A security management system based on face recognition

Abstract

The present invention relates to the technical field of safety management, and specifically discloses a safety management system based on face recognition, which is used to solve the problem of insufficient collaborative processing of air-conditioning air flow safety management and return air and exhaust air utilization in the event of arsine leakage in the existing clean room air flow management. It includes a lithography production module, an air handling unit, a return air adjustment module, an arsine detection sensor, a face recognition system, an exhaust fan, a fresh air unit, five return air and exhaust air channels, four air supply channels, and one air discharge and recovery channel. Based on face recognition technology, it monitors the personnel in each area and provides a return air and exhaust air adjustment plan and a safety area air supply plan in the event of arsine leakage. The present invention uses face recognition technology to manage the personnel permissions in different clean level areas, adjusts the air supply in the safety area in real time, automatically adjusts the negative pressure and exhaust air in the event of arsine leakage, and dynamically adjusts the fresh air and return air mixing ratio based on the air age to reduce air-conditioning energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of security management. More specifically, the present invention relates to a security management system based on face recognition. Background Art

[0002] The air supply, return air, and exhaust air of the air conditioner involved in the present invention do not involve the production exhaust inside the lithography production equipment, do not include its organic exhaust, and are limited to the air outside the equipment and instruments in the working area.

[0003] In the semiconductor factory area for producing wafers, the cleanliness, temperature, humidity, etc. of the air in the production environment of the wafers are guaranteed through a clean room. The clean air-conditioning treatment unit needs to ensure the air cleanliness, temperature, and humidity stability in each area, and effectively control the accumulation of pollutants to meet the strict environmental requirements for the production of precision production equipment such as lithography machines. However, the semiconductor factory has strict requirements for cleanliness, with an energy consumption as high as 1163 W / ㎡, an air change rate greater than or equal to 100 times / h. In a clean room of ISO class 6, in the form of vertical unidirectional airflow organization according to the designed air outlet velocity of the specification, to maintain positive pressure and cleanliness, overcome the filtration resistance of 2 - 3 levels (about 700 Pa), the air change rate of a room with a size of 12 m × 8 m × 3 m is even as high as 300 - 400 times / h. Therefore, energy consumption is a thorny problem for the air-conditioning system in the semiconductor factory. At the same time, although air circulation can reduce the consumption of fresh air, the control effect of pollutant concentration is not good. And completely introducing fresh air to dilute the airflow to fully dilute the dirty airflow to reach the required cleanliness involves the treatment of fresh air load, and the energy-saving effect is not good (air age); in the treatment of return air and exhaust air in different clean-class areas, the filtration load is high, the efficiency is low, and the resource waste is significant. Face recognition technology can realize the authority management and real-time number monitoring of personnel entering different clean-class areas, so as to dynamically adjust the requirements for air supply, return air, and exhaust air, save energy and improve the control accuracy of air cleanliness. Face recognition can also detect abnormal behaviors in areas with higher clean-class requirements, prevent unauthorized personnel from entering sensitive areas, and ensure the safety of the production environment. Arsine leakage is an extremely serious safety accident in the semiconductor factory area for wafer production. Effective air-conditioning airflow management can reduce the severity of the consequences of arsine leakage safety accidents. In the existing clean-room airflow management, there are deficiencies in the coordinated treatment of air-conditioning airflow safety management and the utilization of return air and exhaust air when arsine leaks. How to solve the current deficiencies based on face recognition technology and the deployment of return air and exhaust airflows has become a technical problem that needs to be solved urgently. Summary of the Invention

[0004] To overcome the above-mentioned defects of the prior art, the present invention provides a security management system based on face recognition, which uses face recognition technology to manage the personnel permissions in areas with different cleanliness levels, adjusts the air supply in the safe area in real time, automatically adjusts the negative pressure and exhaust when arsine leaks, dynamically adjusts the fresh air and return air mixing ratio based on the air age, and reduces the air conditioning energy consumption.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A security management system based on face recognition includes a lithography production module, an air handling unit, a return air adjustment module, an arsine detection sensor, a face recognition system, an exhaust fan, a fresh air unit, a shunt device, five return air channels, one exhaust air channel, five supply air channels, and one waste heat recovery channel. The lithography production module has a pretreatment and cleaning area, a lithography coating area, an exposure area, and a development and post-treatment area. The return air adjustment module includes a particle filtration section, a fine dust filtration section, a temperature and humidity adjustment section, a return air mixing section, and an exhaust gas discharge section. The air handling unit includes a first air filtration unit, a second filtration unit, a first return air mixing unit, an electrostatic filtration unit, a harmful gas filtration unit, a second return air mixing unit, an air circulation purification unit, a zoning control unit, a harmful gas deep purification unit, a waste heat recovery unit, and an air discharge port. When the arsine detection sensor detects an arsine leakage signal, it activates the face recognition system to identify the identity and location of the personnel in the leakage area, sends an emergency evacuation instruction to the personnel in the leakage area and guides the designated evacuation route to the safe area. At the same time, it closes the supply air channel in the leakage area, preferentially increases the air supply in the safe area based on the preset dynamic air supply formula to make it meet the air parameter requirements of the safe area, sets the exhaust equipment on the exhaust air channel to the full speed mode, discharges arsine through the exhaust gas discharge section and the air discharge port at the same time, and creates a directional air flow isolation by setting a negative pressure of -20 Pa to -30 Pa within 15 meters of the leakage point. It activates the harmful gas deep purification unit and temporarily closes the waste heat recovery unit. The preset dynamic air supply formula is:

[0007] ;

[0008] In the formula: is the dynamic air supply volume, is the basic air supply volume, which is set based on the minimum air supply design requirement of each area, is the number of personnel, is the personnel activity coefficient, is the cleanliness level coefficient, is the adjustment coefficient.

[0009] As a further aspect of the present invention, the five air supply channels include a first air supply channel, a second air supply channel, a third air supply channel, a fourth air supply channel, and a fifth air supply channel, and their air supplies are the first air supply, the second air supply, the third air supply, the fourth air supply, and the fifth air supply respectively. The five air return channels include a first air return channel, a second air return channel, a third air return channel, a fourth air return channel, and a fifth air return channel, and their air returns are the first air return, the second air return, the third air return, the fourth air return, and the fifth air return respectively.

[0010] As a further aspect of the present invention, the first air return channel is formed by unidirectionally connecting the photolithography coating area with the first air return mixing unit and the electrostatic filtration unit in sequence through an air duct, and is used for transporting the first air return;

[0011] In the first air supply channel, the cleanliness level of the first air supply is ISO3, the temperature control accuracy is ±0.1 °C, the humidity control accuracy is ±2%, and the target air age is 1.5 minutes, and it is used for supplying air to the photolithography coating area.

[0012] As a further aspect of the present invention, the second air return channel is formed by unidirectionally connecting the exposure area with the first air filtration unit, the second filtration unit, and the first air return mixing unit in sequence through an air duct, and is used for transporting the second air return;

[0013] In the second air supply channel, the cleanliness level of the second air supply is ISO2, the temperature control accuracy is ±0.05 °C, the humidity control accuracy is ±1%, and the target air age is 1 minute, and it is used for supplying air to the exposure area.

[0014] As a further aspect of the present invention, the third air return channel is formed by unidirectionally connecting the development and post-treatment area with the second air return mixing unit and the air circulation purification unit through an air duct, and is used for transporting the third air return;

[0015] In the third air supply channel, the cleanliness level of the third air supply is ISO4, the temperature control accuracy is ±0.2 °C, the humidity control accuracy is ±3%, and the target air age is 4 minutes, and it is used for supplying air to the development and post-treatment area.

[0016] As a further aspect of the present invention, the fourth air return channel is formed by unidirectionally connecting the pre-treatment and cleaning area with the partition control unit through an air duct, and is used for transporting the fourth air return;

[0017] In the fourth air supply channel, the cleanliness level of the fourth air supply is ISO8, the temperature control accuracy is ±0.5 °C, the humidity control accuracy is ±5%, and the target air age is 7 minutes, and it is used for supplying air to the pre-treatment and cleaning area.

[0018] As a further aspect of the present invention, the fifth air return channel is formed by unidirectionally connecting the electrostatic filtration unit with the harmful gas filtration unit and the second air return mixing unit through an air duct, and is used for transporting the fifth air return;

[0019] The fifth air supply channel is formed by connecting the air circulation and purification unit with the zoning control unit and the flow splitting device through an air duct, and is used to transport the circulated air adjusted by the return air adjustment device to the zoning control unit.

[0020] As a further solution of the present invention, the zoning control unit is connected to a fresh air unit, and the fresh air unit is used to supply fresh air transmitted to the zoning control unit. In the zoning control unit, the mixing ratio of fresh air and return air is obtained through a preset fresh air ratio adaptive adjustment formula based on the difference between the target air age and the real-time monitored air age. The preset fresh air ratio adaptive adjustment formula is:

[0021] ;

[0022] In the formula: is the fresh air adaptive adjustment ratio, which is used to adjust the opening degree of the valve body in the fresh air duct, is the basic fresh air ratio, which is set according to the requirements of maintaining the set positive pressure value and cleanliness level in each area, is the sensitivity adjustment coefficient, which is obtained based on the historical data analysis of the response speed of the valve bodies in the actual fresh air duct and return air duct, 、 are the target air age and the actually monitored air age respectively. The actually monitored air age is obtained through the gas tracer method. The return air ratio is and is used to adjust the opening degree of the valve body in the return air duct.

[0023] As a further solution of the present invention, the exhaust air channel is formed by connecting the zoning control unit with the harmful gas deep purification unit, and is used to transport the exhaust air. The harmful gas deep purification unit, the waste heat recovery unit, and the air discharge port are connected unidirectionally to form a waste heat recovery channel.

[0024] As a further solution of the present invention, the harmful gas filtration unit is a composite filter formed by combining an activated carbon filter and a HEPA filter.

[0025] As a further solution of the present invention, the first return air mixing unit and the second return air mixing unit adaptively adjust the mixed air ratio based on the change of the fresh air load to control the working conditions of each return air, reduce the fresh air load, and indirectly reduce the energy consumption caused by the fresh air load. The first return air mixing unit mixes the return air provided by the photolithography coating area and the exposure area. The photolithography coating area and the exposure area have high requirements for air cleanliness, which are ISO level 3 and ISO level 2 respectively. The cleanliness, temperature and humidity, and air age are monitored in real time, and the ratio of these two parts of return air is adjusted to ensure that the air quality delivered to these areas meets the standards. When the fresh air load is in the peak area In this case, the demand for fresh air is reduced by increasing the return air volume from the photolithography coating area. The return air from the photolithography coating area is processed, and the proportion of return air from the photolithography coating area that needs to be increased is calculated based on the requirements of cleanliness, temperature and humidity, and air age, and the dependence on the return air from the exposure area is reduced, thereby optimizing the return air utilization effect; the second return air mixing unit is responsible for mixing the second return air and the fifth return air for development and post-processing, and dynamically adjusts the proportion of these two parts of return air according to air quality requirements and actual load conditions. When the fresh air load is large, the return air volume from the development and post-processing areas is increased first to reduce the use of fresh air.

[0026] In order to solve the technical problems existing in the prior art, the present invention proposes the technical effect of the system: the present invention uses face recognition technology to manage the authority of personnel entering areas of different clean levels, and monitors the number of personnel in real time, and dynamically increases the air supply in the safe area according to the personnel, so as to detect unauthorized illegal entry, prevent personnel from violating regulations, and ensure the safety of the production environment. When arsine leaks, the negative pressure within 15 meters of the leakage point is adjusted, full-speed exhaust is started, and gas is treated through exhaust gas emission and deep purification units to reduce the impact of leakage accidents. At the same time, the return air and fresh air ratio is controlled by air age to maintain a balance between air circulation and fresh air load, thereby achieving energy saving while optimizing cleanliness control. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of the system of the present invention;

[0028] In the figure: A, photolithography production module; B, return air adjustment module; C, pre-treatment and cleaning area; D, photolithography coating area; E, exposure area; F, development and post-treatment area; W, air treatment group; 1, first air filtration unit; 2, second filtration unit; 3, first return air mixing unit; 4, electrostatic filtration unit; 5, harmful gas filtration unit; 6, second return air mixing unit; 7, air circulation purification unit; 8, partition control unit; 9, harmful gas deep purification unit; 10, waste heat recovery unit; 31, particle filtration section; 32, fine dust filtration section; 33, temperature and humidity adjustment section; 34, return air mixing section; 35, exhaust gas emission section. DETAILED DESCRIPTION

[0029] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0030] Embodiment 1. To solve the technical problems mentioned in the background art, a specific implementation scheme of the system proposed by the present invention is provided.

[0031] As Figure 1 shown, a security management system based on face recognition proposed by the present invention includes a photolithography production module A, an air treatment group W, and a return air adjustment module B. It also includes an arsine detection sensor, a face recognition system, an exhaust fan, a fresh air unit, a shunt device, five return air channels, one exhaust air channel, five supply air channels, and one waste heat recovery channel. The photolithography production module A has a pretreatment and cleaning area C, a photolithography coating area D, an exposure area E, and a development and post-treatment area F. The return air adjustment module B includes a particle filtration section 31, a fine dust filtration section 32, a temperature and humidity adjustment section 33, a return air mixing section 34, and an exhaust gas emission section 35. The air treatment group W includes a first air filtration unit 1, a second filtration unit 2, a first return air mixing unit 3, an electrostatic filtration unit 4, a harmful gas filtration unit 5, a second return air mixing unit 6, an air circulation purification unit 7, a zoning control unit 8, a harmful gas deep purification unit 9, a waste heat recovery unit 10, and an air discharge port. When the arsine detection sensor detects an arsine leakage signal, the face recognition system is activated to identify the identity and location of the personnel in the leakage area, send an emergency evacuation instruction to the personnel in the leakage area, and guide the designated evacuation route to a safe area. At the same time, the supply air channel in the leakage area is closed, and the air supply in the safe area is preferentially increased based on a preset dynamic air supply formula to meet the air parameter requirements of the safe area. The exhaust equipment on the exhaust air channel is set to the full-speed mode, and the arsine is discharged through the air discharge port at the same time. At the same time, a negative pressure of -20 Pa to -30 Pa is set within 15 meters of the leakage point to create a directional air flow isolation, the harmful gas deep purification unit 9 is activated, and the waste heat recovery unit 10 is temporarily closed. The preset dynamic air supply formula is:

[0032] ;

[0033] In the formula: is the dynamic air supply volume, is the basic air supply volume, set based on the minimum air supply design requirements of each area, is the number of people, is the personnel activity coefficient, is the cleanliness level coefficient, is the adjustment coefficient.

[0034] It should be noted that the five air supply channels include the first air supply channel, the second air supply channel, the third air supply channel, the fourth air supply channel, and the fifth air supply channel, and their air supplies are the first air supply, the second air supply, the third air supply, the fourth air supply, and the fifth air supply respectively. The five air return channels include the first air return channel, the second air return channel, the third air return channel, the fourth air return channel, and the fifth air return channel, and their air returns are the first air return, the second air return, the third air return, the fourth air return, and the fifth air return respectively.

[0035] It should be noted that the first air return channel is formed by unidirectionally connecting the photolithography and coating area D, the first air return mixing unit 3, and the electrostatic filtration unit 4 in sequence through an air duct, and is used to transport the first air return;

[0036] In the first air supply channel, the cleanliness level of the first air supply is ISO3, the temperature control accuracy is ±0.1°C, the humidity control accuracy is ±2%, and the target air age is 1.5 minutes, and it is used to supply air to the photolithography and coating area D.

[0037] It should be noted that the second air return channel is formed by unidirectionally connecting the exposure area E, the first air filtration unit 1, the second filtration unit 2, and the first air return mixing unit 3 in sequence through an air duct, and is used to transport the second air return.

[0038] In the second air supply channel, the cleanliness level of the second air supply is ISO2, the temperature control accuracy is ±0.05°C, the humidity control accuracy is ±1%, and the target air age is 1 minute, and it is used to supply air to the exposure area E.

[0039] It should be noted that the third air return channel is formed by unidirectionally connecting the developing and post-processing area F, the second air return mixing unit 6, and the air circulation and purification unit 7 through an air duct, and is used to transport the third air return;

[0040] In the third air supply channel, the cleanliness level of the third air supply is ISO4, the temperature control accuracy is ±0.2°C, the humidity control accuracy is ±3%, and the target air age is 4 minutes, and it is used to supply air to the developing and post-processing area F.

[0041] It should be noted that the fourth air return channel is formed by unidirectionally connecting the pretreatment and cleaning area C to the zoning control unit 8 through an air duct, and is used to transport the fourth air return;

[0042] In the fourth air supply channel, the cleanliness level of the fourth air supply is ISO8, the temperature control accuracy is ±0.5°C, the humidity control accuracy is ±5%, and the target air age is 7 minutes, and it is used to supply air to the pretreatment and cleaning area C.

[0043] It should be noted that the fifth air return channel is formed by unidirectionally connecting the electrostatic filtration unit 4, the harmful gas filtration unit 5, and the second air return mixing unit 6 through an air duct, and is used to transport the fifth air return;

[0044] The fifth air supply channel is formed by the air circulation purification unit 7 being connected with the partition control unit 8 and the diversion device through the air duct, and is used to transport the circulating air adjusted by the return air adjustment device B to the partition control unit 8.

[0045] It should be noted that the partition control unit 8 is connected to the fresh air unit, and the fresh air unit is used to supply fresh air transmitted to the partition control unit 8. In the partition control unit 8, the mixing ratio of fresh air and return air is based on the difference between the target air age and the real-time monitored air age, and is obtained by a preset fresh air ratio adaptive adjustment formula. The preset fresh air ratio adaptive adjustment formula is:

[0046] ;

[0047] Where: It is used to adjust the opening of the valve in the fresh air duct. The basic fresh air ratio is set according to the positive pressure value and cleanliness level requirements of each area. is the sensitivity adjustment coefficient, which is obtained based on the historical data analysis of the valve response speed in the actual fresh air duct and return air duct. , are the target air age and the actual monitored air age, respectively. The actual monitored air age is obtained by the gas tracer method. The return air ratio is , used to adjust the opening of the valve body in the return air duct.

[0048] It should be noted that the exhaust passage is formed by connecting the partition control unit 8 with the harmful gas deep purification unit 9, which is used to transport exhaust air. The harmful gas deep purification unit 9, the waste heat recovery unit 10, and the air discharge port are unidirectionally connected to form a waste heat recovery passage.

[0049] It should be noted that the harmful gas filtering unit 5 is a composite filter formed by combining an activated carbon filter and a HEPA filter.

[0050] The present invention utilizes face recognition technology to manage the authority of personnel entering areas of different cleanliness levels, and monitors the number of personnel in real time. It increases the air supply in the safe area dynamically according to the personnel, can detect unauthorized illegal entry, prevent personnel from violating regulations, and ensure the safety of the production environment. When arsine leaks, it adjusts the negative pressure within 15 meters of the leak point, starts full-speed exhaust, and processes the gas through exhaust gas emission and deep purification units to reduce the impact of leakage accidents. At the same time, the return air and fresh air ratio is controlled by air age to maintain a balance between air circulation and fresh air load, thereby achieving energy saving while optimizing cleanliness control.

[0051] Embodiment 2. To further optimize the coordination of the overall technical solution, and at the same time, use the system proposed by the present invention to dynamically adjust the return air mixing ratios of the first return air mixing unit 3 and the second return air mixing unit 6 in real time according to the return air parameters of each area under the background of fluctuating fresh air load. The specific implementation scheme is described as follows:

[0052] The first return air mixing unit 3 and the second return air mixing unit 6 adaptively adjust the mixed air ratio based on the change of fresh air load and the return air parameters of each area to control the ratio of each return air, reduce the fresh air load, and reduce the energy consumption caused by the fresh air load. The first return air mixing unit 3 mixes the return air provided by the photolithography coating area D and the exposure area E. The photolithography coating area D and the exposure area E have high requirements for air cleanliness, which are ISO 3 and ISO 2 respectively. The cleanliness, temperature, humidity and air age are monitored in real time, and the ratio of these two parts of return air is adjusted to ensure that the air quality sent to these areas meets the standards, so that the ratio of the return air with higher cleanliness in the return air parameters of each return air is increased. When the fresh air load is in the peak area and the change range of the return air temperature, humidity and cleanliness of the photolithography coating area D from the set value does not exceed 2.5%, the fresh air demand is reduced by increasing the return air volume from the photolithography coating area D. The adjustment ratio is based on the historical supply and return air parameters and fresh air load. According to the machine learning model, while meeting the air parameter requirements of each area, the change rule of the return air volume ratio of the photolithography coating area D with the change of fresh air load is learned, and the adjustment ratio of the return air volume of the photolithography coating area D is obtained, so as to optimize the return air utilization effect; The second return air mixing unit 6 is responsible for mixing the second return air and the fifth return air of the developing and post-processing area F, and dynamically adjusts the ratio of these two parts of return air according to the air quality requirements and actual load conditions. The adjustment mechanism is the same as that of the first return air mixing unit 3. Using the machine learning model, the rule corresponding to the change of the return air ratio of the developing and post-processing area F with the change of fresh air load is learned, and then the return air ratio of the developing and post-processing area F is obtained. When the fresh air load is large, the return air volume from the developing and post-processing area F is preferentially increased to reduce the use of fresh air. The actuators in the first return air mixing unit 3 and the second return air mixing unit 6 are adjusted in time according to the obtained return air volume adjustment ratio.

[0053] The above is only the specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claimed rights.

[0054] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A security management system based on face recognition, comprising a photolithography production module (A), an air handling group (W) and a return air adjustment module (B), characterized in that: The system also includes an arsine detection sensor, a face recognition system, an exhaust fan, a fresh air unit, a diversion device, five return air channels, an exhaust air channel, five air supply channels and a waste heat recovery channel. The photolithography production module (A) has a pre-treatment and cleaning area (C), a photolithography coating area (D), an exposure area (E), a development and post-treatment area (F). The return air adjustment module (B) includes a particle filtering section (31), a fine dust filtering section (32), a temperature and humidity adjustment section (33), a return air mixing section (34), and an exhaust gas emission section (35). The air treatment group (W) includes a first air filtering unit (1), a second filtering unit (2), a first return air mixing unit (3), an electrostatic filtering unit (4), a harmful gas filtering unit (5), a second return air mixing unit (6), an air circulation purification unit (7), a partition control unit (8), and a harmful gas The deep purification unit (9), the waste heat recovery unit (10) and the air discharge port, the arsine detection sensor detects the arsine leakage signal, starts the face recognition system, identifies the identity and location of the personnel in the leakage area, sends an emergency evacuation order to the personnel in the leakage area and guides them to evacuate to the designated safe area, and at the same time closes the air supply channel in the leakage area, and based on the preset dynamic air supply formula, gives priority to increasing the air supply in the safe area so that it meets the air parameter requirements of the safe area, sets the exhaust equipment on the exhaust channel to full speed mode, and discharges arsine through the exhaust gas discharge section (35) and the air discharge port at the same time, and sets the negative pressure within 15 meters of the leakage point to -20Pa to -30Pa, creates directional airflow isolation, starts the harmful gas deep purification unit (9), and temporarily closes the waste heat recovery unit (10). The preset dynamic air supply formula is: ； Where: is the dynamic air supply volume, The basic air supply is set based on the minimum air supply design demand for each area. is the number of personnel, is the personnel activity coefficient, is the cleanliness grade coefficient, is the adjustment coefficient; The first return air channel is formed by the photolithography coating area (D) and the first return air mixing unit (3) and the electrostatic filtering unit (4) being connected in one direction in sequence through the air duct, and is used for conveying the first return air; In the first air supply channel, the cleanliness level of the first air supply is ISO3, the temperature control accuracy is ±0.1°C, the humidity control accuracy is ±2%, and the target air age is 1.5 minutes, which is used to supply air to the photolithography coating area (D); The second return air channel is formed by the exposure area (E) and the first air filter unit (1), the second filter unit (2), and the first return air mixing unit (3) being connected in one direction in sequence through the air duct, and is used for conveying the second return air. In the second air supply channel, the cleanliness level of the second air supply is ISO2, the temperature control accuracy is ±0.05°C, the humidity control accuracy is ±1%, and the target air age is 1 minute, which is used to supply air to the exposure area (E); The third return air channel is formed by the development and post-processing area (F) being connected to the second return air mixing unit (6) and the air circulation purification unit (7) in one direction through an air duct, and is used for conveying the third return air; In the third air supply channel, the cleanliness level of the third air supply is ISO4, the temperature control accuracy is ±0.2℃, the humidity control accuracy is ±3%, and the target air age is 4 minutes. It is used to supply air to the development and post-processing area (F); The fourth return air channel is formed by the pre-treatment and cleaning area (C) through the air duct unidirectionally connected to the partition control unit (8), and is used to transport the fourth return air; In the fourth air supply channel, the cleanliness level of the fourth air supply is ISO8, the temperature control accuracy is ±0.5℃, the humidity control accuracy is ±5%, and the target air age is 7 minutes. It is used to supply air to the pretreatment and cleaning area (C); The fifth return air channel is formed by the electrostatic filter unit (4) being connected to the harmful gas filter unit (5) and the second return air mixing unit (6) in one direction through an air duct, and is used for conveying the fifth return air; The fifth air supply channel is formed by the air circulation purification unit (7) being connected to the partition control unit (8) and the flow dividing device through the air duct, and is used to convey the circulating air adjusted by the return air adjustment device (B) to the partition control unit (8); The partition control unit (8) is connected to the fresh air unit, and the fresh air unit is used to supply fresh air to the partition control unit (8). In the partition control unit (8), the mixing ratio of fresh air and return air is based on the difference between the target air age and the real-time monitored air age, and is obtained by a preset fresh air ratio adaptive adjustment formula. The preset fresh air ratio adaptive adjustment formula is: ； Where: It is used to adjust the opening of the valve in the fresh air duct. The basic fresh air ratio is set according to the positive pressure value and cleanliness level requirements of each area. is the sensitivity adjustment coefficient, which is obtained based on the historical data analysis of the valve response speed in the actual fresh air duct and return air duct. , are the target air age and the actual monitored air age, respectively. The actual monitored air age is obtained by the gas tracer method. The return air ratio is , used to adjust the opening of the valve body in the return air duct.

2. A security management system based on face recognition according to claim 1, characterized in that: The exhaust passage is formed by connecting the partition control unit (8) and the harmful gas deep purification unit (9) and is used to convey exhaust air. The harmful gas deep purification unit (9), the waste heat recovery unit (10) and the air discharge port are unidirectionally connected to form the waste heat recovery passage.

3. A security management system based on face recognition according to claim 1, characterized in that: The harmful gas filtering unit (5) is a composite filter formed by combining an activated carbon filter and a HEPA filter.

Images