Method, device and equipment for controlling clean room to save energy by deploying AI algorithm

Abstract

The invention relates to a method, device and equipment for controlling energy conservation of a clean room with an AI algorithm, and the method comprises the steps: collecting real-time environment parameters and equipment power consumption data through a temperature and humidity sensor, a cleanliness monitor, a dust falling particle catcher and an intelligent electric meter which are disposed in the clean room; associating the real-time environment parameters and the equipment electricity consumption data with production scheduling information to generate a dynamic regulation and control strategy; based on the real-time environment parameters, the equipment electricity consumption data, the dynamic regulation and control strategy and historical energy consumption data, a start-stop strategy of the fan filter unit is calculated through a deep learning model, and the start-stop strategy comprises a three-gear hierarchical regulation and control strategy for an alternating current type FFU. And the voltage and the rotating speed of the fan FFU are dynamically adjusted according to the direct current type FFU, so that intelligent linkage of production requirements and environment control is achieved, and the purposes of reducing cost and increasing efficiency are achieved.

Description

Technical Field

[0001] The present invention relates to the field of industrial automation and intelligent control technology, and in particular to a method, device and equipment for controlling clean room energy saving with an AI algorithm deployed. Background Art

[0002] Cleanrooms are the core infrastructure for high-precision manufacturing in the fields of electronics, medicine, biotechnology, etc., and have strict requirements on the temperature, humidity, cleanliness and air quality of the environment. Traditional cleanroom control systems usually use fixed rules or manual intervention to adjust the operation of fans, filters and fresh air systems, which often have technical problems such as high energy consumption, low control accuracy and high maintenance costs.

[0003] In order to solve the above problems, some studies in recent years have tried to introduce the Internet of Things and artificial intelligence technologies to optimize clean room control strategies through data collection and analysis. However, the existing technologies still have limitations such as low data utilization and single control target. Summary of the invention

[0004] The main purpose of the present invention is to provide a method, device and equipment for controlling clean room energy saving deployed with AI algorithm. By combining real-time environmental parameters, equipment power consumption data and production scheduling information, a deep learning model is used to achieve multi-objective optimization control, significantly reduce energy consumption and improve control accuracy, so as to realize the intelligent linkage between production demand and environmental control and achieve the purpose of reducing costs and increasing efficiency.

[0005] To achieve the above object, the present invention provides a method for controlling clean room energy saving using an AI algorithm, comprising the following steps: The real-time environmental parameters and equipment power consumption data are collected by deploying temperature and humidity sensors, cleanliness monitors, dust particle traps and smart meters in the clean room; the real-time environmental parameters include temperature, humidity, particle concentration and cleanliness data; Associating the real-time environmental parameters, the equipment power consumption data and the production scheduling information to generate a dynamic control strategy, wherein the dynamic control strategy includes the production demand priority, cleanliness control threshold and energy consumption constraint conditions for the future period; Based on the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data, the start-stop strategy of the fan filter unit is calculated through a deep learning model. The start-stop strategy includes a three-speed graded control strategy for the AC FFU and a dynamic adjustment of the voltage and speed of the fan FFU for the DC FFU. The historical energy consumption data includes historical power consumption records of the fan filter unit at different operating frequencies, gears, voltages and speeds. The deep learning model takes minimizing energy consumption and maintaining cleanliness standards as optimization goals, and drives the controller to perform differentiated control.

[0006] Furthermore, the step of collecting equipment power consumption data includes: Real-time monitoring and uploading of power consumption data of the equipment, including power consumption of the fan filter unit in different operating modes, dynamic energy consumption when the fan filter unit gear is adjusted, and additional energy consumption of the filter when maintaining the cleanliness level.

[0007] Furthermore, the step of associating the real-time environmental parameters, the equipment power consumption data and the production scheduling information to generate a dynamic control strategy includes: According to the planned capacity of the production scheduling information, the production plan is divided into a high capacity period and a low capacity period, wherein the high capacity period corresponds to a full production or overload production plan, and the low capacity period corresponds to a half production or shutdown plan; Dynamically adjust the cleanliness control thresholds at different time periods based on the collected real-time cleanliness data and ISO 14644-1 standard; Combined with historical equipment power consumption data, set energy consumption constraints for each period, including allowing energy consumption peaks to not exceed the maximum power consumption under the same historical operating conditions during high-capacity periods, and setting energy consumption caps at 75% of the average power consumption under the same historical operating conditions during low-capacity periods; Generate production demand priorities based on the operating hours and real-time equipment operating status of production scheduling information, including giving priority to ensuring the operation and cleanliness of key fans FFU during the operating hours, and shutting down equipment in non-production areas and reducing the fresh air volume to the minimum level to maintain basic cleanliness during non-operating hours; The cleanliness control threshold, the energy consumption constraint and the production demand priority are integrated to generate a dynamic control strategy.

[0008] Furthermore, the deep learning model is a multi-objective optimization model, including an LSTM network module and a reinforcement learning module; wherein, the LSTM network module generates energy consumption demand and cleanliness change trend for future time periods through time series prediction based on historical environmental parameters, equipment operation data and production scheduling data; the reinforcement learning module uses the cleanliness threshold and energy consumption constraint in the dynamic control strategy as the reward function, and generates real-time control instructions in combination with the prediction results, including fan filter unit opening adjustment value, fan start and stop instructions and equipment energy consumption allocation strategy; the training of the model is achieved through historical data pre-training and real-time data online fine-tuning.

[0009] Furthermore, the step of calculating the start and stop strategy of the fan filter unit based on the deep learning model according to the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data includes: Inputting the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data into a deep learning model; Through the LSTM network module in the deep learning model, based on the input data, the cleanliness change trend, temperature and humidity fluctuation range and energy consumption demand in the future set period are predicted; The reinforcement learning module in the deep learning model is used to calculate the minimum fresh air volume and the start-stop strategy of the fan filter unit that meet the predicted cleanliness requirements according to the dynamic control strategy, wherein the start-stop strategy includes the fan number to be started, the operating frequency, the operating time, and the list of non-production area equipment to be shut down.

[0010] Furthermore, the step of performing the corresponding adjustment operation includes: Based on the start-stop strategy of the fan filter unit, real-time control instructions are generated and executed. The real-time control instructions include hierarchical control instructions for the AC FFU, voltage and speed control instructions for the DC FFU, and physical device drive instructions. The hierarchical control instructions include 1st gear 70% rated power low-speed operation, 2nd gear 80% power balancing mode, and 3rd gear full power full-load operation.

[0011] Furthermore, after the step of performing the corresponding adjustment operation, the method further includes: Collect the actual cleanliness data after adjustment, the power consumption of the corresponding equipment and the fan status to verify whether the actual energy consumption and cleanliness meet the preset targets. If not, trigger the closed-loop feedback optimization, adjust the model parameters and regenerate the control instructions.

[0012] The present invention also provides a clean room energy-saving control device deployed with an AI algorithm, comprising: The data acquisition module is used to collect real-time environmental parameters and equipment power consumption data through temperature and humidity sensors, cleanliness monitors, dust particle catchers, and smart meters deployed in the clean room; An information association module, used to associate the real-time environmental parameters, the equipment power consumption data and the production scheduling information to generate a dynamic control strategy; The strategy generation module is used to calculate the start and stop strategy of the fan filter unit through a deep learning model based on the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data, and perform corresponding adjustment operations.

[0013] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned method for controlling energy saving in a clean room deployed with an AI algorithm are implemented.

[0014] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned method for controlling energy saving of a clean room deployed with an AI algorithm are implemented.

[0015] The energy-saving method, device and equipment for controlling clean rooms with AI algorithms provided by the present invention have the following beneficial effects: the present invention flexibly adjusts the control strategy according to the production demand and actual environmental conditions in different time periods, achieves a balance between production and energy saving, avoids energy waste, and ensures that the cleanliness of the production environment meets the requirements, and through the deep learning model, it takes minimizing energy consumption and maintaining cleanliness as the optimization goal, and can intelligently and dynamically adjust the fan operating frequency and the opening of the fresh air valve according to the actual situation to achieve on-demand air supply. Compared with the traditional fixed mode control, this intelligent control method can significantly improve energy utilization efficiency and reduce energy consumption costs. At the same time, the present invention also drives the controller to perform differentiated control, avoids unnecessary energy consumption, and effectively reduces the energy consumption of the clean room while ensuring that the cleanliness of the clean room meets the standards, has significant energy-saving benefits, and is in line with the concept of green production and sustainable development. Reasonable fan start-stop strategy and operating frequency adjustment also avoid frequent start-stop and excessive operation of equipment, which helps to extend the service life of equipment such as fans and reduce equipment maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of a method for controlling clean room energy saving with an AI algorithm deployed in one embodiment of the present invention; Figure 2 is a structural block diagram of a clean room energy-saving device deployed with an AI algorithm in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0017] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] Reference Figure 1 , which is a flow chart of a method for controlling clean room energy saving with an AI algorithm proposed by the present invention, comprising the following steps: S1, collects real-time environmental parameters and equipment power consumption data through temperature and humidity sensors, cleanliness monitors, dust particle collectors and smart meters deployed in the clean room; S2, associating the real-time environmental parameters, the equipment power consumption data and the production scheduling information to generate a dynamic control strategy; S3, based on the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data, the start and stop strategy of the fan filter unit is calculated through the deep learning model, and the corresponding adjustment operation is performed.

[0020] As described in step S1 above, the real-time temperature and humidity fluctuations in the clean room are monitored by the temperature and humidity sensors deployed in the clean room to ensure compliance with the production process requirements; the cleanliness level is monitored in real time by a laser particle counter (such as ≥0.3μm particle concentration); the dust particle catcher is used to detect the particulate content in the air to assist in evaluating the cleanliness, and the dust particle catcher can capture particles with a particle size range of 0.5 microns or more; the smart meter is used to collect equipment power consumption data in real time, including fan FFU energy consumption (recording power consumption in different operating modes at full speed, half speed, and off), fresh air system dynamic energy consumption (monitoring real-time power consumption when valve opening changes), and filter additional energy consumption (collecting power consumption of the filter equipment required to maintain cleanliness). The Modbus / MQTT protocol is used to connect to the Internet of Things communication module, and the collected data is uploaded to the edge computing device in real time. The smart meter is integrated into the device controller to directly read the current and voltage parameters and calculate the power consumption.

[0021] As described in step S2 above, the real-time data collected in step S1 is associated with the planned capacity and operation period of the production scheduling system (MES / ERP) to generate a dynamic control strategy, which specifically includes: capacity period division, high capacity period corresponds to full production or overload production plan (such as chip manufacturing peak period), and cleanliness priority must be guaranteed; low capacity period corresponds to half production or shutdown plan (such as night maintenance), with energy saving as the main goal. Cleanliness threshold adjustment, according to ISO 14644-1 standard (i.e., international standard for clean rooms and related control environments), the cleanliness level is dynamically set, a stricter cleanliness threshold is used in high capacity period, and a relaxed cleanliness threshold is used in low capacity period (such as Class 5 in high capacity period, Class 7 in low capacity period), if the real-time cleanliness data is lower than the threshold, trigger an early warning and adjust the strategy. Energy consumption constraint setting, the energy consumption peak value in high capacity period is allowed not to exceed the maximum power consumption under the same historical working conditions (such as full speed operation of the fan), and the energy consumption limit is set to 75% of the historical average power consumption in the low capacity period (such as shutting down 50% of non-critical fans). Production priority strategy: Prioritize the operation of key fans FFU (such as around production equipment) during operating hours to ensure that cleanliness meets standards; shut down equipment in non-production areas during non-operating hours and reduce the fresh air volume to the minimum level to maintain basic cleanliness (such as maintaining only 30% fresh air volume).

[0022] As described in step S3 above, based on the real-time data of S1, the dynamic control strategy of S2 and the historical energy consumption data, the control instructions are calculated through the deep learning model and the equipment is driven to execute. Among them, the historical energy consumption data refers to the historical operation power consumption records of the equipment and scenes in the clean room, including the energy consumption data of the fan FFU in different time periods in the past (such as high capacity, low capacity), AC FFU at different operating frequencies, gears and DC FFU at different voltages and speeds, production scheduling-related energy consumption (overall equipment power consumption under historical production plans (such as full production, half production, and shutdown)) and environmental control energy consumption (records of additional energy consumption required to maintain specific temperature, humidity and cleanliness levels). The deep learning model structure includes an LSTM network module, which uses historical environmental parameters (temperature, humidity, cleanliness), equipment operation data (fan frequency, valve opening), and production scheduling as input to predict the energy consumption demand and cleanliness change trend in the next 1-4 hours. The reinforcement learning module uses real-time environmental parameters, equipment power consumption, and production scheduling status as the state space, fan filter unit opening adjustment value (0-100%), fan start and stop instructions (number + frequency) as the action space, and cleanliness standard reward + energy consumption penalty (if it exceeds the constraint threshold) as the reward function. The control instructions are generated through the model output results: fresh air volume adjustment, calculate the minimum fresh air volume that meets the predicted cleanliness requirements, and convert it into the valve opening percentage (for example, if the prediction needs to increase the cleanliness to Class 5, the opening increases from 40% to 60%); fan start and stop strategy, according to the operation period and energy consumption constraints, determine the fan number and operating frequency to be started (for example, start fan A / B during high-capacity period, frequency 50Hz); physical equipment drive, the fan filter unit (FFU) controller receives the gear and frequency instructions, adjusts the gear and frequency of each fan FFU; receives voltage and speed parameters, and controls the operation status of the DC FFU equipment. Collect actual cleanliness data, power consumption, and equipment status after adjustment; if the actual energy consumption exceeds the predicted threshold or the cleanliness does not meet the standard, trigger model parameter optimization (such as adjusting LSTM weights or reinforcement learning reward functions); adapt to environmental changes through online fine-tuning (such as daily incremental training).

[0023] The deep learning model is a multi-objective optimization model. The LSTM module uses mean square error and cross entropy loss to optimize the prediction accuracy. The reinforcement learning module iteratively optimizes the control strategy in a simulation environment through the Q-learning algorithm and adapts to the feature differences of different clean room scenarios through transfer learning.

[0024] Example 1: Company A's equipment configuration includes 6 EC fans with a total power of 55KW. The operation mode is traditional fixed rule control, and the fans run at full speed all day. The problem is high energy consumption and large fluctuations in cleanliness.

[0025] In this embodiment, by deploying temperature and humidity sensors, cleanliness monitors, dust particle collectors and smart meters, environmental parameters and equipment power consumption data are collected in real time; the data is transmitted to the edge computing device through the Modbus protocol. According to the production schedule, high-capacity periods (8:00-20:00) and low-capacity periods (20:00-8:00) are divided; during high-capacity periods: the cleanliness threshold is set to Class 5, and the energy consumption constraint is 95% of the historical maximum power consumption; during low-capacity periods: the cleanliness threshold is relaxed to Class 7, and the energy consumption limit is 65% of the historical average. Through AI dynamic control and execution, the LSTM network module predicts the cleanliness trend and energy consumption demand in the next 4 hours, and the reinforcement learning module generates fan start and stop instructions and fresh air humidity adjustment values; all 6 fans are started during the high-capacity period, and the fresh air valve opening is 80%, and 3 non-critical fans are turned off during the low-capacity period, and the fresh air valve opening is reduced to 30%.

[0026] The comprehensive energy saving rate is 35%; the cleanliness compliance rate is 98% during high-capacity periods and 100% during low-capacity periods.

[0027] Example 2: Company B has 32 EC fans, 37-75KW, and 18 fresh air FFU fans, 11-30KW. The operation mode is manual intervention control, and the energy consumption fluctuates greatly. The problem to be solved is the large number of equipment, high control complexity, and great energy saving potential.

[0028] This embodiment deploys multiple sets of sensors and smart meters to monitor environmental parameters and equipment power consumption in real time, and transmits data to the cloud database through the MQTT protocol. According to the production schedule, the high-capacity period (7:00-19:00) and the low-capacity period (19:00-7:00) are divided; the cleanliness threshold of the high-capacity period is set to Class 5, and the energy consumption constraint is 90% of the historical maximum power consumption; the cleanliness threshold of the low-capacity period is relaxed to Class 8, and the energy consumption limit is 60% of the historical average. The cleanliness trend and energy consumption demand for the next 6 hours are predicted through a deep learning model, and the fan start and stop instructions and fresh air humidity adjustment values ​​are generated at the same time. All 32 fans (frequency 45Hz) are started during the high-capacity period, and the fresh air valve opening is 85%; 20 non-critical fans are turned off during the low-capacity period, and the fresh air valve opening is reduced to 25%.

[0029] The comprehensive energy saving rate is 38%, the cleanliness compliance rate is 97% during high-capacity periods and 100% during low-capacity periods.

[0030] Example 3: Company C has 6 EC fans with a total power of 22KW. The operation mode is fixed rule control, and the fans run at full speed all day. The needs of high energy consumption, low control accuracy and energy saving need to be solved. By deploying temperature and humidity sensors, cleanliness monitors, dust particle collectors and smart meters, environmental parameters and equipment power consumption data are collected in real time; data is transmitted to edge computing devices through the Modbus protocol. According to the production schedule, high-capacity periods (9:00-21:00) and low-capacity periods (21:00-9:00) are divided. The cleanliness threshold of the high-capacity period is set to Class 5, and the energy consumption constraint is 90% of the historical maximum power consumption; the cleanliness threshold of the low-capacity period is relaxed to Class 7, and the energy consumption limit is 60% of the historical average. Predict the cleanliness trend and energy consumption demand in the next 4 hours and generate fan start and stop instructions and fresh air humidity adjustment values. All 6 fans (frequency 40Hz) are started during the high-capacity period, and the fresh air valve opening is 75%; 3 non-critical fans are turned off during the low-capacity period, and the fresh air valve opening is reduced to 20%.

[0031] The comprehensive energy saving rate is 40%, and the cleanliness compliance rate is 99% during high-capacity periods and 100% during low-capacity periods.

[0032] Through the above examples, this technical solution demonstrates significant energy-saving effects (comprehensive energy-saving rate of 35%-40%) and improved cleanliness compliance rate (97%-100%). The present invention is applicable to clean room scenarios of different scales and configurations, and has wide applicability and economic value.

[0033] Reference Figure 2 , is a structural block diagram of a clean room energy-saving device deployed with an AI algorithm in one embodiment of the present invention, including: The data acquisition module is used to collect real-time environmental parameters and equipment power consumption data through temperature and humidity sensors, cleanliness monitors, dust particle catchers, and smart meters deployed in the clean room; An information association module, used to associate the real-time environmental parameters, the equipment power consumption data and the production scheduling information to generate a dynamic control strategy; The strategy generation module is used to calculate the start and stop strategy of the fan filter unit through a deep learning model based on the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data, and perform corresponding adjustment operations.

[0034] For the specific implementation of each module in the above device example, please refer to the above method embodiment, which will not be repeated here.

[0035] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0036] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0037] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0038] In summary, a method, device and equipment for controlling energy saving in a clean room deployed with an AI algorithm, wherein the method includes collecting real-time environmental parameters and equipment power consumption data through temperature and humidity sensors, cleanliness monitors, dust particle catchers and smart meters deployed in the clean room; associating the real-time environmental parameters and the equipment power consumption data with production scheduling information to generate a dynamic control strategy; based on the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and historical energy consumption data, the start-stop strategy of the fan filter unit is calculated through a deep learning model, and the start-stop strategy includes a three-speed graded control strategy for AC FFU, and dynamically adjusting the voltage and speed of the fan FFU for DC FFU, so as to realize the intelligent linkage between production demand and environmental control, and achieve the purpose of reducing costs and increasing efficiency.

[0039] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0040] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0041] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for controlling clean room energy saving using an AI algorithm, characterized in that: The following steps are involved: The real-time environmental parameters and equipment power consumption data are collected by deploying temperature and humidity sensors, cleanliness monitors, dust particle traps and smart meters in the clean room; the real-time environmental parameters include temperature, humidity, particle concentration and cleanliness data; Associating the real-time environmental parameters, the equipment power consumption data and the production scheduling information to generate a dynamic control strategy, wherein the dynamic control strategy includes the production demand priority, cleanliness control threshold and energy consumption constraint conditions for the future period; Based on the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data, the start-stop strategy of the fan filter unit is calculated through a deep learning model, and the corresponding adjustment operation is performed, wherein the start-stop strategy includes a three-speed graded control strategy for the AC FFU, and a dynamic adjustment of the voltage and speed of the fan FFU for the DC FFU. The historical energy consumption data includes historical power consumption records of the fan filter unit at different operating frequencies, gears, voltages and speeds. The deep learning model takes minimizing energy consumption and maintaining cleanliness standards as optimization goals, and drives the controller to perform differentiated control.

2. The method for controlling clean room energy saving with AI algorithm according to claim 1, characterized in that: The step of collecting equipment power consumption data includes: Real-time monitoring and uploading of power consumption data of the equipment, including power consumption of the fan filter unit in different operating modes, dynamic energy consumption when the fan filter unit gear is adjusted, and additional energy consumption of the filter when maintaining the cleanliness level.

3. The method for controlling clean room energy saving with AI algorithm according to claim 1, characterized in that: The step of associating the real-time environmental parameters, the equipment power consumption data and the production scheduling information to generate a dynamic control strategy includes: According to the planned capacity of the production scheduling information, the production plan is divided into a high capacity period and a low capacity period, wherein the high capacity period corresponds to a full production or overload production plan, and the low capacity period corresponds to a half production or shutdown plan; Dynamically adjust the cleanliness control thresholds at different time periods based on the collected real-time cleanliness data and ISO 14644-1 standard; Combined with historical equipment power consumption data, set energy consumption constraints for each period, including allowing energy consumption peaks to not exceed the maximum power consumption under the same historical operating conditions during high-capacity periods, and setting energy consumption caps at 75% of the average power consumption under the same historical operating conditions during low-capacity periods; Generate production demand priorities based on the operating hours and real-time equipment operating status of production scheduling information, including giving priority to ensuring the operation and cleanliness of key fans FFU during the operating hours, and shutting down equipment in non-production areas and reducing the fresh air volume to the minimum level to maintain basic cleanliness during non-operating hours; The cleanliness control threshold, the energy consumption constraint and the production demand priority are integrated to generate a dynamic control strategy.

4. The method for controlling clean room energy saving with AI algorithm according to claim 1, characterized in that: The deep learning model is a multi-objective optimization model, including an LSTM network module and a reinforcement learning module; wherein the LSTM network module generates energy consumption demand and cleanliness change trend for future time periods through time series prediction based on historical environmental parameters, equipment operation data and production scheduling data; the reinforcement learning module uses the cleanliness threshold and energy consumption constraint in the dynamic control strategy as the reward function, and generates real-time control instructions in combination with the prediction results, including the fan filter unit opening adjustment value, the fan start and stop instructions and the equipment energy consumption allocation strategy; the training of the model is achieved through historical data pre-training and real-time data online fine-tuning.

5. The method for controlling clean room energy saving with AI algorithm according to claim 1, characterized in that: The step of calculating the start and stop strategy of the fan filter unit based on the deep learning model according to the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data includes: Inputting the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data into a deep learning model; Through the LSTM network module in the deep learning model, based on the input data, the cleanliness change trend, temperature and humidity fluctuation range and energy consumption demand in the future set period are predicted; The reinforcement learning module in the deep learning model is used to calculate the minimum fresh air volume and the start-stop strategy of the fan filter unit that meet the predicted cleanliness requirements according to the dynamic control strategy, wherein the start-stop strategy includes the fan number to be started, the operating frequency, the operating time, and the list of non-production area equipment to be shut down.

6. The method for controlling clean room energy saving with AI algorithm according to claim 1, characterized in that: The step of performing the corresponding adjustment operation includes: Based on the start-stop strategy of the fan filter unit, real-time control instructions are generated and executed. The real-time control instructions include hierarchical control instructions for the AC FFU, voltage and speed control instructions for the DC FFU, and physical device drive instructions. The hierarchical control instructions include 1st gear 70% rated power low-speed operation, 2nd gear 80% power balancing mode, and 3rd gear full power full-load operation.

7. The method for controlling clean room energy saving with AI algorithm according to claim 1, characterized in that: After the step of performing the corresponding adjustment operation, the method further includes: Collect the actual cleanliness data after adjustment, the power consumption of the corresponding equipment and the fan status to verify whether the actual energy consumption and cleanliness meet the preset targets. If not, trigger the closed-loop feedback optimization, adjust the model parameters and regenerate the control instructions.

8. A clean room energy-saving device deployed with an AI algorithm, characterized in that: include: The data acquisition module is used to collect real-time environmental parameters and equipment power consumption data through temperature and humidity sensors, cleanliness monitors, dust particle catchers, and smart meters deployed in the clean room; An information association module, used to associate the real-time environmental parameters, the equipment power consumption data and the production scheduling information to generate a dynamic control strategy; The strategy generation module is used to calculate the start and stop strategy of the fan filter unit through a deep learning model based on the real-time environmental parameters, the equipment power consumption data, the dynamic control strategy and the historical energy consumption data, and perform corresponding adjustment operations.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method for controlling clean room energy saving with an AI algorithm deployed in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for controlling clean room energy saving with an AI algorithm deployed in any one of claims 1 to 7 are implemented.

Images