A cooling tower efficiency system and a control method thereof

Abstract

The application discloses a cooling tower efficiency increasing system and a control method thereof. The cooling tower efficiency increasing system is provided with a water inlet pipe connected with a water inlet of a cooling tower and a multi-scale bubble generating device including a Venturi tube, a micro-bubble generating unit and a medium-bubble generating unit connected with the water inlet pipe, so that the contact area and time of water and air are increased, the internal structure of the cooling tower does not need to be modified, and the continuous operation of the whole is not affected. The flow of liquid flowing into the multi-scale bubble generating device is controlled by a control device, the micro-bubble generating unit and the medium-bubble generating unit are arranged on the Venturi tube and are communicated with each other, the flow of gas injected into a gas conveying pipe by an air compressor is controlled by the control device, and an air outlet electric control valve on a corresponding branch gas pipe is controlled, so that the flow of bubbles of each scale is accurately adjusted. Through the dynamic coupling effect of the multi-scale bubble formation water-air interface expansion and fluid disturbance combination, the contact area of water and air is expanded, and the cooling efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of cooling tower technology, and in particular to a cooling tower efficiency enhancement system and its control method. Background Technology

[0002] In fields such as water-cooled air conditioning systems and industrial circulating cooling systems, cooling towers serve as core heat dissipation equipment, and their performance directly determines the efficiency and energy consumption level of the entire refrigeration or cooling system. Cooling towers achieve cooling by allowing circulating water to directly contact the air, transferring the heat carried by the water to the atmosphere. Therefore, improving the thermal performance of cooling towers has always been a key focus of ongoing research in this field.

[0003] Existing methods for improving the thermal performance of cooling towers include physical structure optimization and operation control. Physical structure optimization mainly involves improving the internal components of the cooling tower to increase the contact area and time between water and air. Operation control mainly involves adjusting the operating parameters of the cooling tower to achieve energy savings. However, improving the internal components of the cooling tower fails to effectively regulate the heat and mass transfer process at the water-air interface at the microscopic level. Furthermore, adjusting the operating parameters of the cooling tower is mostly based on the designed operating mode settings, failing to establish a deep coupling optimization relationship with real-time changes in ambient temperature, humidity, and heat load, resulting in poor operational adaptability. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a cooling tower efficiency enhancement system and its control method, so as to effectively regulate the heat and mass transfer process at the water-air interface at the micro level and improve the adaptability of operation.

[0005] In a first aspect, the present invention provides a cooling tower efficiency enhancement system, comprising a chiller, a cooling tower, a multi-scale bubble generator, an air compressor, and a control device. The cooling tower has an inlet connected to a water inlet pipe, and its outlet connected to the condenser inlet of the chiller via a return water pipe. The condenser outlet of the chiller is connected to the water inlet pipe via a cooling water transmission pipe and a cooling pump. A first temperature sensor is installed on the water inlet pipe near the cooling tower's inlet, and a second temperature sensor is installed on the return water pipe near the cooling tower's outlet. The cooling tower's air intake... The inlet is equipped with a temperature and humidity sensor; the multi-scale bubble generator is connected to the water inlet pipe and includes a Venturi tube, at least one microbubble generating unit, and at least one medium-sized bubble generating unit. Both ends of the Venturi tube are connected to the water inlet pipe via branch water pipes. One of the branch water pipes is equipped with a first electrically controlled valve, and a second electrically controlled valve is equipped on the section of the water inlet pipe corresponding to the Venturi tube. The microbubble generating unit and the medium-sized bubble generating unit are both located on the Venturi tube and are connected to each other; the air compressor outlet... The outlet is connected to the gas supply pipe, which is connected to the microbubble generating unit and the medium-sized bubble generating unit via branch gas pipes. Each branch gas pipe is equipped with an outlet electrically controlled valve. The control device is electrically connected to the chiller, the cooling tower, the cooling pump, the first temperature sensor, the second temperature sensor, the temperature and humidity sensor, the first electrically controlled valve, the second electrically controlled valve, the outlet electrically controlled valve, and the air compressor. Based on the collected real-time operating status information of the cooling pump, the cooling tower, and the air compressor, as well as the inlet and outlet water temperatures of the cooling tower, and combined with the real-time wet-bulb temperature, a deep reinforcement learning model is used for analysis and processing. This generates corresponding action commands for adjusting the operating parameters of the cooling pump, the cooling tower fan, and the bubble generation ratio and zoning strategy of the multi-scale bubble generating device. The device also controls the operation of the chiller, the cooling tower, the cooling pump, the first temperature sensor, the second temperature sensor, the temperature and humidity sensor, the first electrically controlled valve, the second electrically controlled valve, the air compressor, and the outlet electrically controlled valve.

[0006] Secondly, the present invention also provides a control method for a cooling tower efficiency enhancement system, applied to the aforementioned cooling tower efficiency enhancement system. The control method includes the following steps: acquiring real-time inlet and outlet water temperatures of the cooling tower from corresponding temperature sensors via a control device; acquiring real-time wet-bulb temperatures; and acquiring operating status information of the cooling tower, cooling pump, and air compressor. The control device performs filtering and normalization processing on the acquired real-time data to construct a current state vector; and calculates the real-time heat transfer coefficient and unit cooling energy consumption of the cooling tower based on the current state vector. The control device uses a deep reinforcement learning model to generate action vectors based on the current state vectors; and generates corresponding parameters for the cooling pump, cooling tower fan, and bubble generation ratio and partitioning of the multi-scale bubble generator based on the elements of the action vectors. The system executes and outputs action commands to adjust the strategy; the control device calculates the reward corresponding to the action vector based on the action vector and the current state vector, combined with a reward mechanism; after the action command is executed, the control device acquires the inlet and outlet water temperatures of the cooling tower at the next moment, the wet-bulb temperature at the next moment, and the operating status information of the cooling tower, cooling pump, and air compressor at the next moment. Based on the data information at the next moment, the system performs filtering and normalization to construct an updated state vector; the control device uses the current state vector, action vector, reward, and updated state vector to form a sample, stores the sample in an experience replay buffer, and randomly samples a preset number of samples at preset time intervals in the experience replay buffer. The system evaluates the adjustment based on the average reward monitored by the sampled samples and trains and updates the deep reinforcement learning model.

[0007] The beneficial technical effects of this invention are as follows: The cooling tower efficiency enhancement system of this invention increases the contact area and time between water and air by setting an inlet pipe connected to the water inlet of the cooling tower and a multi-scale bubble generating device, including a Venturi tube, a microbubble generating unit, and a medium-sized bubble generating unit, connected to the inlet pipe. This requires no modification to the internal structure of the cooling tower and does not affect the overall continuous operation. Both ends of the Venturi tube are connected to the inlet pipe via branch water pipes. A first electrically controlled valve on the branch water pipe and a second electrically controlled valve on the corresponding section of the inlet pipe control the flow rate of liquid flowing into the multi-scale bubble generating device. The microbubble generating unit and the medium-sized bubble generating unit are both located on the Venturi tube and interconnected. The control device controls the flow rate of gas injected from the air compressor into the gas delivery pipe and controls the corresponding outlet electrically controlled valve on the branch gas pipe to precisely adjust the flow rate of bubbles at each scale. Through the dynamic coupling effect of multi-scale bubbles forming a water-air interface expansion and fluid disturbance, the contact area between water and air is increased, improving cooling efficiency. Efficiency is achieved by setting up a first temperature sensor and a second temperature sensor to collect the inlet and outlet water temperatures of the cooling tower, respectively, and a temperature and humidity sensor at the air inlet of the cooling tower. A control device connected to each sensor calculates the wet-bulb temperature based on the collected dry-bulb temperature and relative humidity using a humidity measurement formula. This calculation is then combined with the inlet and outlet water temperatures of the cooling tower and analyzed using a deep reinforcement learning model. Simultaneously, real-time operating status information of the cooling tower, cooling pump, and air compressor can be analyzed using the same deep reinforcement learning model to generate corresponding action commands for adjusting the operating parameters of the cooling pump, the cooling tower fan, and the bubble generation ratio and zoning strategy of the multi-scale bubble generator. This enables intelligent optimization of the operating strategy and bubble generation and distribution strategy based on real-time operating conditions, achieving precise control under all operating conditions. This effectively saves energy without manual intervention, improves operational adaptability, and allows for effective control of heat and mass transfer at the water-air interface at the microscopic level. The control method of the cooling tower efficiency enhancement system of this invention also possesses the above-mentioned functions. Attached Figure Description

[0008] 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a connection diagram of the cooling tower efficiency enhancement system provided by the present invention when the chiller is omitted;

[0010] Figure 2A schematic diagram of the multi-scale bubble generator for the cooling tower efficiency enhancement system provided by the present invention;

[0011] Figure 3 This is a schematic diagram of the architecture of a specific embodiment of the cooling tower efficiency enhancement system provided by the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] Please see Figures 1 to 3 , Figure 1This is a connection diagram of the cooling tower efficiency enhancement system provided by the present invention when the chiller is omitted. The cooling tower efficiency enhancement system includes a chiller 17, a cooling tower 11, a multi-scale bubble generator 12, an air compressor 13, and a control device 16. The inlet of the cooling tower 11 is connected to an inlet pipe 111. The outlet of the cooling tower 11 is connected to the condenser inlet of the chiller 17 via a return pipe 112. The condenser outlet of the chiller 17 is connected to the inlet pipe 111 via a cooling water transmission pipe 171 and a cooling pump 172. A first temperature sensor is installed on the inlet pipe 111 near the inlet of the cooling tower 11. A [missing information - likely a temperature sensor or a temperature sensor] is installed on the return pipe 112 near the outlet of the cooling tower 11. A second temperature sensor is provided at the air inlet of the cooling tower 11; the multi-scale bubble generator 12 is connected to the water inlet pipe 111 and includes a Venturi tube 121, at least one microbubble generating unit, and at least one medium-sized bubble generating unit. Both ends of the Venturi tube 121 are connected to the water inlet pipe 111 via branch water pipes 101. A first electrically controlled valve 102 is provided on one of the branch water pipes 101, and a second electrically controlled valve 103 is provided on the section of the water inlet pipe 111 corresponding to the Venturi tube 121. The microbubble generating unit and the medium-sized bubble generating unit are both located on the Venturi tube 121. The air compressor 13 is connected to the air supply pipe 131, which is connected to the microbubble generating unit and the medium bubble generating unit via branch air pipes 132. Each branch air pipe 132 is equipped with an outlet solenoid valve 133. The control device 16 is electrically connected to the chiller 17, the cooling tower 11, the cooling pump 172, the first temperature sensor, the second temperature sensor, the temperature and humidity sensor, the first solenoid valve 102, the second solenoid valve 103, the outlet solenoid valve 133, and the air compressor 13. It is used to monitor the real-time operating status of the cooling pump 172 and the cooling tower 11. Information, the working status information of the air compressor 13, and the inlet and outlet water temperatures of the cooling tower 11, combined with real-time wet-bulb temperatures, are analyzed and processed using a deep reinforcement learning model to generate corresponding action commands for adjusting the operating parameters of the cooling pump 172, the operating parameters of the fan of the cooling tower 11, and the bubble generation ratio and zoning strategy of the multi-scale bubble generator 12. These commands control the operation of the chiller 17, the cooling tower 11, the cooling pump 172, the first temperature sensor, the second temperature sensor, the temperature and humidity sensor, the first solenoid valve 102, the second solenoid valve 103, the air compressor 13, and the outlet solenoid valve 133.

[0014] The operating status information of the cooling pump 172 includes its power and cooling water flow rate. The power of the cooling pump 172 can be obtained through a power transmitter, and the cooling water flow rate can be detected and collected by a flow sensor installed at the output end of the cooling pump 172. The operating status information of the cooling tower 11 includes the inlet and outlet water temperature difference, the fan speed, the fan power, and the water pump power. The inlet and outlet water temperature difference can be calculated based on data collected by the first and second temperature sensors. Both the first and second temperature sensors can be platinum resistance temperature sensors. The fan speed can be obtained through an encoder installed on the fan motor, and the water pump power can be obtained through a water pump frequency converter or a power transmitter. The operating status information of the air compressor 13 includes the air flow rate, which can be obtained through a flow sensor. The wet-bulb temperature is acquired by a temperature and humidity sensor installed at the air inlet of the cooling tower 11. This sensor is either a dry-bulb or wet-bulb sensor or a capacitive sensor. The wet-bulb temperature is calculated by directly measuring the dry-bulb temperature and relative humidity, combined with a psychrometric formula. The bubble generation ratio and zoning strategy of the multi-scale bubble generator 12 can be adjusted in real time by the control device 16 using a frequency converter to regulate the gas flow rate of the air compressor 13 and the opening of the outlet solenoid valve 133. All sensor signals can be aggregated by a programmable logic controller (PLC) or an IoT gateway and transmitted to the control device 16.The cooling tower efficiency enhancement system increases the contact area and time between water and air by setting up an inlet pipe 111 connected to the inlet of the cooling tower 11 and a multi-scale bubble generator 12, including a Venturi tube 121, a microbubble generating unit, and a medium-scale bubble generating unit, connected to the inlet pipe 111. This requires no modification to the internal structure of the cooling tower 11 and does not affect the overall continuous operation. Both ends of the Venturi tube 121 are connected to the inlet pipe 111 via branch water pipes 101. This is coordinated with a first electrically controlled valve 102 installed on the branch water pipe 101 and a corresponding valve installed on the inlet pipe 111 corresponding to the Venturi tube 121. The second electrically controlled valve 103 on pipe section 1 controls the flow rate of liquid flowing into the multi-scale bubble generator 12 via the control device 16. Both the microbubble generator and the medium-sized bubble generator are located on the Venturi tube 121 and are interconnected. Combined with the control device 16 controlling the flow rate of gas injected from the air compressor 13 into the air delivery pipe 131 and coordinating with the control of the corresponding branch air pipe 132's outlet electrically controlled valve 133, the flow rate of bubbles at each scale is precisely adjusted. Through the dynamic coupling effect of multi-scale bubble formation combining water-air interface expansion and fluid disturbance, the contact area between water and air is increased, improving cooling efficiency. By setting up a first temperature sensor and a second temperature sensor to collect the inlet and outlet water temperatures of the cooling tower 11, respectively, and installing a temperature and humidity sensor at the air inlet of the cooling tower 11, the control device 16 connected to each sensor calculates the wet-bulb temperature based on the collected dry-bulb temperature and relative humidity using a humidity measurement formula. This calculation is then combined with the inlet and outlet water temperatures of the cooling tower 11 and analyzed using a deep reinforcement learning model. Furthermore, it can also incorporate real-time operating status information of the cooling tower 11, the cooling pump 172, and the air compressor 13 for further analysis. The reinforcement learning model is used for analysis and processing to generate corresponding action commands for adjusting the operating parameters of the cooling pump 172, the operating parameters of the fan of the cooling tower 11, and the bubble generation ratio and zoning strategy of the multi-scale bubble generator 12. This enables intelligent optimization of the operating strategy and the bubble generation and distribution strategy based on real-time operating conditions, environment, and heat load, achieving precise control under all operating conditions. This effectively saves energy without human intervention, improves the adaptability of operation, and can effectively control the heat and mass transfer at the water-air interface at the microscopic level, realizing the intelligence, efficiency, and energy saving of the cooling tower 11.

[0015] Preferably, in some embodiments, the number of multi-scale bubble generators 12 is two, and they are spaced apart along the output direction of the water inlet pipe 111 and connected to the water inlet pipe 111 to divide the water inlet pipe 111 into zones for air distribution. Combined with the control device 16 controlling the flow rate of gas injected by the air compressor 13 into the air delivery pipe 131 and coordinating with the control of the outlet solenoid valve 133 on the corresponding branch air pipe 132, the bubble flow rate of each zone is precisely adjusted to achieve uniform bubble distribution and maximize the heat exchange area. Of course, in other embodiments, the number of multi-scale bubble generators 12 may be one or more.

[0016] Preferably, the first electrically controlled valve 102 is located on the branch water pipe 101 connected to the water inlet end of the corresponding Venturi tube 121, so as to better control the start and stop of liquid flow into the Venturi tube 121.

[0017] Specifically, in this embodiment, the microbubble generating unit is used to generate bubbles with an average diameter ranging from 20 to 200 μm, and the medium-sized bubble generating unit is used to generate bubbles with an average diameter ranging from 0.5 to 2 mm. The average diameter of the bubbles generated by different microbubble generating units may vary, but the average diameter of the generated bubbles is always within the range of 20 to 200 μm. Similarly, the average diameter of the bubbles generated by different medium-sized bubble generating units may vary, but the average diameter of the generated bubbles is always within the range of 0.5 to 2 mm. The specific surface area of ​​the bubbles with an average diameter ranging from 20 to 200 μm provides a larger heat and mass transfer interface, and they can continuously release dissolved gases during their ascent, promoting the renewal and breakup of the water film. The fluid disturbance generated during the ascent of the bubbles with an average diameter ranging from 0.5 to 2 mm can effectively promote the breakup and regeneration of the water film on the packing material, further enhancing heat and mass exchange. The interaction between microbubbles and medium-sized bubbles forms a dynamic coupling cyclic enhancement mechanism to achieve synergistic effects.

[0018] Specifically, the venturi tube 121 is sequentially divided along the output direction into a first ejector section 1211, a first diffuser section 1212, a second ejector section 1213, and a second diffuser section 1214. The microbubble generating unit is located in the first ejector section 1211 and / or the second ejector section 1213, and the medium bubble generating unit is located in the first ejector section 1211 and / or the second ejector section 1213. In conjunction with... Figure 1 and Figure 2The microbubble generating unit is located in the second ejector section 1213, and the medium bubble generating unit is located in the first ejector section 1211. Of course, in some embodiments, the microbubble generating unit may be located in the first ejector section 1211 and the medium bubble generating unit may be located in the second ejector section 1213; or, the microbubble generating unit may be located in both the first and second ejector sections 1211 and the medium bubble generating unit may be located in the first ejector section 1211; or, the medium bubble generating unit may be located in both the first and second ejector sections 1213 and the microbubble generating unit may be located in the first ejector section 1211; or, the microbubble generating unit may be located in both the first and second ejector sections 1211 and the medium bubble generating unit may be located in the second ejector section 1213; or, the medium bubble generating unit may be located in both the first and second ejector sections 1211 and the microbubble generating unit may be located in the second ejector section 1213.

[0019] Preferably, in this embodiment, the first injector section 1211 is provided with at least one first air inlet pipe 14, the air outlet section of the first air inlet pipe 14 passes through the venturi tube 121 and is located in the first injector section 1211, the air outlet section of the first air inlet pipe 14 is provided with first air outlets 141 evenly distributed along the radial direction of the venturi tube 121, the air inlet section of the first air inlet pipe 14 extends out of the venturi tube 121, and the air inlet section of the first air inlet pipe 14 is connected to the corresponding branch air pipe 132; the second injector section 1 At least one second air inlet pipe 15 is provided on 213. The air outlet section of the second air inlet pipe 15 passes through the Venturi tube 121 and is located in the second injector section 1213. The air outlet section of the second air inlet pipe 15 is provided with second air outlets 151 evenly distributed along the radial direction of the Venturi tube 121. The air inlet section of the second air inlet pipe 15 extends out of the Venturi tube 121 and is connected to the corresponding branch air pipe 132. The diameter of the first air outlet 141 is different from the diameter of the second air outlet 151.

[0020] In this embodiment, the intake pipe sections of the first intake pipe 14 and the second intake pipe 15 are respectively connected to the corresponding branch pipes 132, which in turn connect to the gas delivery pipe 131. The control device 16 controls the flow rate of the gas injected into the output pipe by the air compressor 13 and controls the operation of the corresponding outlet electric control valve 133, so that the gas is injected into the Venturi tube 121 through the corresponding branch pipes 132, thereby generating a cavitation effect when the liquid flows through, realizing the injection of bubbles. The size of the bubbles is related to the diameter of the corresponding outlet, and multi-scale bubbles are generated by using outlets of multiple sizes. In this embodiment, the first injector section 1211 is provided with two first intake pipes 14, and the second injector section 1213 is provided with two second intake pipes 15. The diameter of the outlet of the intake pipe on the Venturi tube 121 decreases along the output direction. Of course, in other embodiments, the diameter of the outlet of the intake pipe on the Venturi tube 121 can increase along the output direction.

[0021] Preferably, combined with Figures 1 to 2 The first air outlets 141 located on the same first air intake pipe 14 have the same diameter, while the first air outlets 141 located on different first air intake pipes 14 may have different diameters. Similarly, the second air outlets 151 located on the same second air intake pipe 15 have the same diameter, while the second air outlets 151 located on different second air intake pipes 15 may have different diameters. Of course, in other embodiments, the first air outlets 141 located on different first air intake pipes 14 may have the same diameter, and the second air outlets 151 located on different second air intake pipes 15 may have the same diameter.

[0022] Preferably, when the first air outlet 141 is used to generate bubbles with an average diameter range of 20 to 200 μm, the first air inlet pipe 14 corresponding to the first air outlet 141 is the microbubble generating unit; when the second air outlet 151 is used to generate bubbles with an average diameter range of 20 to 200 μm, the second air inlet pipe 15 corresponding to the second air outlet 151 is the microbubble generating unit; when the first air outlet 141 is used to generate bubbles with an average diameter range of 0.5 to 2 mm, the first air inlet pipe 14 corresponding to the first air outlet 141 is the medium-sized bubble generating unit; when the second air outlet 151 is used to generate bubbles with an average diameter range of 0.5 to 2 mm, the second air inlet pipe 15 corresponding to the second air outlet 151 is the medium-sized bubble generating unit.

[0023] Specifically, the inner diameters of the first diffuser section 1212 and the second diffuser section 1214 both decrease first and then gradually increase along the output direction.

[0024] Specifically, in combination Figure 3In some embodiments, the cooling tower efficiency enhancement system further includes a fan coil unit 19. The input end of the fan coil unit 19 is connected to the evaporator outlet of the chiller 17 via a chilled water transfer pipe 173. A chilled water pump 18 is installed on the chilled water transfer pipe 173. The output end of the fan coil unit 19 is connected to the evaporator inlet of the chiller 17 via a recovery pipe 191. A control device 16 can be electrically connected to the fan coil unit 19 and the chilled water pump 18 to control their operation.

[0025] Specifically, the control method of the cooling tower efficiency enhancement system of the present invention is applied to the above-mentioned cooling tower efficiency enhancement system, and the control method of the cooling tower efficiency enhancement system includes the following steps:

[0026] The control device acquires the real-time inlet and outlet water temperatures of the cooling tower from the corresponding temperature sensors, as well as the real-time wet-bulb temperature, and the operating status information of the cooling tower, cooling pump, and air compressor.

[0027] The control device filters and normalizes the collected real-time data to construct a current state vector. Based on this vector, the real-time heat transfer coefficient and unit cooling energy consumption of the cooling tower are calculated. The collected real-time data includes the real-time wet-bulb temperature, the operating status of the cooling tower, the operating status of the cooling pump, and the operating status of the air compressor, as well as the real-time inlet and outlet water temperatures of the cooling tower obtained from the corresponding temperature sensors. The calculated real-time heat transfer coefficient and unit cooling energy consumption of the cooling tower serve as the core evaluation indicators for control strategy optimization.

[0028] The control device uses a deep reinforcement learning model to generate action vectors based on the current state vector. Based on each element of the action vector, it generates and outputs action commands to adjust the operating parameters of the cooling pump, the cooling tower fan, and the bubble generation ratio and zoning strategy of the multi-scale bubble generator. The deep reinforcement learning model can employ a Deep Deterministic Policy Gradient (DDPG) algorithm based on the Actor-Critic (policy-value) framework, including an Actor (policy) network and a Critic (value) network. The Actor (policy) network selects actions based on the current state, while the Critic (value) network evaluates the actions. The action vectors can include microbubble ratio, medium bubble ratio, the opening degree of the first solenoid valve, the opening degree of the second solenoid valve, the opening degree of the outlet solenoid valve, and the fan speed setpoint, etc.

[0029] The control device calculates the reward corresponding to the action vector based on the action vector and the current state vector, combined with the reward mechanism.

[0030] After the execution of the action command is completed, the control device acquires the inlet and outlet water temperatures of the cooling tower at the next moment from the corresponding temperature sensor, acquires the wet bulb temperature at the next moment, and acquires the working status information of the cooling tower, the cooling pump, and the air compressor at the next moment. After filtering and normalizing the data information at the next moment, the updated state vector is constructed.

[0031] The control device forms samples based on the current state vector, action vector, reward, and updated state vector, stores the samples in the experience replay buffer, randomly samples a preset number of samples at preset time step intervals in the experience replay buffer, evaluates the adjustment based on the average reward monitored by the sampled samples, and trains and updates the deep reinforcement learning model.

[0032] The real-time data may also include the generation ratio of bubbles at various scales, which can be obtained by linearly converting the output valve of the corresponding branch gas pipe. The reward function for the reward mechanism can be based on minimizing unit cooling energy consumption and constraining the cooling tower's outlet water temperature. It is calculated by multiplying the absolute value of the difference between the real-time cooling tower outlet water temperature and the initialized preset outlet water temperature by a preset temperature deviation penalty coefficient. The negative of the sum of unit cooling energy consumption and this product is then used as the corresponding reward value. Unit cooling energy consumption is the ratio of consumed electrical energy to removed heat, which can be obtained by dividing the sum of the cooling tower's fan power and water pump power by the heat exchange rate. The real-time heat transfer coefficient of a cooling tower can be calculated based on a cooling tower heat transfer model. The heat transfer coefficient represents the heat transfer rate per unit area per unit temperature difference. The mass flow rate of the cooling water can be calculated based on the collected flow rate and density of the cooling water. The mass flow rate of the cooling water is calculated as the product of the specific heat capacity of the water and the inlet and outlet temperature difference of the cooling water as the heat transfer. The ratio between the temperature difference between the inlet water temperature and the wet-bulb temperature of the cooling tower and the temperature difference between the outlet water temperature and the wet-bulb temperature of the cooling tower is calculated, and the logarithm of this ratio is obtained. The difference between the temperature difference between the inlet water temperature and the wet-bulb temperature of the cooling tower and the temperature difference between the outlet water temperature and the wet-bulb temperature of the cooling tower is calculated, and this difference is divided by the ratio to obtain the logarithmic mean temperature difference. The logarithmic mean temperature difference is calculated as the product of the theoretical heat transfer area of ​​the cooling tower packing, and the heat transfer is divided by this product to obtain the heat transfer coefficient.

[0033] The control method of the cooling tower efficiency enhancement system controls the flow rate of liquid flowing into the multi-scale bubble generator through a control device. The microbubble generator and the medium-sized bubble generator are both located on venturi tubes and interconnected. The control device also controls the flow rate of gas injected into the air pipe by the air compressor and coordinates with the control of the outlet valves on the corresponding branch air pipes to precisely adjust the flow rate of bubbles at each scale. Through the dynamic coupling effect of multi-scale bubbles forming a water-air interface expansion and fluid disturbance, the contact area between water and air is increased, improving cooling efficiency. A first temperature sensor and a second temperature sensor are installed to collect the inlet and outlet water temperatures of the cooling tower, respectively. A temperature and humidity sensor is also installed at the air inlet of the cooling tower. The control device, connected to each sensor, uses the collected dry-bulb temperature and humidity data to determine the optimal flow rate. The system calculates wet-bulb temperature using a humidity measurement formula and analyzes it using a deep reinforcement learning model, taking into account the inlet and outlet water temperatures of the cooling tower. It also analyzes and processes real-time operating status information of the cooling tower, cooling pump, and air compressor using the same model. This generates action commands to adjust the operating parameters of the cooling pump, the cooling tower fan, and the bubble generation ratio and zoning strategy of the multi-scale bubble generator. This enables intelligent optimization of operating strategies and bubble generation and distribution strategies based on real-time operating conditions, achieving precise control under all operating conditions. This results in effective energy saving without manual intervention, improved operational adaptability, and effective control of heat and mass transfer at the water-air interface at the microscopic level.

[0034] Preferably, in some embodiments, before the step of acquiring the real-time inlet and outlet water temperatures of the cooling tower collected by the temperature sensor through the control device, the method further includes:

[0035] The control device initializes the chiller, cooling tower, first temperature sensor, second temperature sensor, temperature and humidity sensor, first electric control valve, second electric control valve and exhaust electric control valve, and initializes the control parameters, and loads the pre-trained deep reinforcement learning model.

[0036] The initialization of sensors involves zero-point calibration and range checks. Initialization of the chiller and cooling tower involves setting the corresponding frequency converters to soft-start mode. The fan speed is initialized to 75% of its rated speed. The initialization of each electrically controlled valve involves zeroing the valve and then slowly opening it to its initial opening degree, typically 0%. The initialization control parameters can be the weights of the reward function for the initialization reward mechanism.

[0037] Preferably, the operating mode of the cooling tower efficiency enhancement system can also be initialized, including initializing the generation ratio of microbubbles and medium bubbles to 50%, uniformly distributing the opening degree of the electrically controlled valves, and initializing the preset outlet water temperature of the cooling tower and the data acquisition frequency according to the seasonal default settings. The seasonal default setting is to increase the ambient temperature by 2°C, and the data acquisition frequency can be initialized to once every 5 seconds.

[0038] In summary, the cooling tower efficiency enhancement system of the present invention increases the contact area and time between water and air by setting up an inlet pipe connected to the water inlet of the cooling tower and a multi-scale bubble generating device including a Venturi tube, a microbubble generating unit, and a medium-sized bubble generating unit connected to the inlet pipe. This requires no modification to the internal structure of the cooling tower and does not affect the overall continuous operation. Both ends of the Venturi tube are connected to the inlet pipe via branch water pipes. A first electrically controlled valve on the branch water pipe and a second electrically controlled valve on the corresponding section of the inlet pipe control the flow rate of liquid flowing into the multi-scale bubble generating device. The microbubble generating unit and the medium-sized bubble generating unit are both located on the Venturi tube and interconnected. The control device controls the flow rate of gas injected from the air compressor into the gas delivery pipe and controls the corresponding outlet electrically controlled valve on the branch gas pipe to precisely adjust the flow rate of bubbles at each scale. Through the dynamic coupling effect of multi-scale bubbles forming a water-air interface expansion and fluid disturbance, the contact area between water and air is increased, improving cooling efficiency. The system employs a first and a second temperature sensor to collect the inlet and outlet water temperatures of the cooling tower, respectively, and a temperature and humidity sensor at the air inlet. A control device connected to each sensor calculates the wet-bulb temperature using the collected dry-bulb temperature and relative humidity, combined with a humidity measurement formula. This calculation is then analyzed using a deep reinforcement learning model, incorporating the inlet and outlet water temperatures. Furthermore, real-time operating status information from the cooling tower, cooling pump, and air compressor is also analyzed using the same model to generate corresponding action commands for adjusting the operating parameters of the cooling pump, the cooling tower fan, and the bubble generation ratio and zoning strategy of the multi-scale bubble generator. This enables intelligent optimization of operating strategies and bubble generation and distribution strategies based on real-time operating conditions, achieving precise control across all operating conditions. This results in effective energy savings, requires no manual intervention, improves operational adaptability, and allows for effective microscopic control of heat and mass transfer at the water-air interface. The control method of this cooling tower efficiency enhancement system also possesses the aforementioned functions.

[0039] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A cooling tower efficiency enhancement system, characterized in that, include: Refrigeration unit; The cooling tower has an inlet pipe connected to its inlet, and its outlet is connected to the condenser inlet of the chiller via a return pipe. The condenser outlet of the chiller is connected to the inlet pipe via a cooling water transmission pipe and a cooling pump. A first temperature sensor is installed on the inlet pipe near the inlet of the cooling tower, and a second temperature sensor is installed on the return pipe near the outlet of the cooling tower. A temperature and humidity sensor is installed at the air inlet of the cooling tower. A multi-scale bubble generator, connected to the water inlet pipe, includes a Venturi tube, at least one microbubble generating unit, and at least one medium bubble generating unit. Both ends of the Venturi tube are connected to the water inlet pipe via branch water pipes. A first electrically controlled valve is provided on one of the branch water pipes, and a second electrically controlled valve is provided on the section of the water inlet pipe corresponding to the Venturi tube. The microbubble generating unit and the medium bubble generating unit are both located on the Venturi tube, and the microbubble generating unit is connected to the medium bubble generating unit. An air compressor, the air outlet of which is connected to an air supply pipe, the air supply pipe being connected to the microbubble generating unit and the medium bubble generating unit respectively through branch air pipes, and an air outlet electric control valve being provided on the branch air pipe; The control device is electrically connected to the chiller, the cooling tower, the cooling pump, the first temperature sensor, the second temperature sensor, the temperature and humidity sensor, the first electrically controlled valve, the second electrically controlled valve, the outlet electrically controlled valve, and the air compressor. It is used to analyze and process real-time operating status information of the cooling pump, the cooling tower, and the air compressor, as well as the inlet and outlet water temperatures of the cooling tower, combined with real-time wet-bulb temperatures, using a deep reinforcement learning model. This generates corresponding action commands to adjust the operating parameters of the cooling pump, the cooling tower fan, and the bubble generation ratio and zoning strategy of the multi-scale bubble generator. The device also controls the operation of the chiller, the cooling tower, the cooling pump, the first temperature sensor, the second temperature sensor, the temperature and humidity sensor, the first electrically controlled valve, the second electrically controlled valve, the air compressor, and the outlet electrically controlled valve. The Venturi tube is sequentially divided into a first ejector section, a first diffuser section, a second ejector section, and a second diffuser section along the output direction. The microbubble generating unit is located in the first ejector section and / or the second ejector section, and the medium bubble generating unit is located in the first ejector section and / or the second ejector section. The inner diameters of both the first and second diffuser sections decrease first and then gradually increase along the output direction.

2. The cooling tower efficiency enhancement system according to claim 1, characterized in that, The microbubble generating unit is used to generate bubbles with an average diameter ranging from 20 to 200 μm, and the medium bubble generating unit is used to generate bubbles with an average diameter ranging from 0.5 to 2 mm.

3. The cooling tower efficiency enhancement system according to claim 1, characterized in that, The first injector section is provided with at least one first air inlet pipe. The air outlet section of the first air inlet pipe passes through the venturi tube and is located in the first injector section. First air outlets are evenly distributed along the radial direction of the venturi tube on the air outlet section of the first air inlet pipe. The air inlet section of the first air inlet pipe extends out of the venturi tube and is connected to a corresponding branch air pipe. The second injector section is provided with at least one second air inlet pipe. The air outlet section of the second air inlet pipe passes through the venturi tube and is located in the second injector section. Second air outlets are evenly distributed along the radial direction of the venturi tube on the air outlet section of the second air inlet pipe. The air inlet section of the second air inlet pipe extends out of the venturi tube and is connected to a corresponding branch air pipe. The diameter of the first air outlet is different from the diameter of the second air outlet.

4. The cooling tower efficiency enhancement system according to claim 3, characterized in that, The diameter of the first air outlet located on the same first air inlet pipe is the same, the diameter of the first air outlet located on different first air inlet pipes is different, the diameter of the second air outlet located on the same second air inlet pipe is the same, and the diameter of the second air outlet located on different second air inlet pipes is different.

5. The cooling tower efficiency enhancement system according to claim 3, characterized in that, When the first air outlet is used to generate bubbles with an average diameter range of 20–200 μm, the first air inlet pipe corresponding to the first air outlet is the microbubble generating unit; when the second air outlet is used to generate bubbles with an average diameter range of 20–200 μm, the second air inlet pipe corresponding to the second air outlet is the microbubble generating unit; when the first air outlet is used to generate bubbles with an average diameter range of 0.5–2 mm, the first air inlet pipe corresponding to the first air outlet is the medium-sized bubble generating unit; when the second air outlet is used to generate bubbles with an average diameter range of 0.5–2 mm, the second air inlet pipe corresponding to the second air outlet is the medium-sized bubble generating unit.

6. The cooling tower efficiency enhancement system according to claim 1, characterized in that, The number of multi-scale bubble generators is two.

7. The cooling tower efficiency enhancement system according to claim 1, characterized in that, The cooling tower efficiency enhancement system also includes a fan coil unit. The input end of the fan coil unit is connected to the evaporator outlet of the chiller through a chilled water transmission pipe. A chilled water pump is installed on the chilled water transmission pipe. The output end of the fan coil unit is connected to the evaporator inlet of the chiller through a recovery pipe.

8. A control method for a cooling tower efficiency enhancement system, characterized in that, The cooling tower efficiency enhancement system according to any one of claims 1-7, wherein the control method of the cooling tower efficiency enhancement system comprises the following steps: The control device acquires the real-time inlet and outlet water temperatures of the cooling tower from the corresponding temperature sensors, as well as the real-time wet-bulb temperature, and the operating status information of the cooling tower, cooling pump, and air compressor. The control device filters and normalizes the collected real-time data to construct the current state vector, and calculates the real-time heat transfer coefficient and unit cooling energy consumption of the cooling tower based on the current state vector. The control device uses a deep reinforcement learning model to generate action vectors based on the current state vector. Based on each element of the action vector, it generates and outputs action commands to adjust the operating parameters of the cooling pump, the operating parameters of the cooling tower fan, and the bubble generation ratio and zoning strategy of the multi-scale bubble generator. The control device calculates the reward corresponding to the action vector based on the action vector and the current state vector, combined with the reward mechanism. After the execution of the action command is completed, the control device acquires the inlet and outlet water temperatures of the cooling tower at the next moment from the corresponding temperature sensor, acquires the wet bulb temperature at the next moment, and acquires the working status information of the cooling tower, the cooling pump, and the air compressor at the next moment. After filtering and normalizing the data information at the next moment, the updated state vector is constructed. The control device forms samples based on the current state vector, action vector, reward, and updated state vector, stores the samples in the experience replay buffer, randomly samples a preset number of samples at preset time step intervals in the experience replay buffer, evaluates the adjustment based on the average reward monitored by the sampled samples, and trains and updates the deep reinforcement learning model.

Images