blind decision-making. By constructing a virtual twin decision-making environment, the system can preview the long-term impact of intervention actions in the digital twin. This enables the system to evaluate the effect of decisions in advance, adjust strategies in a timely manner, and enhance its adaptability to complex environmental changes. By reshaping the policy through policy gradients, the optimized policy in the virtual environment is migrated to the physical system, achieving more efficient energy management and more precise environmental control. The causal reinforcement intervention module is introduced, combining counterfactual causal inference and virtual twin technology to form an innovative decision optimization mechanism. This mechanism is pioneering in intelligent environmental control systems and provides new ideas and methods for decision optimization of similar systems. The transfer learning rate η can be adaptively adjusted according to the stability of the workshop, ensuring the rationality and effectiveness of policy updates. This adaptive mechanism enables the system to maintain good performance under different working conditions and further improves the intelligence level of the system.

[0108] Further, the linkage unit includes:

[0109] A radar micro-motion sensing system that distinguishes between short-term stays and long-term departures of personnel through Doppler feature analysis;

[0110] A thermal inertia prediction module that constructs a thermal radiation attenuation model after the equipment stops based on the material heat conduction equation;

[0111] A power mapping module that calculates the FFU power attenuation curve according to the thermal inertia model to control the temperature and humidity fluctuations within the process threshold.

[0112] Among them, in a chip manufacturing workshop, the accurate control of temperature and humidity is crucial. The chip manufacturing process has almost stringent requirements for environmental cleanliness and stability. Fluctuations in temperature and humidity not only affect the normal operation of production equipment but may also cause irreversible damage to the quality and performance of chips.

[0113] The temperature control in a chip dust-free workshop is usually set within the range of \(22^{\circ}C \pm 2^{\circ}C\). Too high or too low temperature may lead to a decline in equipment performance or even failure. The stable control of temperature helps reduce the thermal stress during chip manufacturing and lower the product defect rate.

[0114] The humidity control in a chip dust-free workshop is usually set within the range of \(45\% \pm 5\%\). A high-humidity environment helps reduce the accumulation of static charges and lower the damage caused by electrostatic discharge to chips.

[0115] Based on the embodiments provided in the present application, the thermal inertia model predicts the residual heat decay curve of the equipment, realizes the progressive adjustment of the FFU power, and avoids the influence of sudden changes in temperature and humidity on product quality.

[0116] Further, the deviation traceability analysis module of the feedback unit includes:

[0117] A causal discovery engine that uses the PC algorithm based on conditional independence to identify the key causal links of environmental control deviations;

[0118] An incremental knowledge distillation component that converts the traceability results into a lightweight knowledge graph and injects it into the prediction model adjustment process.

[0119] Based on the embodiments provided in the present application, incremental knowledge distillation retains the core features of the historical model (such as the environmental response differences between the morning shift and the evening shift), preventing knowledge forgetting during model iteration.

[0120] Such as Figure 3 shown, optionally, the present application provides a method for applying an intelligent constant humidity control FFU integrated system in a dust-free workshop, including:

[0121] S301, Deploy temperature and humidity sensors, production equipment heat load monitoring modules, and meteorological data interfaces inside the dust-free workshop to collect real-time internal and external environmental parameters of the workshop and the operating status of the equipment;

[0122] S302, Integrate historical production data and real-time environmental fluctuation characteristics, and use the trained model with time series prediction ability to output the temperature and humidity change trends and key adjustment time windows of each partition of the workshop in the future time period;

[0123] S303, Based on real-time production process requirements, environmental sensitivity thresholds, and energy consumption constraint conditions, dynamically select the target operation strategy from the process priority mode, energy-saving mode, and emergency mode;

[0124] S304, Use an optimization algorithm to adjust the wind field parameters of the FFU unit under the target operation strategy, including fan speed gradient configuration, air outlet angle collaborative control, and airflow circulation path optimization, to minimize the energy consumption of the intelligent constant humidity control FFU integrated system while maintaining uniform temperature and humidity distribution.

[0125] The intelligent constant humidity control FFU integrated system provided by the present application can be applied to the following engineering projects: dust-free workshops, constant temperature and humidity workshops, high-precision constant temperature and humidity laboratories, medical device dust-free workshops, food dust-free workshops, pharmaceutical dust-free workshops, health product dust-free workshops, CV dust-free workshops, coating dust-free workshops, hundred-thousand-class dust-free workshops, thousand-class dust-free workshops, and hundred-class dust-free workshops.

[0126] In the embodiments of the present application, the system and method of the present invention are applied in a certain dust-free workshop. Equipment such as high-efficiency air outlets, FFUs, air showers, goods showers, weighing rooms, laminar flow hoods, transfer windows, and air shower transfer windows are installed in the workshop. The intelligent constant humidity control FFU integrated system is used to precisely control the temperature and humidity in the workshop. When personnel and materials enter and leave the workshop, the air showers and goods showers can effectively remove the dust and microorganisms attached to the surface, while the weighing room provides a stable environment for precise weighing. The transfer windows and air shower transfer windows avoid the influence of external air on the temperature, humidity, and cleanliness in the workshop during the process of item transfer.

[0127] Based on the embodiments provided in the present application, the intelligent constant humidity control FFU integrated system and its application method in the dust-free workshop can effectively solve the problems existing in the prior art in practical applications, and have significant advantages and broad application prospects.

[0128] It should be noted that in the present application, the embodiments implemented on the side of the intelligent constant humidity control FFU integrated system can be mutually referred to the embodiments implemented on the side of the application method of the intelligent constant humidity control FFU integrated system in the dust-free workshop, and the present application will not elaborate one by one.

[0129] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structural or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. Intelligent constant humidity control FFU integrated system, characterized in that, include: The environmental sensing unit, consisting of temperature and humidity sensors, a production equipment heat load monitoring module, and a meteorological data interface, is used to collect real-time environmental parameters inside and outside the workshop and equipment operating status; The prediction unit is used to train and generate a model with time series prediction capabilities by integrating historical production data with real-time environmental fluctuation characteristics. This model outputs the temperature and humidity change trends and key adjustment time windows for each workshop partition in the future time period. A decision-making unit, which dynamically selects a target operation strategy from process priority mode, energy-saving mode, and emergency mode based on real-time production process requirements, environmental sensitivity thresholds, and energy consumption constraints; The parameter optimization unit is used to adjust the wind field parameters of the FFU unit under the target operation strategy using an optimization algorithm to minimize system energy consumption while maintaining uniform temperature and humidity distribution; the adjustment includes fan speed gradient configuration, coordinated control of air outlet angles, and optimization of airflow circulation paths.

2. The intelligent constant humidity control FFU integrated system according to claim 1, characterized in that, The intelligent constant humidity control FFU integrated system also includes: The linkage unit integrates the UWB personnel positioning system and the infrared heat source detection device. When it detects that there is no personnel activity in the target zone for 10 consecutive minutes and the thermal radiation intensity of the equipment is lower than the threshold, it triggers the FFU power step-by-step attenuation and the temperature and humidity adjustment equipment sleep command in the zone; The feedback unit collects the environmental response data of each partition and the execution status of the equipment, builds a deviation tracing model through the Bayesian network, and feeds back the data to the prediction unit and the parameter optimization unit for iterative model updating.

3. The intelligent constant humidity control FFU integrated system according to claim 1, wherein The production equipment heat load monitoring module in the environmental perception unit is connected to the PLC controller via an industrial bus to analyze the equipment start and stop signals, material processing volume and heat load change curve in real time; the environmental perception unit includes: The redundant verification network uses a hexagonal cellular topology to cover all zones in the workshop, with each zone configured with three sets of cross-verified temperature and humidity sensor nodes; Meteorological data cleaning module, which performs wavelet transform denoising and dynamic time alignment on the temperature, humidity, and air pressure data from external weather stations; The production equipment interface module collects the equipment operating frequency, motor torque and cooling water flow data in the PLC controller in real time through the OPC UA protocol.

4. The intelligent constant humidity control FFU integrated system according to claim 2, wherein The prediction unit includes: The spatiotemporal hypergraph network module models each workshop partition as a hypergraph node, connects partitions with environmental communication relationships through hyperedges, and uses spatiotemporal convolution to extract temperature and humidity diffusion characteristics across partitions; The time window predictor, based on the hidden Markov jump diffusion model, analyzes the time intervals between historical regulation events and the Poisson distribution of production activities, and predicts the triggering probability and duration of future key regulation time windows; The feature fusion module performs tensor fusion on the spatial correlation features output by the hypergraph network and the time window prediction results to generate a partitioned temperature and humidity evolution map with uncertainty intervals.

5. The intelligent constant humidity control FFU integrated system according to claim 4, wherein The prediction unit further includes: The physical constraint network generates physically reasonable training data for extreme operating conditions by integrating the thermodynamic equations of workshop equipment with fluid dynamics boundary conditions, including scenarios such as sudden equipment overload, vent blockage, and sudden moisture absorption by materials. The search module reconstructs the hypergraph network topology based on the real-time collected partition layout change data. Its search strategies include: Dynamically optimize hyperedge connections through a differentiable architecture search algorithm to adapt the hypergraph network to partitioning and reorganization; Introducing topology persistence constraints to ensure smooth transition of network structures between adjacent production batches; A meta-knowledge distillation engine is introduced to encode the workshop environment response rules implicit in the historical prediction model into a lightweight rule set, which serves as the model initialization prior in new scenarios.

6. The intelligent constant humidity control FFU integrated system according to claim 5, wherein, The decision-making unit includes: The attention reinforcement learning framework uses the hypergraph features and time window probabilities output by the prediction unit as state inputs, and extracts environmental sensitivity features at different spatial scales through a hierarchical attention mechanism; Fuzzy differential game controller, builds a game model between the production department and the equipment department, and uses fuzzy differential equations to solve the Pareto optimal strategy; The risk-constrained strategy distillation module extracts time window constraint rules from the event-driven model of the prediction unit and encodes them into action masks for reinforcement learning to limit the selection of high-risk strategies.

7. The intelligent constant humidity control FFU integrated system according to claim 6, wherein The decision-making unit also includes: A heterogeneous strategy fusion module is configured to: extract key propagation path features from the hypergraph network of the prediction unit to construct a probability map for environmental fluctuation propagation; employ a random walk graph attention mechanism to simulate the propagation impact of different decision strategies in the probability map to generate a risk heat map; and dynamically adjust the game model utility function by fusing the risk heat map with real-time production yield data through a dual-stream gating network. The causal reinforcement intervention module executes a closed-loop operation chain: when the feedback unit detects an environmental deviation, it uses a counterfactual causal reasoning algorithm to deduce the optimal set of intervention actions; constructs a virtual twin decision-making environment to preview the long-term impact of the intervention actions in the digital twin; and reshapes the strategy through policy gradients to migrate the optimization strategy in the virtual environment to the physical system.

8. The intelligent constant humidity control FFU integrated system according to claim 2, wherein The linkage unit comprises: The radar micro-motion sensing system uses Doppler feature analysis to distinguish between short-term stays and long-term departures of personnel; Thermal inertia prediction module, which builds a thermal radiation attenuation model after equipment shutdown based on the material heat conduction equation; The power mapping module calculates the FFU power attenuation curve based on the thermal inertia model to control temperature and humidity fluctuations to not exceed the process threshold.

9. The intelligent constant humidity control FFU integrated system according to claim 2, characterized in that, The deviation tracing and analysis module of the feedback unit includes: Causal discovery engine, which uses a PC algorithm based on conditional independence to identify the key causal links of environmental control deviations; The incremental knowledge distillation component converts the traceability results into a lightweight knowledge graph and injects it into the prediction model adjustment process.

10. Application method of intelligent constant humidity control FFU integrated system in dust-free workshop, characterized in that, include: Deploy temperature and humidity sensors, production equipment heat load monitoring modules, and meteorological data interfaces in the clean room to collect real-time environmental parameters inside and outside the workshop and equipment operating status; By integrating historical production data with real-time environmental fluctuation characteristics and using a trained model with time series prediction capabilities, the system outputs the temperature and humidity change trends and key adjustment time windows for each workshop partition in the future time period. Dynamically select the target operation strategy from process priority mode, energy-saving mode and emergency mode based on real-time production process requirements, environmental sensitivity thresholds and energy consumption constraints; The wind field parameters of the FFU unit under the target operation strategy are adjusted by using an optimization algorithm, including the configuration of the fan speed gradient, the collaborative control of the air outlet angle, and the optimization of the air flow circulation path, so as to minimize the energy consumption of the intelligent constant humidity control FFU integrated system while maintaining the uniform distribution of temperature and humidity.

Images