Intelligent clean air conditioner automatic control system and integrated control method

Abstract

The application discloses an intelligent clean air conditioner automatic control system and an integrated control method, and belongs to the technical field of clean environment control. In view of the problems of dispersed control, high energy consumption and insufficient dynamic response of the existing clean air conditioner system, the application integrates intelligent control algorithms and hardware facilities, constructs a multi-parameter collaborative optimization model, and realizes real-time and accurate regulation and control of the temperature, humidity, air flow speed and pressure of a clean workshop. The system comprises a central controller, a sensor network, an actuator and a control software module, dynamically adjusts air valves and variable air volume devices through data fusion and machine learning algorithms, reduces air resistance and turbulence, and optimizes air flow organization efficiency. At the same time, the system supports remote monitoring, fault self-diagnosis and self-adaptive energy-saving mode, and significantly improves the operation stability and energy efficiency ratio. The integrated control method of the application solves the pain points of the insufficient intelligence and high maintenance cost of the traditional system, and has significant energy-saving benefits and environmental adaptability.

Description

Technical Field

[0001] This invention relates to the field of clean environment control technology, and in particular to intelligent clean air conditioning automatic control systems and integrated control methods. Background Technology

[0002] In cleanroom environment control (such as pharmaceuticals, electronics manufacturing, and biological laboratories), air conditioning systems must maintain constant temperature, humidity, pressure differential, and airflow velocity to ensure the cleanliness and stability of the production environment. Traditional cleanroom air conditioning control systems mostly employ classic PID control algorithms, which, while possessing basic adjustment capabilities, face the following technical bottlenecks under complex dynamic conditions:

[0003] 1. The challenge of balancing control precision and energy consumption

[0004] Traditional PID parameters are fixed and cannot be adjusted in real time according to environmental fluctuations, resulting in a sharp increase in energy consumption during high-precision control (such as frequent start-stop of fans), or sacrificing control stability for energy saving (such as the accumulation of temperature and humidity deviations).

[0005] 2. Lack of multi-parameter collaborative optimization

[0006] Existing technologies mostly target the independent control of a single parameter (such as temperature), lacking the integrated analysis and coordinated control of multiple variables such as temperature, humidity, pressure difference, and wind speed, resulting in low airflow organization efficiency (η<0.8).

[0007] 3. Insufficient fault diagnosis and emergency response capabilities

[0008] Traditional systems rely on threshold alarms, which cannot predict potential faults (such as sensor drift or actuator jamming). Furthermore, the delay in switching to safe mode in emergency situations such as sudden pressure drops is relatively high (>1 second), which may cause clean environment failure.

[0009] 4. Limitations of intelligence and adaptive capabilities

[0010] Existing solutions rarely incorporate machine learning and multi-objective optimization algorithms, making it difficult to dynamically optimize PID parameters and energy consumption weights, resulting in a year-on-year decline in the long-term operating energy efficiency ratio (COP).

[0011] As the goals are advanced, cleanroom air conditioning systems need to simultaneously meet the requirements of high-precision control, low-energy operation, and intelligent operation and maintenance. Although some research has attempted to introduce fuzzy control or data fusion technology (such as the Chinese patent "A Multi-parameter Coordinated Control Method for Air Conditioning"), the following shortcomings still exist:

[0012] The fuzzy rule base is fixed and cannot be updated online to adapt to changes in operating conditions;

[0013] The optimization objective is singular, failing to take into account multiple dimensions such as energy consumption, control accuracy, and equipment lifespan.

[0014] The edge computing and cloud collaboration capabilities are weak, and the utilization rate of historical data is low.

[0015] To address the aforementioned issues, this invention proposes an intelligent cleanroom air conditioning automatic control system and integrated control method. Through dynamic data fusion, adaptive adjustment of fuzzy PID parameters, multi-objective optimization model, and edge-cloud collaborative management, the control performance and intelligence level of the cleanroom air conditioning system are improved. Summary of the Invention

[0016] In view of the aforementioned existing problems, the present invention is proposed.

[0017] Therefore, the present invention provides an intelligent clean air conditioning automatic control system and integrated control method that solves the problems of balancing control accuracy and energy consumption, lack of multi-parameter collaborative optimization, and insufficient fault diagnosis and emergency response capabilities.

[0018] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0019] In a first aspect, the present invention provides an intelligent cleanroom air conditioning automatic control system, comprising:

[0020] Central controller, distributed sensor network, actuators and control software modules;

[0021] The central controller adopts a multi-core processor architecture, has a built-in real-time operating system (RTOS), and is configured with communication interfaces including RS-485, CAN bus and Ethernet.

[0022] The distributed sensor network consists of a temperature sensor (accuracy ±0.1℃), a humidity sensor (accuracy ±1%RH), a differential pressure sensor (range 0-500Pa), and a wind speed sensor (range 0-10m / s). It is installed in the four corners and core area of ​​the cleanroom, with a sampling frequency of 10Hz, to collect environmental parameters of the cleanroom in real time.

[0023] The actuators include an electric proportional damper (opening resolution 0.1%), a variable frequency fan (speed range 10%-100%), and a PID-controlled humidifier / dehumidifier (control accuracy ±2%RH), the opening degree or speed of which is adjusted by the control signal output by the central controller.

[0024] The control software module is deployed on the central controller and integrates data fusion, fuzzy PID control and multi-objective optimization algorithms, and supports remote interaction via Modbus-TCP protocol.

[0025] The control software module performs weighted processing on the multi-sensor data using a data fusion algorithm. The weight allocation formula is as follows:

[0026]

[0027] Where, σ i Let be the measurement standard deviation of the i-th type of sensor.

[0028] In a preferred embodiment of the intelligent clean air conditioning automatic control system described in this invention, the central controller employs a fuzzy PID control algorithm, and its output control quantity u(t) is:

[0029]

[0030] Where, e(t) = P target -P real For environmental parameter deviation, P target P is the target environmental parameter for the cleanroom. real K represents the actual environmental parameters of the cleanroom. p ,K i ,K d K is the dynamically adjusted gain coefficient. p For proportional gain, K i (for integral gain), K d For differential gain, K p (t)∈[0.5,2.0],K i (t)∈[0.01,0.1],K d (t)∈[0.05,0.5].

[0031] As a preferred embodiment of the intelligent clean air conditioning automatic control system of the present invention, the objective function of the multi-objective optimization model of the control software module is:

[0032]

[0033] The constraints include: V∈[0.2,0.8]m / s, T∈[20,26];

[0034] In the formula, E base E is the baseline energy consumption (unit: kWh). current P represents the energy consumption for the current cycle (unit: kWh). k,range The baseline environmental parameters for the cleanroom are α = 0.6 and β = 0.4, and the weights are optimized every 30 minutes using the NSGA-II algorithm.

[0035] In a preferred embodiment of the intelligent clean air conditioning automatic control system described in this invention, the relationship between the air volume Q and the opening degree θ of the electric proportional damper is as follows:

[0036]

[0037] in,

[0038] ρ = 1.2 kg / m, opening θ ∈ [0%, 100%], and accuracy is guaranteed by a 12-bit DAC module.

[0039] As a preferred embodiment of the intelligent clean air conditioning integrated control method of the present invention, it includes the following steps:

[0040] S1. Environmental parameters P are collected in real time at a frequency of 10Hz through a distributed sensor network, and noise is eliminated by Kalman filtering;

[0041] S2. Calculate the weighted environmental parameter comprehensive value based on the data fusion algorithm. Weight W i Dynamically updated based on sensor health status;

[0042] S3. Query the fuzzy rule table based on e(t) and γ(t), and output the real-time control parameter K. p ,K i ,K d The fuzzy PID controller generates control signals U = {u1, u2, u3}, which correspond to the valve opening, fan speed and humidification power, respectively. The actuator is dynamically adjusted according to the control signal U.

[0043] S4. Every 30 minutes, call the NSGA-II algorithm to optimize α and β, update the PID parameter thresholds, and update the control parameter K through a multi-objective optimization model. p ,K i ,K d And feedback is sent to step S3;

[0044] S5. Steps S1-S4 are executed sequentially in a loop with a control cycle of 100ms. When an emergency condition is detected (such as a sudden drop in differential pressure exceeding 20%), an interruption mechanism is triggered, skipping S4 and directly using preset safety parameters. Historical data is stored on edge computing nodes, supporting backtracking analysis and model retraining, with a data retention period of 30 days.

[0045] As a preferred embodiment of the intelligent clean air conditioning integrated control method of the present invention, the step of acquiring environmental parameters P in real time at a frequency of 10Hz through a distributed sensor network and eliminating noise through Kalman filtering includes the following steps.

[0046] Temperature T, humidity H, pressure difference ΔP, and wind speed V in the cleanroom are synchronously collected at a frequency of 10Hz through a distributed sensor network. The temperature sensor is a PT100 platinum resistance thermometer, the humidity sensor is a capacitive polymer film sensor, the pressure difference sensor is a microelectromechanical system (MEMS) structure, and the wind speed sensor is a hot wire type.

[0047] The original data is preprocessed using Kalman filtering. The filtering formula is as follows:

[0048]

[0049] In the formula, Let P be the filter environment parameter at time k. k The environmental parameters at time k, Let K be the filter environment parameters at time k-1. k Let H be the Kalman gain, H be the observation matrix, and Q be the noise covariance matrix, where I is the identity matrix.

[0050] The data storage format is 32-bit floating-point numbers, and CRC-16 checksum is used to ensure transmission integrity.

[0051] As a preferred embodiment of the intelligent clean air conditioning integrated control method of the present invention, wherein: the weighted environmental parameter comprehensive value is calculated based on the data fusion algorithm. Weight W i Dynamic updates based on sensor health status include the following steps:

[0052] The weight W is dynamically adjusted based on the sensor's health status. i Health status score S i The calculation formula is

[0053]

[0054] Where R represents the number of recent failures;

[0055] The corrected fusion weight is W i ′=W i ·S i The comprehensive value is calculated as follows:

[0056]

[0057] If a certain sensor S i If the value is less than 0.5, the redundant sensor switching is triggered, and the fusion computing logic is updated.

[0058] As a preferred embodiment of the intelligent clean air conditioning integrated control method of the present invention, wherein: the real-time control parameter K is output by querying the fuzzy rule table according to e(t) and γ(t). p ,K i ,K d A fuzzy PID controller is used to generate control signals U = {u1, u2, u3}, corresponding to the valve opening, fan speed, and humidification power, respectively. The actuator is dynamically adjusted according to the control signal U, including the following steps.

[0059] The fuzzy input variables are environmental parameter deviation e(t) and deviation change rate γ(t), which are divided into 7 fuzzy sets (NB, NM, NS, ZO, PS, PM, PB).

[0060] The fuzzy rule table uses the Mamdani inference mechanism. Example rules are as follows:

[0061] IFe(t)is PB ANDγ(t)is NB THEN K p is PB,K i is NB,K d is PS;

[0062] Defuzzification employs the centroid method, outputting precise PID parameters:

[0063]

[0064] Where μ j Let be the activation degree of the j-th fuzzy rule;

[0065] The calculation period of the control signal U is 100ms, and it is output to the actuator through the DA conversion module.

[0066] As a preferred embodiment of the intelligent clean air conditioning integrated control method of the present invention, wherein: the NSGA-II algorithm is called every 30 minutes to optimize α and β and update the PID parameter thresholds, and the control parameter K is updated through a multi-objective optimization model. p ,K i ,K d And feedback is sent to step S3, including the following steps:

[0067] The NSGA-II algorithm was used to optimize the objective function weights α and β, with a population size of 50, a crossover probability of 0.8, and a mutation probability of 0.1.

[0068] A Pareto front solution set is generated every 30 minutes. The optimal solution considering both energy consumption and parameter deviation is selected, and the PID parameter thresholds are updated accordingly.

[0069]

[0070] in, The threshold values ​​for the control parameters before the update. The updated control parameter threshold is δ, which is the adaptive learning rate coefficient used to control the magnitude of PID parameter updates, and ΔE is the energy consumption optimization rate, representing the energy consumption E in the current cycle. current Compared with the baseline energy consumption E base The relative differences;

[0071] The optimization results are uploaded to the cloud monitoring platform via the OPC UA protocol and a visual report is generated.

[0072] As a preferred embodiment of the intelligent clean air conditioning integrated control method of the present invention, steps S1-S4 are executed sequentially in a cyclical manner with a control cycle of 100ms; when an emergency condition is detected (such as a sudden pressure drop exceeding 20%), an interrupt mechanism is triggered, skipping S4 and directly using preset safety parameters; historical data is stored in edge computing nodes, supporting backtracking analysis and model retraining, with a data retention period of 30 days, including the following steps.

[0073] (a) Periodic control loop execution:

[0074] Steps S1 to S4 are executed sequentially in a loop, with a fixed control cycle of 100ms, triggered by a high-precision hardware timer (model STM32F4xx, clock source error ±0.001%) built into the central controller.

[0075] Within each cycle, data acquisition takes ≤5ms, data fusion and fuzzy PID calculation take ≤80ms, and the remaining time is used for asynchronous iterative calculation of multi-objective optimization.

[0076] (b) Emergency Operational Interruption Detection and Response:

[0077] Real-time monitoring of the differential pressure parameter ΔP; if |ΔP| is satisfied... t -ΔP t-1 ∣ / ΔP t-1 If the differential pressure drops by ≥20% for 3 consecutive sampling cycles (30ms), it is determined to be an emergency condition of sudden pressure drop.

[0078] Trigger the interrupt service routine (ISR), immediately pause the current control cycle, skip optimization step S4, and call the preset safety parameter set U. safe ={θ = 60%, N = 50%, P humid =30%};

[0079] In safe mode, the actuator (103) receives U at a fixed frequency of 10Hz. safe until environmental parameters stabilize (ΔP volatility < 5% for 10 seconds);

[0080] (c) Edge data storage and intelligent management:

[0081] Historical data is stored on edge computing nodes, configured with an ARM Cortex-A72 processor and NVMe SSD. It is encrypted with timestamps (AES-256) and written to the time-series database (InfluxDB). The stored fields include T, H, ΔP, V, U, η, and fault flag bits. T is the real-time temperature data in the cleanroom, in degrees Celsius (°C), collected by a temperature sensor. H is the real-time relative humidity data in the cleanroom, in percentage (%), collected by a humidity sensor. ΔP is the pressure difference data between the cleanroom and the external environment, in Pascals (Pa), measured by a differential pressure sensor to monitor airflow direction and airtightness. V is the airflow velocity data in the cleanroom, in meters per second (m / s), collected by a wind speed sensor to evaluate airflow organization efficiency. U is the set of control signals for the actuators, output by the central controller, used to dynamically adjust the system operating status. η is the airflow uniformity index.

[0082] The data retention strategy is a rolling deletion mechanism, which automatically cleans up records older than 30 days and generates a compressed archive file (ZIP + CRC32 checksum) and uploads it to the cloud;

[0083] It supports model retraining based on the TensorFlow Lite framework. Every 7 days, 10% of the samples (uniformly distributed) are extracted from historical data to update the weights of the LSTM prediction model. The optimized model is then updated to the central controller via OTA.

[0084] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the intelligent clean air conditioning integrated control method as described in the first aspect of the present invention.

[0085] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the intelligent clean air conditioning integrated control method as described in the first aspect of the present invention.

[0086] The beneficial effects of this invention are:

[0087] This invention, based on Kalman filtering and dynamic data fusion, effectively eliminates sensor noise, reducing the overall parameter measurement error to within ±1%. Adaptive fuzzy PID control comprehensively improves control accuracy and dynamic response capabilities. The NSGA-II algorithm dynamically balances energy consumption and accuracy targets, increasing the system's coefficient of performance (COP) by 15%-20% and reducing annual overall energy consumption by approximately 12%-18%. An adaptive energy-saving mode automatically switches to a low-power state (fan speed reduced to 40%) when environmental parameters are stable, improving energy efficiency by 25%. Based on an LSTM prediction model and the 3σ criterion, fault detection accuracy is ≥95%, and the interruption response delay under emergency conditions (such as a sudden 20% pressure drop) is ≤30ms, avoiding the misjudgments and delays (>1 second) of traditional threshold alarms. Historical data is encrypted and stored (AES-256) and supports model retraining (updated every 7 days), reducing maintenance costs by 30% and extending equipment lifespan by 10%-15%. By predicting the airflow uniformity index (η≥0.9) using a BP neural network, the clean area coverage rate is improved by 20%-30% compared to traditional empirical adjustment (η≈0.7). Through coordinated adjustment of parameters such as temperature, humidity, pressure difference, and wind speed, the compliance rate of cleanrooms with the ISO 14644 standard has increased from 80% to over 98%. Remote real-time monitoring is achieved via the Modbus-TCP / OPC UA protocol, and firmware and algorithms can be upgraded online, reducing downtime for maintenance by more than 50%. This invention systematically solves the shortcomings of traditional cleanroom air conditioning systems in terms of control accuracy, energy consumption, reliability, and intelligence level through dynamic data fusion, fuzzy PID adaptive control, multi-objective optimization, and intelligent operation and maintenance mechanisms. Experimental verification shows that it can significantly reduce operation and maintenance costs, improve energy efficiency ratio, and provide a stable and reliable environmental guarantee for high-cleanliness industrial scenarios, demonstrating significant economic benefits and social value. Attached Figure Description

[0088] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0089] Figure 1 This is a schematic diagram of an intelligent cleanroom air conditioning automatic control system in Example 1;

[0090] Figure 2 This is a flowchart of an intelligent clean air conditioning integrated control method in Example 2. Detailed Implementation

[0091] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0092] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0093] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0094] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides an intelligent cleanroom air conditioning automatic control system, comprising:

[0095] Central controller 101, distributed sensor network 102, actuator 103 and control software module 104;

[0096] The central controller 101 adopts a multi-core processor architecture, has a built-in real-time operating system (RTOS), and is equipped with communication interfaces including RS-485, CAN bus and Ethernet;

[0097] The distributed sensor network 102 consists of a temperature sensor (accuracy ±0.1℃), a humidity sensor (accuracy ±1%RH), a differential pressure sensor (range 0-500Pa), and a wind speed sensor (range 0-10m / s). It is installed in the four corners and core area of ​​the cleanroom, with a sampling frequency of 10Hz, to collect environmental parameters of the cleanroom in real time.

[0098] The actuator 103 includes an electric proportional damper A1 with an opening resolution of 0.1%, a variable frequency fan A2 with a speed range of 10%-100%, and a PID-controlled humidifier / dehumidifier A3 with a control accuracy of ±2%RH. Its opening or speed is adjusted by the control signal output by the central controller 101.

[0099] The control software module 104 is deployed on the central controller 101, integrating data fusion, fuzzy PID control and multi-objective optimization algorithms, and supporting remote interaction via Modbus-TCP protocol;

[0100] The control software module 104 performs weighted processing on the multi-sensor data using a data fusion algorithm. The weight allocation formula is as follows:

[0101]

[0102] Where, σ iLet be the measurement standard deviation of the i-th type of sensor.

[0103] The central controller 101 adopts a fuzzy PID control algorithm, and its output control quantity u(t) is:

[0104]

[0105] Where, e(t) = P target -P real For environmental parameter deviation, P target P is the target environmental parameter for the cleanroom. real K represents the actual environmental parameters of the cleanroom. p ,K i ,K d K is the dynamically adjusted gain coefficient. p For proportional gain, K i (for integral gain), K d For differential gain, K p (t)∈[0.5,2.0],K i (t)∈[0.01,0.1],K d (t)∈[0.05,0.5].

[0106] The objective function of the multi-objective optimization model of control software module 104 is:

[0107]

[0108] The constraints include: V∈[0.2,0.8]m / s, T∈[20,26];

[0109] In the formula, E base E is the baseline energy consumption (unit: kWh). current P represents the energy consumption for the current cycle (unit: kWh). k,range The baseline environmental parameters for the cleanroom are α = 0.6 and β = 0.4, and the weights are optimized every 30 minutes using the NSGA-II algorithm.

[0110] The relationship between the air volume Q and the opening degree θ of the electric proportional damper A1 is as follows:

[0111]

[0112] in,

[0113] ρ = 1.2 kg / m, opening θ ∈ [0%, 100%], and accuracy is guaranteed by a 12-bit DAC module.

[0114] Example 2, refer to Figure 2 This is the second embodiment of the present invention, which provides an intelligent

[0115] clean

[0116] The integrated control method for clean air conditioning includes the following steps:

[0117] S1. Environmental parameters P are collected in real time at a frequency of 10Hz through a distributed sensor network 102, and noise is eliminated by Kalman filtering;

[0118] Data acquisition process:

[0119] Temperature T, humidity H, pressure difference ΔP and wind speed V in the cleanroom are synchronously collected at a frequency of 10Hz through a distributed sensor network 102. The temperature sensor is a PT100 platinum resistance thermometer, the humidity sensor is a capacitive polymer film sensor, the pressure difference sensor is a microelectromechanical system (MEMS) structure, and the wind speed sensor is a hot wire type.

[0120] Sensor selection:

[0121] Temperature sensor: PT100 platinum resistance thermometer (model: Omega PT100-1K), accuracy ±0.1℃, range -20℃~80℃ (model: Omega PT100-1K), installed in the four corners of the cleanroom and around the core equipment (8 nodes per 100㎡), connected to the central controller via shielded cable;

[0122] Humidity sensor: Honeywell HIH-4000 capacitive sensor is selected, with an accuracy of ±1% and a range of 0-100%. It is installed in the air conditioner return air vent and supply air duct (2 nodes per duct) to avoid direct contact with condensate.

[0123] Differential pressure sensor: Sensirion SDP800 MEMS sensor, accuracy ±1Pa, range 0-500Pa, installed on both sides of the partition wall between the clean room and the buffer room (1 pair per partition wall), data is transmitted via RS-485 bus.

[0124] Wind speed sensor: TSI 8475 hot-wire anemometer, resolution 0.01m / s, range 0-10m / s, installed 0.5 meters below the high-efficiency air outlet (1 per outlet), sampling frequency 50Hz;

[0125] Synchronization mechanism:

[0126] The distributed sensor network synchronously collects data at a frequency of 10Hz via the LoRa wireless communication protocol.

[0127] (middle

[0128] The sensor data is transmitted from a 433MHz clock frequency and a 125kHz bandwidth to the central controller (STM32F407VG). The central controller triggers 10Hz synchronous acquisition through a hardware timer (TIM2, 72MHz clock). Each sensor node uses a GPS module (ublox NEO-M8N) to achieve microsecond-level time synchronization.

[0129] Preprocessing method: Kalman filtering is used to eliminate noise. The filtering parameters are set as follows: process noise covariance Q = 0.01I and observation noise covariance R = 0.05I. The output smoothed environmental parameter set is P = {T, H, ΔP, V}, where T is the real-time temperature in the clean room in degrees Celsius (°C), H is the real-time relative humidity in the clean room in percentage (%), ΔP is the pressure difference between the clean room and the adjacent area in Pascals (Pa), and V is the airflow velocity in the clean room in meters per second (m / s).

[0130] Kalman filter implementation:

[0131] The original data is preprocessed using Kalman filtering. The filtering formula is as follows:

[0132]

[0133] In the formula, Let P be the filter environment parameter at time k. k The environmental parameters at time k, Let K be the filter environment parameters at time k-1. k Let H be the Kalman gain, H be the observation matrix, and Q be the noise covariance matrix, where I is the identity matrix.

[0134] Data integrity verification:

[0135] The data storage format is 32-bit floating-point numbers, and CRC-16 checksum is used to ensure transmission integrity. If the checksum fails, a retransmission mechanism is triggered (up to 3 times) to ensure that the packet loss rate is ≤0.5%.

[0136] S2. Calculate the weighted environmental parameter comprehensive value based on the data fusion algorithm. Weight W i Dynamically updated based on sensor health status;

[0137] Dynamic weight allocation: initial weights are

[0138] Where σ T =0.05,σ H =0.5,σ ΔP =2,σ V =0.1;

[0139] Based on sensor health status score Where R represents the number of recent failures; the corrected fusion weight is W. i ′=W i ·S i .

[0140] Redundancy switching mechanism:

[0141] If a certain sensor S i If the value is less than 0.5, redundant sensor switching is triggered, automatically activating the backup sensor (such as a dual redundant sensor group installed at the four corners), and the fused composite value is recalculated.

[0142] Verification and Calibration:

[0143] Offline calibration: Sensors are calibrated on-site weekly using a standard instrument (Fluke 754 process calibrator) to correct deviations exceeding ±2%.

[0144] Online verification: Compare fusion values ​​P every 1 hour fused If the deviation from the manually measured value is greater than 3%, an alarm will be triggered and the data will be logged.

[0145] S3. Query the fuzzy rule table based on e(t) and γ(t), and output the real-time control parameter K. p ,K i ,K d The fuzzy PID controller generates control signals U = {u1, u2, u3}, which correspond to the valve opening, fan speed and humidification power, respectively. The actuator is dynamically adjusted according to the control signal U.

[0146] Fuzzy input variable: Environmental parameter deviation e(t) = P target -P real And the rate of change of deviation γ(t) is divided into 7 fuzzy sets (NB, NM, NS, ZO, PS, PM, PB);

[0147] Output variable: PID gain coefficient K p ,K i ,K d The ranges are K p ∈[0.5,2.0]、K i ∈[0.01,0.1]、K d ∈[0.05,0.5];

[0148] The fuzzy rule table uses the Mamdani inference mechanism. Example rules are as follows:

[0149] IFe(t)is PB ANDγ(t)is NB THEN K pis PB,K i is NB,K d is PS;

[0150] Defuzzification employs the centroid method, outputting precise PID parameters:

[0151]

[0152] Where μ j The activation degree of the j-th fuzzy rule is determined by the minimum value of the input membership degree;

[0153] The calculation cycle of the control signal U is 100ms. The control signal U drives the electric air valve (opening resolution 0.1%), the variable frequency fan (speed range 10%-100%) and the humidifier (control accuracy ±2%RH) respectively, and is output to the actuator (103) through the DA conversion module.

[0154] Control signal U = {u1, u2, u3} is generated:

[0155] Electric air valve opening: Output resolution: 0.1%.

[0156] Variable frequency fan speed: Where D = 0.5m, η = 0.75, speed regulation error < ±1%, Q target The target air volume for the electric proportional damper.

[0157] Humidification power: u3 = sat(K) p ·(H target -H real The saturation function limits the output range (0%, 100%).

[0158] S4. Every 30 minutes, call the NSGA-II algorithm to optimize α and β, update the PID parameter thresholds, and update the control parameter K through a multi-objective optimization model. p ,K i ,K d And feedback is sent to step S3;

[0159] Multi-objective optimization model update control parameter optimization process:

[0160] Objective function:

[0161]

[0162] The constraints include: V∈[0.2,0.8]m / s, T∈[20,26];

[0163] In the formula, E base E is the baseline energy consumption (unit: kWh). currentP represents the energy consumption for the current cycle (unit: kWh). k,range The baseline environmental parameters for the cleanroom are α = 0.6 and β = 0.4, and the weights are optimized every 30 minutes using the NSGA-II algorithm.

[0164] Algorithm implementation:

[0165] The NSGA-II algorithm is used to optimize the objective function weights α and β, where α∈[0,1], β∈[0,1], the population size is 50, the crossover probability is 0.8, and the mutation probability is 0.1.

[0166] Every 30 minutes, the Pareto front solution set is generated by calculating the distance between non-dominated ranking and crowding degree. The optimal solution combining energy consumption and parameter bias is selected, and the top 10% of individuals are chosen as the optimal solution set. The PID parameter thresholds are then updated.

[0167]

[0168] in, The threshold values ​​for the control parameters before the update. The updated control parameter threshold is δ, which is the adaptive learning rate coefficient used to control the magnitude of PID parameter updates, and ΔE is the energy consumption optimization rate, representing the energy consumption E in the current cycle. current Compared with the baseline energy consumption E base The relative differences

[0169]

[0170] The optimization results are uploaded to the cloud monitoring platform via the OPC UA protocol and a visual report is generated.

[0171] S5. Steps S1-S4 are executed sequentially in a loop with a control cycle of 100ms. When an emergency condition is detected (such as a sudden drop in differential pressure exceeding 20%), an interruption mechanism is triggered, skipping S4 and directly using preset safety parameters. Historical data is stored in edge computing node 105, supporting backtracking analysis and model retraining, with a data retention period of 30 days.

[0172] (a) Periodic control loop execution:

[0173] Steps S1 to S4 are executed sequentially in a loop. The control cycle is fixed at 100ms and is triggered by the high-precision hardware timer (model STM32F407VG, clock source error ±0.001%) built into the central controller 101. The interrupt priority is configured to the highest level.

[0174] Within each cycle, data acquisition takes ≤5ms, data fusion and fuzzy PID calculation take ≤80ms, and the remaining time is used for asynchronous iterative calculation of multi-objective optimization.

[0175] (b) Emergency Operational Interruption Detection and Response:

[0176] Real-time monitoring of the differential pressure parameter ΔP; if |ΔP| is satisfied... t -ΔP t-1 ∣ / ΔP t-1 If the differential pressure drops by ≥20% for 3 consecutive sampling cycles (30ms), it is determined to be an emergency condition of sudden pressure drop.

[0177] Trigger the interrupt service routine (ISR), immediately pause the current control cycle, skip optimization step S4, and call the preset safety parameter set U. safe ={θ = 60%, N = 50%, P humid =30%}; stored in EEPROM (model: AT24C256) to ensure that parameters are not lost after power failure;

[0178] In safe mode, actuator 103 receives U at a fixed frequency of 10Hz. safe until environmental parameters stabilize (ΔP volatility < 5% for 10 seconds);

[0179] (c) Edge data storage and intelligent management:

[0180] Edge storage configuration:

[0181] Hardware: NVIDIA Jetson Nano (4GB RAM) + 1TB NVMe SSD (Samsung 970EVO).

[0182] Software: Data is stored in InfluxDB time-series structure, encrypted using AES-256-CBC mode, and the key is rotated every 24 hours.

[0183] Historical data is stored on edge computing node 105, encrypted with timestamps (AES-256), and then written to the time-series database (InfluxDB). The stored fields include T, H, ΔP, V, U, η, and fault flag bits. T is the real-time temperature data in the cleanroom, in degrees Celsius (°C), collected by a temperature sensor. H is the real-time relative humidity data in the cleanroom, in percentage (%), collected by a humidity sensor. ΔP is the pressure difference data between the cleanroom and the external environment, in Pascals (Pa), measured by a differential pressure sensor to monitor airflow direction and airtightness. V is the airflow velocity data in the cleanroom, in meters per second (m / s), collected by a wind speed sensor to evaluate airflow organization efficiency. U is the set of control signals for the actuators, output by the central controller, used to dynamically adjust the system operating status. η is the airflow uniformity index.

[0184] The data retention strategy is a rolling deletion mechanism, which automatically cleans up records older than 30 days and generates a compressed archive file (ZIP + CRC32 checksum) and uploads it to the cloud;

[0185] Model retraining:

[0186] Every 7 days, 10% of the samples (approximately 300,000 records) are randomly selected from historical data. An LSTM prediction model (128 hidden nodes) is trained using the TensorFlow Lite framework. Once the training error is less than 2%, the model is pushed to the central controller via OTA.

[0187] This embodiment also provides a computer device applicable to an integrated control method for intelligent cleanroom air conditioning, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the integrated control method for intelligent cleanroom air conditioning as proposed in the above embodiment.

[0188] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0189] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the intelligent clean air conditioning integrated control method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0190] In summary, this invention, based on Kalman filtering and dynamic data fusion, effectively eliminates sensor noise, reducing the overall parameter measurement error to within ±1%. Adaptive fuzzy PID control comprehensively improves control accuracy and dynamic response capabilities. The NSGA-II algorithm dynamically balances energy consumption and accuracy targets, increasing the system's coefficient of performance (COP) by 15%-20% and reducing annual overall energy consumption by approximately 12%-18%. An adaptive energy-saving mode automatically switches to a low-power state (fan speed reduced to 40%) when environmental parameters are stable, improving energy efficiency by 25%. Based on an LSTM prediction model and the 3σ criterion, fault detection accuracy is ≥95%, and the interruption response delay under emergency conditions (such as a sudden 20% pressure drop) is ≤30ms, avoiding the misjudgments and delays (>1 second) of traditional threshold alarms. Historical data is encrypted and stored (AES-256) and supports model retraining (updated every 7 days), reducing maintenance costs by 30% and extending equipment lifespan by 10%-15%. By predicting the airflow uniformity index (η≥0.9) using a BP neural network, the clean area coverage rate is improved by 20%-30% compared to traditional empirical adjustment (η≈0.7). Through coordinated adjustment of parameters such as temperature, humidity, pressure difference, and wind speed, the compliance rate of cleanrooms with the ISO 14644 standard has increased from 80% to over 98%. Remote real-time monitoring is achieved via the Modbus-TCP / OPC UA protocol, and firmware and algorithms can be upgraded online, reducing downtime for maintenance by more than 50%. This invention systematically solves the shortcomings of traditional cleanroom air conditioning systems in terms of control accuracy, energy consumption, reliability, and intelligence level through dynamic data fusion, fuzzy PID adaptive control, multi-objective optimization, and intelligent operation and maintenance mechanisms. Experimental verification shows that it can significantly reduce operation and maintenance costs, improve energy efficiency ratio, and provide a stable and reliable environmental guarantee for high-cleanliness industrial scenarios, demonstrating significant economic benefits and social value.

[0191] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An intelligent cleanroom air conditioning automatic control system, characterized in that, include: Central controller (101), distributed sensor network (102), actuator (103) and control software module (104); The central controller (101) adopts a multi-core processor architecture, has a built-in real-time operating system (RTOS), and is configured with communication interfaces including RS-485, CAN bus and Ethernet; The distributed sensor network (102) consists of a temperature sensor (accuracy ±0.1℃), a humidity sensor (accuracy ±1%RH), a differential pressure sensor (range 0-500Pa), and a wind speed sensor (range 0-10m / s). It is installed in the four corners and core area of ​​the cleanroom, with a sampling frequency of 10Hz, to collect environmental parameters of the cleanroom in real time. The actuator (103) includes an electric proportional damper (A1, opening resolution 0.1%), a variable frequency fan (A2, speed range 10%-100%), and a PID-controlled humidifier / dehumidifier (A3, control accuracy ±2%RH), the opening degree or speed of which is adjusted by the control signal output by the central controller (101). The control software module (104) is deployed on the central controller (101), integrating data fusion, fuzzy PID control and multi-objective optimization algorithms, and supporting remote interaction via Modbus-TCP protocol; The control software module (104) performs weighted processing on the multi-sensor data through a data fusion algorithm, and the weight allocation formula is as follows: in, Let be the measurement standard deviation of the i-th type of sensor.

2. The intelligent cleanroom air conditioning automatic control system as described in claim 1, characterized in that, The central controller (101) adopts a fuzzy PID control algorithm, and its output control quantity u(t) is: Where e(t) = P target -P real For environmental parameter deviation, P target P is the target environmental parameter for the cleanroom. real These are the actual environmental parameters of the cleanroom. , , The gain coefficient is dynamically adjusted. For proportional gain, (for integral gain) For differential gain, ∈[0.5,2.0], ∈[0.01,0.1], ∈[0.05,0.5].

3. The intelligent cleanroom air conditioning automatic control system as described in claim 1, characterized in that, The objective function of the multi-objective optimization model of the control software module (104) is: The constraints include: V∈[0.2,0.8]m / s, T∈[20,26]; In the formula, E base Baseline energy consumption (unit: kWh) Energy consumption for the current cycle (unit: kWh) The baseline environmental parameters for the cleanroom are α=0.6, β=0.4, and the weights are optimized every 30 minutes using the NSGA-II algorithm.

4. The intelligent cleanroom air conditioning automatic control system as described in claim 1, characterized in that, The relationship between the air volume Q and the opening degree θ of the electric proportional damper (A1) is as follows: in, ; ρ=1.2kg / m, opening θ∈[0%,100%], accuracy is guaranteed by a 12-bit DAC module.

5. An integrated control method for intelligent cleanroom air conditioning, which is based on an intelligent cleanroom air conditioning automatic control system as described in any one of claims 1 to 4, characterized in that, Includes the following steps, S1. Environmental parameters P are collected in real time at a frequency of 10Hz through a distributed sensor network (102), and noise is eliminated by Kalman filtering; S2. Calculate the weighted environmental parameter comprehensive value based on the data fusion algorithm. Weight W i Dynamically updated based on sensor health status; S3. Based on e(t) and Query the fuzzy rule table and output the real-time control parameter K. p ,K i ,K d The fuzzy PID controller generates control signals U={u1,u2,u3}, which correspond to the valve opening, fan speed and humidification power, respectively. The actuator is dynamically adjusted according to the control signal U. S4. Every 30 minutes, call the NSGA-II algorithm to optimize α and β, update the PID parameter thresholds, and update the control parameter K through a multi-objective optimization model. p ,K i ,K d And feedback is sent to step S3; S5. Steps S1-S4 are executed sequentially in a loop with a control cycle of 100ms. When an emergency condition is detected (such as a sudden drop in differential pressure exceeding 20%), an interruption mechanism is triggered, skipping S4 and directly using the preset safety parameters. Historical data is stored in the edge computing node (105), supporting backtracking analysis and model retraining, with a data retention period of 30 days.

6. The intelligent clean air conditioning integrated control method as described in claim 5, characterized in that, The process of acquiring environmental parameters P in real time at a frequency of 10Hz through a distributed sensor network (102) and eliminating noise through Kalman filtering includes the following steps: Temperature T, humidity H, pressure difference ΔP and wind speed V in the cleanroom are synchronously collected at a frequency of 10Hz through a distributed sensor network (102). The temperature sensor is a PT100 platinum resistance thermometer, the humidity sensor is a capacitive polymer film sensor, the pressure difference sensor is a microelectromechanical system (MEMS) structure, and the wind speed sensor is a hot wire type. The original data is preprocessed using a Kalman filter. The filtering formula is as follows: In the formula, Let k be the filtering environment parameters at time k. The environmental parameters at time k, Let K be the filter environment parameters at time k-1. k Let H be the Kalman gain, H be the observation matrix, and Q be the noise covariance matrix, where I is the identity matrix. The data storage format is 32-bit floating-point numbers, and CRC-16 checksum is used to ensure transmission integrity.

7. The intelligent clean air conditioning integrated control method as described in claim 5, characterized in that, The weighted environmental parameter comprehensive value is calculated based on the data fusion algorithm. Weight W i Dynamic updates based on sensor health status include the following steps: The weight W is dynamically adjusted based on the sensor's health status. i Health status score S i The calculation formula is Where R represents the number of recent failures; The corrected fusion weight is W i ′=W i S i The comprehensive value is calculated as follows: If a certain sensor If the value is less than 0.5, the redundant sensor switching is triggered, and the fusion computing logic is updated.

8. The intelligent clean air conditioning integrated control method as described in claim 5, characterized in that, The statement based on e(t) and Query the fuzzy rule table and output the real-time control parameter K. p ,K i ,K d A fuzzy PID controller is used to generate control signals U={u1,u2,u3}, corresponding to the valve opening, fan speed, and humidification power, respectively. The actuator is then dynamically adjusted based on the control signal U, including the following steps. Fuzzy input variables: environmental parameter deviation e(t) and rate of change of deviation Divided into 7 fuzzy Sets (NB, NM, NS, ZO, PS, PM, PB); The fuzzy rule table uses the Mamdani inference mechanism. Example rules are as follows: IF e(t) is PB AND is NB THEN K p is PB, K i is NB, K d is PS; Defuzzification uses the centroid method to output accurate PID parameters: in Let be the activation degree of the j-th fuzzy rule; The calculation period of the control signal U is 100ms, and it is output to the actuator (103) through the DA conversion module.

9. The intelligent clean air conditioning integrated control method as described in claim 5, characterized in that, The NSGA-II algorithm is invoked every 30 minutes to optimize α and β, and the PID parameter thresholds are updated. The control parameter K is then updated through a multi-objective optimization model. p ,K i ,K d And feedback is sent to step S3, including the following steps: The NSGA-II algorithm is used to optimize the objective function weights α and β, with a population size of 50 and a crossover probability of 0.

8. Mutation probability 0.1; A Pareto front solution set is generated every 30 minutes. The optimal solution considering both energy consumption and parameter deviation is selected, and the PID is updated accordingly. Parameter threshold: in, The threshold values ​​for the control parameters before the update. For the updated control parameter thresholds, The adaptive learning rate coefficient is used to control the magnitude of PID parameter updates. Energy optimization rate, representing the energy consumption E in the current cycle. current Compared with the baseline energy consumption E base The relative differences; The optimization results are uploaded to the cloud monitoring platform via the OPC UA protocol and a visual report is generated.

10. The intelligent cleanroom air conditioning integrated control method as described in claim 5, characterized in that, Steps S1-S4 are executed sequentially in a loop with a control cycle of 100ms. When an emergency condition is detected (such as a sudden drop in differential pressure exceeding 20%), an interrupt mechanism is triggered, skipping S4 and directly using preset safety parameters. Historical data is stored on the edge computing node (105), supporting retrospective analysis and model retraining. The data retention period is 30 days, including the following steps. (a) Periodic control loop execution: Steps S1 to S4 are executed sequentially in a loop, with a fixed control cycle of 100ms, triggered by a high-precision hardware timer (model STM32F4xx, clock source error ±0.001%) built into the central controller (101). Within each cycle, data acquisition time is ≤5ms, and data fusion and fuzzy PID calculation time is ≤80ms. The remaining time is used for asynchronous iterative computation of multi-objective optimization; (b) Emergency Operational Interruption Detection and Response: Real-time monitoring of differential pressure parameter ΔP, if it satisfies |Δ -Δ ∣ / Δ ≥20% for 3 consecutive months If the sampling period is 30ms, it is determined to be an emergency situation of sudden pressure drop; Trigger the interrupt service routine (ISR), immediately pause the current control cycle, skip optimization step S4, and call the preset safety parameter set U. safe ={θ=60%,N=50%,P humid =30%}; In safe mode, the actuator (103) receives U at a fixed frequency of 10Hz. safe Continue until environmental parameters stabilize (ΔP volatility < 5% for 10 seconds). (c) Edge data storage and intelligent management: Historical data is stored on an edge computing node (105), configured with an ARM Cortex-A72 processor and NVMe SSD. It is encrypted with timestamps (AES-256) and written to a time-series database (InfluxDB). The stored fields include T, H, ΔP, V, U, η and a fault flag bit. T is the real-time temperature data in the cleanroom, in degrees Celsius (°C), collected by a temperature sensor. H is the real-time relative humidity data in the cleanroom, in percentage (%), collected by a humidity sensor (S2). ΔP is the pressure difference data between the cleanroom and the external environment, in Pascals (Pa), measured by a pressure difference sensor, used to monitor the airflow direction and airtightness. V is the airflow velocity data in the cleanroom, in meters per second (m / s), collected by a wind speed sensor, used to evaluate the airflow organization efficiency. U is the control signal set of the actuator, output by the central controller (101), used to dynamically adjust the system operating status. η is the airflow uniformity index. The data retention strategy is a rolling deletion mechanism, which automatically cleans up records older than 30 days and generates a compressed archive file (ZIP + CRC32 checksum) and uploads it to the cloud; It supports model retraining based on the TensorFlow Lite framework. Every 7 days, 10% of the samples (uniformly distributed) are extracted from the historical data to update the weights of the LSTM prediction model. The optimized model is then updated to the central controller (101) via OTA.

Images