Control method for indirect evaporative cooling system, and system

Abstract

The present application provides a control method for an indirect evaporative cooling system, and a system. The control method comprises: establishing an indirect evaporative cooling system cooling model and an indirect evaporative cooling system operating power consumption model; obtaining training data; training the indirect evaporative cooling system cooling model and the indirect evaporative cooling system operating power consumption model on the basis of the training data, to obtain a determined indirect evaporative cooling system cooling model and a determined indirect evaporative cooling system operating power consumption model; and optimizing the operating cost of the indirect evaporative cooling system at least on the basis of the determined models, to obtain data of the output parameters of the indirect evaporative cooling system cooling model corresponding to the optimized operating cost, thereby controlling the operation of the indirect evaporative cooling system. The present application can ensure that the operating cost of the indirect evaporative cooling system is optimal under different climates and at different locations, and the operation of the indirect evaporative cooling system is controlled on the basis of indoor supply air temperature and indoor return air temperature, thereby quickly and accurately ensuring that the cooling capacity meets requirements of a unit to be cooled.

Description

Control methods and systems for indirect evaporative cooling systems Technical Field

[0001] This application relates to indirect evaporative cooling systems, and more particularly to control methods and systems for indirect evaporative cooling systems. Background Technology

[0002] To keep pace with the global trend of digitalization, data centers are developing at an unprecedented pace, making their energy consumption an increasingly significant issue. For example, research indicates that data centers account for 1.5% of total societal electricity consumption, with air conditioning systems alone accounting for nearly 40% of this energy consumption. This makes data center energy usage efficiency (PUE) a key factor influencing their performance, making energy conservation and consumption reduction a primary consideration in the initial planning stages of data centers. Indirect evaporative cooling solutions, which utilize natural cooling sources to improve energy efficiency, have been demonstrated and applied in numerous large-scale data centers both domestically and internationally due to their energy efficiency and high applicability. Summary of the Invention

[0003] The inventors observed that current indirect evaporative cooling systems / units have relatively traditional control logic. For example, they activate different operating modes based on the outdoor air's dry-bulb and wet-bulb temperatures. More specifically, when the outdoor dry-bulb temperature is below a first set value, the outdoor fan system is activated, allowing heat exchange only between the air inside the unit being cooled (e.g., the machine room) and the outdoor air (outside the unit being cooled) through the air-to-air heat exchange core. When the outdoor dry-bulb temperature is above the first set value and the wet-bulb temperature is below a second set value, a water mist spraying system is activated to spray water or water mist onto the air-to-air heat exchange core, lowering the outdoor air temperature before heat exchange with the air inside the unit being cooled. When the outdoor wet-bulb temperature is above the second set value, meaning the spraying mode cannot meet the temperature requirements of the unit being cooled, the direct expansion refrigeration system is activated to further reduce the air temperature of the unit being cooled.

[0004] Currently, most indirect evaporative cooling systems / units on the market have adopted variable frequency fans and compressors to achieve energy saving and consumption reduction. However, the inventors discovered that the traditional control logic of indirect evaporative cooling systems / units only relies on the outdoor air dry and wet bulb temperatures to activate different operating modes, which has the following drawbacks: Firstly, my country's climate distribution is diverse, and this system does not take into account climate diversity, nor does it consider the differences in electricity and water costs at different locations, thus failing to guarantee the lowest total energy consumption / cost for system operation; secondly, the heat exchange efficiency of air-to-air heat exchangers in wet mode is related to outdoor air parameters. Activating the direct expansion refrigeration system based on a predetermined wet bulb temperature will result in the cooling capacity failing to meet the requirements of the cooled unit as the outdoor temperature increases.

[0005] To address the aforementioned problems, this application provides a control method and system for an indirect evaporative cooling system. Specifically, this application acquires multiple sets of control parameter data to control the indirect evaporative cooling system, meeting the requirements of the cooled unit. Based on these multiple sets of control parameter data, the operating cost of the indirect evaporative cooling system is optimized, ensuring that the total operating cost of the indirect evaporative cooling system is optimal under a certain set of control parameter data. In optimizing the operating cost of the indirect evaporative cooling system, this application considers the total cost of electricity and water consumption under different climates and locations, thereby ensuring that the total operating cost of the indirect evaporative cooling system is always optimal (e.g., minimum) under different climates and locations. In different climates, such as different regions, the unit price of electricity and water may differ (even significantly), thus even the same set of control parameter data can result in different (even significantly different) operating costs for the indirect evaporative cooling system. For example, an energy-saving mode may be suitable for areas with abundant water resources, while a water-saving mode may be suitable for areas with scarce water resources, depending on the total cost of electricity and water consumption in those areas. This application considers the total operating cost of the indirect evaporative cooling system under different climates and locations, while also taking into account that the control parameter data meets the needs of the cooled unit.

[0006] Furthermore, this application controls the operation of the indirect evaporative cooling system to quickly and accurately meet the needs of the cooled unit, such as its cooling capacity requirement, based at least on the indoor supply air temperature provided by the indirect evaporative cooling system to the cooled unit and the indoor return air temperature returning from the cooled unit to the indirect evaporative cooling system. This application uses the indoor supply air temperature output by the indirect evaporative cooling system as a monitoring target to ensure that the output meets the temperature requirement received by the cooled unit, and uses the indoor return air temperature returning from the cooled unit to the indirect evaporative cooling system as a monitoring target to determine whether the cooled unit has actually met its temperature requirement. When both the indoor supply air temperature and the indoor return air temperature are detected to meet the requirements, it indicates that the indirect evaporative cooling system has accurately provided the required cooling capacity to the cooled unit; otherwise, the actual cooling capacity required by the cooled unit has not been met. For example, when the indoor supply air temperature output by the indirect evaporative cooling system does not meet the temperature requirement received by the cooled unit, it may indicate that the operation of the indirect evaporative cooling system is inappropriate, and therefore, the operation of the indirect evaporative cooling system needs to be adjusted. When the indoor return air temperature from the cooled unit to the indirect evaporative cooling system does not meet the actual temperature requirement of the cooled unit, it may indicate an increase in the actual load on the cooled unit or a component failure in the cooled unit. Therefore, it is necessary to adjust the operation of the indirect evaporative cooling system accordingly. Based on the monitored indoor supply air temperature and indoor return air temperature, this application can adjust the operation of the indirect evaporative cooling system accordingly to accurately meet the cooling capacity requirements of the cooled unit.

[0007] More specifically, according to a first aspect of this application, this application provides a control method for an indirect evaporative cooling system. The control method includes the following steps S1-S4. In step S1, a refrigeration model and an operating power consumption model of the indirect evaporative cooling system are created. The refrigeration model includes input parameters and output parameters, and the operating power consumption model includes input parameters and output parameters. The output parameters of the refrigeration model include control parameters of the indirect evaporative cooling system. The output parameters of the operating power consumption model include the operating power consumption of various parts of the indirect evaporative cooling system. In step S2, training data is acquired, including data on the input and output parameters of the refrigeration model and the operating power consumption model. In step S3, the refrigeration model and the operating power consumption model are trained based on the acquired training data to obtain a determined refrigeration model and an determined operating power consumption model for the indirect evaporative cooling system. In step S4, the operating cost of the indirect evaporative cooling system is optimized based at least on the determined refrigeration model and the determined power consumption model of the indirect evaporative cooling system, so as to obtain the output parameter data of the refrigeration model of the indirect evaporative cooling system corresponding to the optimization of the operating cost of the indirect evaporative cooling system, thereby controlling the operation of the indirect evaporative cooling system.

[0008] According to a first aspect of this application, the output parameters of the refrigeration model of the indirect evaporative cooling system are used as input parameters of the power consumption model of the indirect evaporative cooling system.

[0009] According to the first aspect of this application, the input parameters of the refrigeration model of the indirect evaporative cooling system include the indoor supply air temperature and the indoor return air temperature.

[0010] According to the first aspect of this application, in step S4, steps S4.1-S4.3 are performed to obtain data on the output parameters of the indirect evaporative cooling system refrigeration model corresponding to the optimization of the operating cost of the indirect evaporative cooling system. In step S4.1, field data of the input parameters of the indirect evaporative cooling system refrigeration model are obtained. In step S4.2, the obtained field data is input to the determined indirect evaporative cooling system refrigeration model to obtain a dataset of the output parameters of the indirect evaporative cooling system refrigeration model. In step S4.3, the obtained dataset of the output parameters of the indirect evaporative cooling system refrigeration model is input to the indirect evaporative cooling system operating cost optimization model to obtain data on the output parameters of the indirect evaporative cooling system refrigeration model corresponding to the optimization of the operating cost of the indirect evaporative cooling system.

[0011] According to a first aspect of this application, the indirect evaporative cooling system operating cost optimization model is configured to: obtain an operating power consumption set composed of the operating power consumption of each part of the indirect evaporative cooling system based on the dataset of output parameters of the refrigeration model of the indirect evaporative cooling system and the operating power consumption model of the indirect evaporative cooling system; obtain an operating cost set of the indirect evaporative cooling system based on the obtained operating power consumption set; and optimize the obtained operating cost set to obtain the data of the output parameters of the refrigeration model of the indirect evaporative cooling system corresponding to the operating cost optimization of the indirect evaporative cooling system.

[0012] According to a first aspect of this application, the operating power consumption of each part of the indirect evaporative cooling system includes electrical power consumption and water consumption, and the operating cost of the indirect evaporative cooling system includes the electrical cost generated by the electrical power consumption and the water cost generated by the water consumption.

[0013] According to a first aspect of this application, the indirect evaporative cooling system is configured to cool a unit. The indirect evaporative cooling system includes a water mist spraying system. Output parameters of the refrigeration model of the indirect evaporative cooling system include the rotational speed of the nozzle moving motor of the water mist spraying system, the nozzle moving motor being configured to control the movement of the nozzles of the water mist spraying system. The control method further includes obtaining a motor speed limit model. The motor speed limit model is configured to obtain a speed limit of the nozzle moving motor after the outdoor air is sprayed by the water mist spraying system based on the outdoor air dry-bulb temperature and outdoor air dry-air humidity of the unit being cooled. In step S4, the speed limit of the nozzle moving motor after the outdoor air is sprayed by the water mist spraying system at the current outdoor air dry-bulb temperature and current outdoor air dry-air humidity is obtained, and the speed of the nozzle moving motor when the outdoor air is sprayed by the water mist spraying system is limited within the corresponding speed limit of the nozzle moving motor to optimize the operating cost of the indirect evaporative cooling system.

[0014] According to the first aspect of this application, the motor speed limit model is obtained by the following operations: creating and obtaining a temperature and humidity model between the humidity content and air temperature of outdoor air after being sprayed by the water mist system, wherein the temperature and humidity model uses the current outdoor dry-bulb temperature and the current outdoor dry air humidity content as constants; creating and obtaining a temperature and humidity limit model between the humidity limit and air temperature limit when the outdoor air is saturated after being sprayed by the water mist system; creating and obtaining a speed and humidity limit model between the speed limit of the nozzle moving motor and the humidity limit when the air is saturated; and combining the obtained temperature and humidity model, the temperature and humidity limit model, and the speed and humidity limit model to obtain the motor speed limit model.

[0015] According to a first aspect of this application, the indirect evaporative cooling system further includes a direct expansion refrigeration system and a fan system. The input parameters of the refrigeration model of the indirect evaporative cooling system include the target parameters of the cooled unit and the operating parameters of the indirect evaporative cooling system. The output parameters of the refrigeration model of the indirect evaporative cooling system include control parameters for operating the water mist system, the direct expansion refrigeration system, and the fan system. The output parameters of the power consumption model of the indirect evaporative cooling system include the power consumption and water consumption of the water mist system, the power consumption of the direct expansion refrigeration system, and the power consumption of the fan system. The operating cost of the indirect evaporative cooling system includes the sum of the power and water costs of the water mist system, the power costs of the direct expansion refrigeration system, and the power costs of the fan system.

[0016] According to a first aspect of this application, the direct expansion refrigeration system includes a compressor, a condenser, an evaporator, and a flow regulating valve. The fan system includes an indoor fan and an outdoor fan. The target parameter of the cooled unit is the target temperature of the cooled unit. The operating parameters of the indirect evaporative cooling system include the outdoor air temperature, outdoor air humidity, indoor supply air temperature, indoor supply air humidity, indoor return air temperature, and indoor return air humidity of the cooled unit. The control parameters for operating the water mist system include the rotational speed of the nozzle moving motor; the control parameters for operating the direct expansion refrigeration system include the rotational speed of the compressor and the opening degree of the flow regulating valve; and the control parameters for operating the fan system include the fan speed of the outdoor fan. The power consumption of the water mist system includes the power consumption of the nozzle moving motor; the water consumption of the water mist system includes the water volume sprayed by the nozzle; the power consumption of the direct expansion refrigeration system includes the power consumption of the compressor; and the power consumption of the fan system includes the power consumption of the indoor fan and the outdoor fan. The electricity cost of the water spray system includes the electricity cost generated by the power consumption of the nozzle moving motor; the water cost of the water spray system includes the water cost generated by the water volume of the nozzle; the electricity cost of the direct expansion refrigeration system includes the electricity cost generated by the power consumption of the compressor; and the electricity cost of the fan system includes the electricity cost generated by the power consumption of the indoor fan and the outdoor fan.

[0017] According to the first aspect of this application, both the refrigeration model of the indirect evaporative cooling system and the power consumption model of the indirect evaporative cooling system are fully connected deep neural network models.

[0018] According to the first aspect of this application, the backpropagation algorithm and the gradient descent optimization algorithm are used to train the refrigeration model of the indirect evaporative cooling system and the power consumption model of the indirect evaporative cooling system.

[0019] According to a first aspect of this application, the indirect evaporative cooling system operating cost optimization model is configured to use a genetic algorithm to optimize the operating cost of the indirect evaporative cooling system, wherein the fitness function in the genetic algorithm is obtained based on the determined indirect evaporative cooling system operating power consumption model.

[0020] According to the first aspect of this application, the indirect evaporative cooling system is applied to a test bench, and data obtained from the test bench is used as training data to train the refrigeration model and the power consumption model of the indirect evaporative cooling system, thereby obtaining a determined refrigeration model and the determined power consumption model of the indirect evaporative cooling system. The indirect evaporative cooling system is then applied to a cooled unit in the field, and data obtained from the cooled unit in the field is used as further training data to further train the determined refrigeration model and the determined power consumption model of the indirect evaporative cooling system, thereby adjusting the determined refrigeration model and the determined power consumption model of the indirect evaporative cooling system, and thus obtaining the final determined refrigeration model and the power consumption model of the indirect evaporative cooling system.

[0021] According to a second aspect of this application, a computing system is provided. The computing system includes a processor configured to execute the aforementioned control method to obtain a determined refrigeration model of an indirect evaporative cooling system and a determined operating power consumption model of the indirect evaporative cooling system.

[0022] According to a third aspect of this application, a control system is provided. The control system includes an indirect evaporative cooling system module and an indirect evaporative cooling system operating cost optimization module. The indirect evaporative cooling system operating cost optimization module is connected to the indirect evaporative cooling system module. The indirect evaporative cooling system operating cost optimization module is configured to execute the aforementioned control method to optimize the operating cost of the indirect evaporative cooling system based on the current operating condition data of the indirect evaporative cooling system and the indirect evaporative cooling system module, thereby obtaining control data of the indirect evaporative cooling system corresponding to the optimized operating cost, and thus controlling the operation of the indirect evaporative cooling system. The indirect evaporative cooling system module includes a determined indirect evaporative cooling system refrigeration model and a determined indirect evaporative cooling system operating power consumption model obtained through the aforementioned control method. Alternatively, the indirect evaporative cooling system module and the indirect evaporative cooling system operating cost optimization module each include a determined indirect evaporative cooling system refrigeration model and a determined indirect evaporative cooling system operating power consumption model obtained through the aforementioned control method.

[0023] According to a third aspect of this application, the obtained determined refrigeration model and the determined operating power consumption model of the indirect evaporative cooling system are deployed to the indirect evaporative cooling system module using the aforementioned calculation system. Alternatively, the obtained determined refrigeration model and the determined operating power consumption model of the indirect evaporative cooling system are deployed to the indirect evaporative cooling system module and the indirect evaporative cooling system operating cost optimization module, respectively, using the aforementioned calculation system.

[0024] According to a third aspect of this application, the control system includes a processor configured to execute the aforementioned control method to obtain a determined refrigeration model of the indirect evaporative cooling system and a determined operating power consumption model of the indirect evaporative cooling system, and to deploy the obtained determined refrigeration model of the indirect evaporative cooling system and the determined operating power consumption model of the indirect evaporative cooling system to the indirect evaporative cooling system module, or to deploy them respectively to the indirect evaporative cooling system module and the indirect evaporative cooling system operating cost optimization module.

[0025] According to a fourth aspect of this application, an indirect evaporative cooling system is provided. The indirect evaporative cooling system includes a water mist system, a direct expansion refrigeration system, a fan system, a detection device, and a controller. The detection device is connected to at least one of the water mist system, the direct expansion refrigeration system, and the fan system to detect operating condition data of the at least one. The controller is configured to control the operation of at least one of the water mist system, the direct expansion refrigeration system, and the fan system based on a control signal received from the aforementioned control system. The control system is configured to optimize the operating cost of the indirect evaporative cooling system based on the operating condition data detected by the detection device to obtain control data of the indirect evaporative cooling system corresponding to the optimized operating cost, thereby generating the control signal. Attached Figure Description

[0026] The accompanying drawings are not to scale. In the drawings, each identical or nearly identical component shown in different figures is indicated by the same reference numerals. For clarity, not every component may be labeled in every drawing. In the drawings:

[0027] Figure 1 shows a block diagram of the overall system structure for controlling the indirect evaporative cooling system according to this application;

[0028] Figure 2 shows a block diagram of an embodiment of the indirect evaporative cooling system and detection device shown in Figure 1;

[0029] Figure 3 shows a general flowchart of an embodiment of the control method for the indirect evaporative cooling system shown in Figure 1;

[0030] Figure 4 shows a block diagram of an embodiment of the refrigeration model and power consumption model of the indirect evaporative cooling system created in step 304 of Figure 3;

[0031] Figure 5 shows a detailed flowchart of an embodiment of step 308 shown in Figure 3;

[0032] Figure 6 shows a detailed flowchart of one embodiment of step 310 shown in Figure 3;

[0033] Figure 7 shows a detailed flowchart of an embodiment of step 606 shown in Figure 6;

[0034] Figure 8 shows a flowchart of an embodiment of the method for obtaining a motor speed limit model;

[0035] Figure 9 shows the enthalpy-humidity diagram of air;

[0036] Figure 10 shows the block diagram structure of the PID control of the controller shown in Figure 1; and

[0037] Figure 11 shows a block diagram of the computing system and control system shown in Figure 1. Detailed Implementation

[0038] Various specific embodiments of this application will now be described with reference to the accompanying drawings, which form part of this specification. It should be understood that, where possible, the same or similar reference numerals used in this application refer to the same parts.

[0039] Figure 1 shows a block diagram of the overall system structure for controlling an indirect evaporative cooling system according to this application. This application creates and determines a refrigeration model and an operating power consumption model for the indirect evaporative cooling system. Based on the determined refrigeration model and operating power consumption model, and field data of the indirect evaporative cooling system 100, array control parameter data of the indirect evaporative cooling system 100 is obtained. Based on the obtained array control parameter data, the operating cost of the indirect evaporative cooling system 100 is optimized to obtain optimized operating cost (e.g., optimal / lowest total operating cost) control parameter data of the indirect evaporative cooling system 100, such as a set of control parameter data, thereby controlling the operation of the indirect evaporative cooling system 100.

[0040] As shown in Figure 1, the overall system for controlling the indirect evaporative cooling system includes an indirect evaporative cooling system 100, a computing system 110, a control system 111, and a detection device 112. The indirect evaporative cooling system 100 is configured to cool a unit 106. In one embodiment, the unit 106 is a data center, such as a server room. In other embodiments, the unit 106 includes other suitable units that need to be cooled by the indirect evaporative cooling system. The indirect evaporative cooling system 100 includes a controller 101 and a unit assembly 102. The controller 101 is configured to control the operation of the unit assembly 102 to provide the required cooling capacity to the unit 106. The unit assembly 102 includes a water mist system 103, a direct expansion cooling system 104, and a fan system 105. An embodiment of the specific structure of the unit assembly 102 is detailed in Figure 2.

[0041] The detection device 112 is connected to the indirect evaporative cooling system 100 and configured to detect and acquire the operating parameters of the indirect evaporative cooling system 100. The detection device 112 includes several sensors, such as a temperature sensor, a humidity sensor, a speed sensor, an opening degree sensor, a flow sensor, etc. In some embodiments, the detection device 112 is located outside the indirect evaporative cooling system 100. In some embodiments, the detection device 112 is part of the indirect evaporative cooling system 100.

[0042] The computing system 110 is configured to create a refrigeration model and an operating power consumption model for an indirect evaporative cooling system, acquire training data, and train the refrigeration model and operating power consumption model based on the acquired training data to obtain a determined refrigeration model and an operating power consumption model for the indirect evaporative cooling system. The computing system 110 is connected to the detection device 112 via a connection line 115 and is configured to acquire operating parameter data from the detection device 112 via the connection line 115, and use this data as training data. The computing system 110 is connected to the control system 111 via a connection line 113 and is configured to send the obtained determined refrigeration model and operating power consumption model for the indirect evaporative cooling system as a module package to the control system 111 via the connection line 113. In one embodiment, the indirect evaporative cooling system 100 is applied to a test bench, which includes a unit 106 to be cooled. At the test bench, the indirect evaporative cooling system 100 operates to cool the unit 106. This application acquires operating condition parameter data of the indirect evaporative cooling system 100 during operation as training data. The computing system 110 uses this acquired training data to train a refrigeration model and a power consumption model for the indirect evaporative cooling system, thereby obtaining a determined refrigeration model and a determined power consumption model for the indirect evaporative cooling system. In one embodiment, the computing system 110 is a computer. In other embodiments, the computing system 110 includes other suitable devices or equipment. In one embodiment, the overall system for controlling the indirect evaporative cooling system does not include the computing system 110; instead, the control system 111 performs the operations of the computing system 110 in this embodiment.

[0043] Considering that the determined refrigeration model and power consumption model of the indirect evaporative cooling system obtained from the test bench may deviate when the indirect evaporative cooling system 100 is applied to the user's actual cooling unit 106, this application further obtains the operating condition parameter data of the indirect evaporative cooling system 100 when it is applied to the user's actual cooling unit 106 to further train the model, so as to obtain an indirect evaporative cooling system refrigeration model and power consumption model adapted to the field of the cooled unit 106. Specifically, after obtaining the determined refrigeration model and power consumption model of the indirect evaporative cooling system from the test bench, this application further applies the indirect evaporative cooling system 100 to the user's actual cooling unit 106 in the field, and also applies the detection device 112 to the field. The detection device 112 obtains the operating condition parameter data of the indirect evaporative cooling system 100 at the cooled unit 106 as further training data. The computing system 110 further trains the determined indirect evaporative cooling system refrigeration model and the determined indirect evaporative cooling system operating power consumption model obtained on the test bench based on the further training data, and adjusts (e.g., fine-tunes) the determined indirect evaporative cooling system refrigeration model and the determined indirect evaporative cooling system operating power consumption model to obtain the final determined indirect evaporative cooling system refrigeration model and the indirect evaporative cooling system operating power consumption model.

[0044] The control system 111 is configured to optimize the operating cost of the indirect evaporative cooling system 100 based at least on a determined refrigeration model and a determined power consumption model of the indirect evaporative cooling system, to obtain control parameter data of the indirect evaporative cooling system 100 for optimizing its operating cost, thereby controlling the operation of the indirect evaporative cooling system 100. The control system 111 is connected to the detection device 112 via a connection line 116 and is configured to acquire real-time operating condition parameter data from the detection device 112 via the connection line 116, and use this data as field data. The control system 111 is connected to the indirect evaporative cooling system 100 via a connection line 114 and is configured to send the obtained control parameter data of the indirect evaporative cooling system 100 as a control signal to the indirect evaporative cooling system 100, such as its controller 101, via the connection line 114 to control the operation of the indirect evaporative cooling system 100. In one embodiment, the control parameter data of the indirect evaporative cooling system 100 is the output parameter data of the corresponding indirect evaporative cooling system refrigeration model. In other embodiments, the control parameter data of the indirect evaporative cooling system 100 includes other suitable parameter data. The operating cost of the indirect evaporative cooling system 100 includes electricity costs and water costs.

[0045] In optimizing the operating cost of the indirect evaporative cooling system 100, this application considers the total cost of electricity and water consumption under different climates and locations, thereby ensuring that the total operating cost of the indirect evaporative cooling system 100 is optimal, i.e., the lowest. Simultaneously, this application controls the operation of the water mist system 103, the direct expansion refrigeration system 104, and the fan system 105 (see Figure 2) of the indirect evaporative cooling system 100 based at least on the indoor supply air temperature provided by the indirect evaporative cooling system 100 to the cooled unit 106 and the indoor return air temperature returning from the cooled unit 106 to the indirect evaporative cooling system 100. This ensures that the indirect evaporative cooling system 100 operates in the required mode, thereby accurately guaranteeing that the cooling capacity meets the needs of the cooled unit 106.

[0046] Figure 2 shows a block diagram of an embodiment of the indirect evaporative cooling system 100 and the detection device 112 shown in Figure 1.

[0047] As shown in Figure 2, the unit components 102 of the indirect evaporative cooling system 100 include a water mist system 103, a direct expansion refrigeration system 104, and a fan system 105. The fan system 105 includes an indoor fan 18 and an outdoor fan 19.

[0048] The water mist system 103 includes an air-to-air heat exchanger 1, a hose connector 2, a water mist component support plate 10, a nozzle movement motor 11, a chain 12, a nozzle 13, a solenoid valve 14, a pressure regulating valve 15, a water pipe 16, a water pipe interface 21, and a slide bar 22. One end of the water pipe 16 is connected to an external water supply system (soft water) via the water pipe interface 21, and the other end is connected to a spiral hose 17. The solenoid valve 14 and the pressure regulating valve 15 are fixedly connected in sequence in the middle of the water pipe 16. The spiral hose 17 supplies water to the nozzle 13 through the hose connector 2. The solenoid valve 14 is configured to control the start and stop of the water mist system 103, and the pressure regulating valve 15 is configured to maintain a constant working pressure of the nozzle 13. The hose connector 2 is also connected to the slide bar 22, the other side of which is connected to the chain 12, and the other side of the chain 12 is connected to the nozzle movement motor 11. The nozzle movement motor 11 is configured to control the movement of the nozzle 13. The nozzle moving motor 11 controls the operation of the hose connector 2 and the nozzle 13 by manipulating the slide bar 22 via the chain 12. On one hand, water mist is sprayed onto the surface of the air-to-air heat exchanger 1, reducing the temperature of the outdoor air (secondary air) through evaporative cooling. The air-to-air heat exchanger 1 exchanges heat with the indoor return air (air returning from the cooled unit, such as the machine room, to the air-to-air heat exchanger 1, i.e., primary air), reducing the temperature of the primary air. Simultaneously, the water film on the surface of the air-to-air heat exchanger 1 also further reduces the temperature of the primary air through evaporative cooling. On the other hand, controlling the dwell time of the slide bar 22 controls the working time of the nozzle 13, i.e., the amount of water mist sprayed. Components such as the nozzle moving motor 11, solenoid valve 14, pressure regulating valve 15, and water pipe 16 are fixed on the water mist spray assembly support plate 10.

[0049] The direct expansion refrigeration system 104 (also known as the DX system) includes an evaporator 3, a gas-liquid separator 5, a compressor 6, an oil separator 7, a condenser 4, a liquid receiver 8, a flow control valve 9, and a refrigerant line 23. In one embodiment, the flow control valve 9 is an expansion valve / capillary tube. In other embodiments, the flow control valve 9 includes other suitable structures. The outlet of the evaporator 3 is sequentially connected to the gas-liquid separator 5, the compressor 6, the oil separator 7, the condenser 4, the liquid receiver 8, and the flow control valve 9 via the refrigerant line 23, and the outlet of the flow control valve 9 is connected to the inlet of the evaporator 3. In the direct expansion refrigeration system 104, liquid refrigerant directly evaporates in the coil of the evaporator 3, absorbing heat from the indoor return air (primary air) outside the coil, lowering the temperature of the primary air, and achieving refrigeration. After absorbing heat, the refrigerant becomes low-temperature, low-pressure refrigerant vapor. This vapor is then adiabatically compressed by compressor 6, becoming high-temperature, high-pressure superheated vapor. It then enters condenser 4 for constant-pressure cooling, releasing heat to the secondary air (outdoor air). The refrigerant is cooled into subcooled liquid refrigerant, which is then adiabatically throttled by flow regulating valve 9 to become low-pressure refrigerant. Finally, it flows into evaporator 3 to complete one refrigeration cycle. Gas-liquid separator 5 is configured to prevent liquid refrigerant leaving evaporator 3 from entering compressor 6, thus avoiding liquid slugging. Oil separator 7 is configured to separate the lubricating oil mixed with refrigerant vapor at the outlet of compressor 6 and uses its own control device to return the lubricating oil to the compressor's oil collection tank. Receiver 8 is configured to compensate for condenser liquid level changes caused by load variations and to separate liquid and gaseous refrigerant at the outlet of condenser 4.

[0050] Figure 2 shows the layout of the detection device 112 installed at the indirect evaporative cooling system 100. As shown in Figure 2, an outdoor air intake temperature and humidity sensor A is installed at the inlet of outdoor air (secondary air) and before the low-pressure water mist system 103. This sensor is configured to detect the outdoor air temperature T. A outdoor air humidity (RH) A A temperature sensor B is installed between the inlet of the air-to-air heat exchanger 1 and the inlet of the condenser 4. This sensor is configured to detect the temperature of the outdoor air (secondary air) at the inlet of the condenser 4. A condenser temperature sensor C is installed between the outlet of the condenser 4 and the outdoor fan 19. This sensor is configured to detect the temperature of the outdoor air (secondary air) at the outlet of the condenser 4. An indoor return air temperature and humidity sensor D is installed between the inlet of the indoor return air (primary air) and the air-to-air heat exchanger 1. This sensor is configured to detect the indoor return air temperature T. D and indoor return air humidity (RH) D A temperature sensor E is installed between the inlet of the air-to-air heat exchanger 1 and the inlet of the evaporator 3. This sensor is configured to detect the temperature of the indoor return air at the inlet of the evaporator 3. An indoor supply air temperature and humidity sensor F is installed on the rear side of the indoor side fan 18. This sensor is configured to detect the indoor supply air temperature T. F Indoor air supply humidity (RH) FThe parameter data detected by the aforementioned sensors can be used to control the normal operation of the indirect evaporative cooling system 100, and can also be used for model training in this application. This application is not limited to the aforementioned sensors, but also includes other suitable sensors for controlling other suitable operations of the indirect evaporative cooling system 100. For example, this application also includes a compressor speed sensor, positioned near the compressor 6, for detecting the speed of the compressor 6; a fan speed sensor, positioned near the outdoor fan 19, for detecting the fan speed of the outdoor fan 19; a motor speed sensor, positioned near the nozzle moving motor 11, for detecting the speed of the nozzle moving motor 11; and an opening degree sensor, positioned near the flow regulating valve 9, for detecting the opening degree of the flow regulating valve 9, etc. In other embodiments, this application detects the aforementioned parameters by setting other suitable sensors.

[0051] Although Figure 2 illustrates the arrangement of the various components of the unit assembly 102 of the indirect evaporative cooling system 100 and the detection device 112, this application is not limited to these specific components of the unit assembly 102 and the specific arrangement of the detection device 112. In other embodiments, the unit assembly 102 of the indirect evaporative cooling system 100 includes other suitable components or structures. In other embodiments, the detection device 112 includes other suitable sensors and other arrangements.

[0052] Figure 3 shows a general flowchart of an embodiment of the control method for the indirect evaporative cooling system 100 shown in Figure 1.

[0053] As shown in Figure 3, the control method 300 of the indirect evaporative cooling system 100 starts at step 302 and then proceeds to step 304.

[0054] In step 304, a refrigeration model and an operating power consumption model for the indirect evaporative cooling system are created. The refrigeration model includes input and output parameters, as does the operating power consumption model. The output parameters of the refrigeration model include control parameters for the indirect evaporative cooling system 100. The output parameters of the operating power consumption model include the operating power consumption of each component of the indirect evaporative cooling system 100. Then, the process proceeds from step 304 to step 306. In one embodiment, the output parameters of the refrigeration model are used as input parameters for the operating power consumption model. The operating power consumption of each component of the indirect evaporative cooling system 100 includes electrical power consumption and water consumption, and the operating cost of the indirect evaporative cooling system 100 includes the electrical cost generated by the electrical power consumption and the water cost generated by the water consumption.

[0055] In step 306, training data is acquired, including the input and output parameters of the refrigeration model of the indirect evaporative cooling system and the input and output parameters of the power consumption model of the indirect evaporative cooling system. Then, the process proceeds from step 306 to step 308.

[0056] In step 308, the refrigeration model and the power consumption model of the indirect evaporative cooling system are trained based on the acquired training data to obtain a determined refrigeration model and a determined power consumption model of the indirect evaporative cooling system. Then, the process proceeds from step 308 to step 310.

[0057] In step 310, the operating cost of the indirect evaporative cooling system 100 is optimized based at least on a determined refrigeration model and a determined power consumption model of the indirect evaporative cooling system, to obtain the output parameter data of the refrigeration model of the indirect evaporative cooling system 100 (i.e., the control parameter data of the indirect evaporative cooling system 100) corresponding to the optimized operating cost of the indirect evaporative cooling system 100, thereby controlling the operation of the indirect evaporative cooling system 100. Then, the process proceeds from step 310 to step 312. A detailed flowchart of an embodiment of step 310 is shown in Figure 6.

[0058] At step 312, it is determined whether the indirect evaporative cooling system 100 needs to be shut down. If it does not need to be shut down, the process proceeds from step 312 to step 310 to continue controlling the operation of the indirect evaporative cooling system 100. If it needs to be shut down, the process proceeds from step 312 to step 314 to terminate the execution of the control method 300 for the indirect evaporative cooling system 100.

[0059] Figure 4 shows a block diagram of an embodiment of the refrigeration model F(x) and the power consumption model G(y) of the indirect evaporative cooling system created in step 304 of Figure 3. As previously described, in step 304 of Figure 3, the refrigeration model and the power consumption model of the indirect evaporative cooling system are created.

[0060] As shown in Figure 4, in one embodiment, the refrigeration model F(x) of the indirect evaporative cooling system selects the following input parameters: x=(T S ,T A ,RH A ,T F ,RH F ,T D ,RH D ),

[0061] Among them, T S The required operating temperature of the cooled unit 106 (i.e., the target parameter of the cooled unit 106), T A Outdoor air temperature, RHA Outdoor air humidity, T F Indoor supply air temperature, RH F For indoor air humidity, T D Indoor return air temperature, RH D This refers to the humidity of the indoor return air.

[0062] Furthermore, the refrigeration model F(x) for the indirect evaporative cooling system selects the following output parameters: y=(N P ,n 冷凝 ,n 喷头 D PF ),

[0063] Where N P For the compressor 6 of the direct expansion refrigeration system 104, the speed, n 冷凝 For the fan speed of outdoor fan 19, n 喷头 The rotational speed of the nozzle moving motor 11, D PF This refers to the opening degree of the flow regulating valve 9. This application controls the operation of the indirect evaporative cooling system 100 to maintain the indoor supply air temperature T. F and indoor return air temperature T D It meets the requirements and can accurately ensure that the cooling capacity meets the cooling capacity requirements of the cooled unit 106.

[0064] Furthermore, the power consumption model G(y) of the indirect evaporative cooling system selects the output parameters of the refrigeration model F(x) of the indirect evaporative cooling system as its input parameters. That is, the input parameters of the power consumption model G(y) of the indirect evaporative cooling system are as follows: y=(N P ,n 冷凝 ,n 喷头 D PF ),

[0065] Furthermore, the power consumption model G(y) for the indirect evaporative cooling system is selected with the following output parameters: z = (P 喷头电机 ,P IT风机 ,P EA风机 ,P 压缩机 W 喷头 ),

[0066] Among them, P 喷头电机 P is the power consumption of the nozzle moving motor 11. IT风机 This refers to the power consumption of the indoor fan 18, P. EA风机 This refers to the power consumption of the outdoor fan 19, P. 压缩机 This is the power consumption of compressor 6, in W. 喷头 This refers to the water volume sprayed from nozzle 13.

[0067] In one embodiment, both the refrigeration model F(x) and the power consumption model G(y) of the indirect evaporative cooling system are fully connected deep neural network models. This application sets the deep neural network model to have m neurons in the input layer, n neurons in the output layer, and k hidden layers. The number of neurons in each hidden layer can be determined using a linear function L = g(m).

[0068] Figure 5 shows a detailed flowchart of one embodiment of step 308 (training the model) shown in Figure 3. As previously described, at step 308 in Figure 3, the refrigeration model and the power consumption model of the indirect evaporative cooling system are trained based on the acquired training data to obtain a determined refrigeration model and a determined power consumption model of the indirect evaporative cooling system. In one embodiment, both the refrigeration model F(x) and the power consumption model G(y) of the indirect evaporative cooling system are fully connected deep neural network models. In other embodiments, the refrigeration model F(x) and the power consumption model G(y) of the indirect evaporative cooling system include other suitable models.

[0069] As shown in Figure 5, the process transitions from step 306 in Figure 3 to step 502 in Figure 5. In step 502, the initial weight values ​​and bias values ​​of the deep neural network model are set. Then, the process transitions from step 502 to step 504. In one embodiment, the initial weight values ​​and bias values ​​φ of the deep neural network model F(x) are set respectively. f And the set φ of initial weights and biases of the deep neural network model G(y). g .

[0070] At step 504, the operating parameter data of the indirect evaporative cooling system 100 is acquired. Then, the process proceeds from step 504 to step 506. In one embodiment, the operating parameter data of the indirect evaporative cooling system 100 includes the input parameter data of the refrigeration model F(x) of the indirect evaporative cooling system (e.g., the operating temperature T of the cooled unit 106). S Outdoor air temperature T A Outdoor air humidity (RH) A Indoor air supply temperature T F Indoor air supply humidity (RH) F Indoor return air temperature T D Indoor return air humidity (RH) D The data) and output parameter data (e.g., the rotational speed N of the compressor 6 in the direct expansion refrigeration system 104). P The fan speed n of outdoor side fan 19 冷凝 The rotational speed n of the nozzle moving motor 11 喷头 The opening degree D of flow regulating valve 9 PFThe operating parameter data of the indirect evaporative cooling system 100 also includes the input parameter data of the indirect evaporative cooling system operating power consumption model G(y) (e.g., the speed N of the compressor 6 of the direct expansion refrigeration system 104). P The fan speed n of outdoor side fan 19 冷凝 The rotational speed n of the nozzle moving motor 11 and the opening degree D of the nozzle and flow regulating valve 9. PF ) and output parameter data (e.g., power consumption P of nozzle moving motor 11) 喷头电机 The power consumption P of the indoor side fan 18 IT风机 The power consumption P of outdoor fan 19 EA风机 The power consumption P of compressor 6 压缩机 The water spray volume W of nozzle 13 喷头 ).

[0071] At step 506, the output of each hidden layer neuron node is calculated based on the activation function. Then, the process proceeds from step 506 to step 508. In one embodiment, the output of each hidden layer neuron node is calculated based on the activation function of the deep neural network model F(x), and the output of each hidden layer neuron node is calculated based on the activation function of the deep neural network model G(y).

[0072] At step 508, the output of the output layer neurons is calculated based on the activation function. Then, the process proceeds from step 508 to step 510. In one embodiment, the output of the output layer neurons is calculated based on the activation function of the deep neural network model F(x), and the output of the output layer neurons is calculated based on the activation function of the deep neural network model G(y).

[0073] At step 510, the error Loss between the predicted output of the output layer and the actual output of the system is calculated. Then, the process proceeds from step 510 to step 512. In one embodiment, the error Loss between the predicted output (predicted value) of the output layer of the deep neural network model (indirect evaporative cooling system refrigeration model) F(x) and the actual output (actual value / true value) of the indirect evaporative cooling system 100 is calculated. f Furthermore, the error Loss between the predicted output (predicted value) of the output layer of the deep neural network model (indirect evaporative cooling system operating power consumption model) G(y) and the actual output (actual value / true value) of the indirect evaporative cooling system 100 is calculated. g In one embodiment, this application uses a loss function to quantify the difference (loss) between the predicted value output by the deep neural network model and the actual value output by the indirect evaporative cooling system 100. By minimizing this difference (loss), the parameters of the deep neural network model can be optimized and the prediction performance of the deep neural network model can be improved. In one embodiment, the loss function Loss of the deep neural network models F(x) and G(y) is...f Loss g They are as follows:

[0074] Wherein, F′(x) is the predicted value output by the refrigeration model F(x) of the indirect evaporative cooling system, f(x) is the actual value output by the indirect evaporative cooling system 100, and x represents the input parameter of the refrigeration model F(x); G′(y) is the predicted value output by the power consumption model G(y) of the indirect evaporative cooling system, g(y) is the actual value output by the indirect evaporative cooling system 100, and y is both the output parameter of the refrigeration model F(x) and the input parameter of the power consumption model G(y). In other embodiments, the loss functions of the deep neural network models F(x) and G(y) include other suitable functions.

[0075] At step 512, the error Loss is determined. f Loss g Does it meet the required value? In one embodiment, the error Loss of the deep neural network model F(x) is determined. f Is it less than or equal to the error threshold ε? f And determine the error loss of the deep neural network model G(y). g Is it less than or equal to the error threshold ε? g If the error Loss meets the required value, it indicates that the error Loss has converged, and the weight values ​​and bias values ​​of the deep neural network models F(x) and G(y) are obtained, that is, the determined refrigeration model F(x) and power consumption model G(y) of the indirect evaporative cooling system are obtained. Therefore, the process proceeds from step 512 to step 310 in Figure 3. If the error Loss does not meet the required value, it indicates that the error Loss has not converged, and the process proceeds from step 512 to step 514 to perform the operation of updating the weights and biases of the deep neural network model. In one embodiment, this application uses the backpropagation algorithm and the gradient descent optimization algorithm to update the weights and biases of the deep neural network model. For example, during training, the gradient of the weights and biases is associated by calculating the loss function, and then the gradient descent algorithm is used to update the weights and biases to minimize the loss function. An embodiment of this method for updating the weights and biases of the deep neural network model is detailed in steps 514-518. In other embodiments, this application may use other suitable algorithms to update the weights and biases of the deep neural network model.

[0076] At step 514, the error of the hidden layer is calculated. Then, the process proceeds from step 514 to step 516. In one embodiment, the error of the hidden layer of the deep neural network model F(x) is calculated, and the error of the hidden layer of the deep neural network model G(y) is also calculated.

[0077] At step 516, the error gradient is calculated. Then, the process proceeds from step 516 to step 518. In one embodiment, the error gradient of the deep neural network model F(x) and the error gradient of the deep neural network model G(y) are calculated.

[0078] At step 518, the weight values ​​and bias values ​​are adjusted according to the learning function. Then, the process proceeds from step 518 to step 506. In one embodiment, the weight values ​​and bias values ​​of the deep neural network model F(x) are adjusted according to the learning function, and the weight values ​​and bias values ​​of the deep neural network model G(y) are also adjusted according to the learning function. This application uses the operating parameter data of the indirect evaporative cooling system 100 to perform supervised learning on the deep neural network model to obtain the loss function Loss. f With Loss g The set φ of weights and biases of the deep neural network models F(x) and G(y) that reaches the minimum value. f With φ g .

[0079] In one embodiment, this application uses an optimizer, such as the Adam optimizer, as an adaptive learning rate optimization algorithm. The learning rate is adjusted based on the gradient information from the previous time step. When the gradient at the previous time step is small, the learning rate is increased; when the gradient at the previous time step is large, the learning rate is decreased. This dynamic adjustment of the learning rate accelerates the training of the neural network model, allowing the weights and biases of the deep neural network model to quickly reach their optimal values, thereby maximizing the loss function. f With Loss g Fastest convergence. For example, the optimizer updates the weights and biases according to the following formula:

[0080] Where, φ t It is the set of weights and biases of the model before the update; φ t71 It is the set of weight values ​​and bias values ​​of the updated model; β1 and β2 are the gradients of the model's weights and biases at position t; β1 and β2 are the decay coefficients of the two exponentially weighted averages, typically taken as constants; m t-1 and v t-1 It is the moving average of the gradient before bias correction, initialized to a zero matrix; m t and v t η is the moving average after gradient bias correction; η is the learning rate, and ε is set to 1e-8 to avoid division by zero in the above formula (2.1).

[0081] The set φ of weights and biases of the above deep neural network models F(x) and G(y) f With φg Applied to φ respectively t The weight and deviation values ​​of the next position are updated according to the above formula (2.1).

[0082] Furthermore, the gradient momentum at the next position is calculated by updating the following formula (2.2), and the gradient variance at the next position is calculated by updating the following formula (2.3):

[0083] Where, m t It is the current momentum, m t-1 It represents the momentum at the next position, where the momentum here represents the exponentially weighted moving average of the gradient; v t It is the variance of the current position, v t-1 This is the variance at the next position, where the variance represents the exponentially weighted moving average of the squared gradient. Initialize m. t v is a zero matrix, initialized to zero. t It is a zero matrix.

[0084] By iteratively applying the three formulas (2.1), (2.2), and (2.3), the weight and deviation values ​​of the previous position are continuously used to calculate the weight and deviation values ​​of the next position until the loss is reached. f With Loss g convergence.

[0085] This application uses a portion of the training data as a validation set. It also periodically evaluates model performance on the validation set to detect overfitting or underfitting. Detection accuracy and loss metrics are used to measure model performance. If inaccurate or overfitting is observed, the learning rate, batch size, and model complexity are adjusted based on the validation set performance to improve model performance.

[0086] Figure 6 shows a detailed flowchart of an embodiment of step 310 (obtaining control parameter data for system operating cost optimization) shown in Figure 3. As previously described, at step 310 in Figure 3, the operating cost of the indirect evaporative cooling system 100 is optimized based at least on a determined refrigeration model and a determined power consumption model of the indirect evaporative cooling system, in order to obtain the output parameter data of the refrigeration model of the indirect evaporative cooling system 100 corresponding to the operating cost optimization (i.e., the control parameter data of the indirect evaporative cooling system 100), thereby controlling the operation of the indirect evaporative cooling system 100.

[0087] As shown in Figure 6, the process transitions from step 308 in Figure 3 to step 602 in Figure 6. At step 602, the field data of the input parameters for the refrigeration model of the indirect evaporative cooling system are obtained. Then, the process transitions from step 602 to step 604.

[0088] In step 604, the acquired field data is input into the determined indirect evaporative cooling system refrigeration model to obtain a dataset of the output parameters of the indirect evaporative cooling system refrigeration model. Then, the process proceeds from step 604 to step 606.

[0089] In step 606, the dataset of output parameters of the acquired indirect evaporative cooling system refrigeration model is input into the indirect evaporative cooling system operating cost optimization model to obtain the output parameter data of the indirect evaporative cooling system refrigeration model corresponding to the operating cost optimization of the indirect evaporative cooling system 100, thereby controlling the operation of the indirect evaporative cooling system 100. Then, the process proceeds from step 606 to step 312 in Figure 3. In one embodiment, the indirect evaporative cooling system operating cost optimization model is configured to: obtain an operating power consumption set composed of the operating power consumption of each part of the indirect evaporative cooling system 100 based on the dataset of output parameters of the acquired indirect evaporative cooling system refrigeration model and the indirect evaporative cooling system operating power consumption model; obtain an operating cost set of the indirect evaporative cooling system 100 based on the obtained operating power consumption set; and optimize the obtained operating cost set to obtain the output parameter data of the indirect evaporative cooling system refrigeration model corresponding to the operating cost optimization of the indirect evaporative cooling system 100.

[0090] The obtained output parameter data of the indirect evaporative cooling system refrigeration model F(x), i.e., the control parameter data of the indirect evaporative cooling system 100, can be used by the controller 101 of the indirect evaporative cooling system 100 (e.g., through the actuator of the indirect evaporative cooling system 100) to control the operation of the indirect evaporative cooling system 100, thereby adjusting the temperature of the cooled unit 106 (e.g., a data center) to meet the requirements. The output parameter data of the indirect evaporative cooling system refrigeration model F(x) includes multiple sets of values ​​equal to or close to the actual requirements, i.e., output parameter dataset / control parameter dataset. If these values ​​(output parameter dataset / control parameter dataset) make the temperature of the data center meet the temperature requirements (e.g., reach the temperature threshold, or be within the temperature threshold range), then they can all be considered as solutions. The solution set composed of all solutions is: Y = Q((N P ,n 冷凝 ,n 喷头 D PF ))1},i∈{1,2,3,4,5,…I}.

[0091] Each solution (i.e., each set of output parameter data / control parameter data) corresponds to a total operating cost for the indirect evaporative cooling system 100. 总The optimal cost, such as the minimum cost, is calculated for the corresponding total operating cost set of the solution set. In one embodiment, this application uses a genetic algorithm to globally search the above solution set (i.e., the output parameter dataset / control parameter dataset of the indirect evaporative cooling system refrigeration model F(x)) to quickly solve for the control parameter data of the indirect evaporative cooling system 100 that satisfies the cooling requirements of the cooled unit 106 (e.g., a data center) and has the lowest total operating cost of the indirect evaporative cooling system 100.

[0092] In one embodiment, the total operating cost of the indirect evaporative cooling system 100 = electricity price * electricity consumption + water price * water volume, calculated as follows: Cost 总 =C E ×(P 喷头电机 +P IT风机 +P EA风机 +P 压缩机 )+C a ×W 喷头 ,

[0093] Among them, C E It's the local electricity price, P 喷头电机 P is the power consumption of the nozzle moving motor 11. IT风机 This refers to the power consumption of the indoor fan 18, P. EA风机 This refers to the power consumption of the outdoor fan 19, P. 压缩机 This refers to the power consumption of compressor 6, C. a That's the local tap water price, W 喷头 This refers to the water spray volume of nozzle 13. The total operating cost of the indirect evaporative cooling system 100 per unit time can be calculated using this formula. This application uses the power consumption model G(y) of the indirect evaporative cooling system as the optimization objective, and the operating parameters output by the refrigeration model F(x) of the indirect evaporative cooling system as the input parameters of the power consumption model G(y). The power consumption of each part of the indirect evaporative cooling system 100, including the power consumption P of the nozzle moving motor 11, is obtained through the power consumption model G(y). 喷头电机 The power consumption P of the indoor side fan 18 IT风机 The power consumption P of outdoor fan 19 EA风机 The power consumption P of compressor 6 压缩机 The water spray volume W of nozzle 13 喷头 The total operating cost of the indirect evaporative cooling system 100 can be calculated based on the above total operating cost formula and the power consumption of each part of the obtained indirect evaporative cooling system 100.

[0094] Since the input parameters of the power consumption model G(y) of the indirect evaporative cooling system are multiple sets of control parameters that satisfy the refrigeration condition, determining which set of control parameters makes the indirect evaporative cooling system 100 operate most energy-efficiently requires calculating the system's operating power consumption through the power consumption model G(y) to obtain the system's operating cost and ensure that the total system operating cost is minimized. This application uses the power consumption model G(y) of the indirect evaporative cooling system as a black-box optimization objective and employs a genetic algorithm to find the control parameter data that optimizes (e.g., minimizes) the system's operating cost. An embodiment of this genetic algorithm optimization calculation is shown in detail in Figure 7.

[0095] Figure 7 shows a detailed flowchart of one embodiment of step 606 shown in Figure 6.

[0096] As shown in Figure 7, the process transitions from step 604 in Figure 6 to step 702 in Figure 7. At step 702, each individual in the population is initialized, i.e., the control parameters of the indirect evaporative cooling system 100 (i.e., the input parameters of the indirect evaporative cooling system operating power consumption model G(y)), such as the rotational speed N of the compressor 6 of the direct expansion refrigeration system 104. P The fan speed n of outdoor side fan 19 冷凝 The rotational speed n of the nozzle moving motor 11 喷头 The opening degree D of flow regulating valve 9 PF Then, proceed from step 702 to step 704.

[0097] At step 704, the fitness value of each acquired individual (i.e., the control parameters of the indirect evaporative cooling system 100) is calculated according to the fitness function. In one embodiment, the fitness function in the genetic algorithm is obtained based on a determined indirect evaporative cooling system operating power consumption model G(y). For example, the fitness function is the same as the determined indirect evaporative cooling system operating power consumption model G(y). Then, the process proceeds from step 704 to step 706.

[0098] In step 706, individuals with higher fitness values ​​are selected from among the individuals and "retained". Then, the process proceeds from step 706 to step 708.

[0099] At step 708, the retained individuals are cross-crossed in pairs to generate new individuals. Then, the process proceeds from step 708 to step 710.

[0100] At step 710, the new individuals generated after crossover are subjected to random mutation. Then, the process proceeds from step 710 to step 712.

[0101] At step 712, the randomly mutated individuals form a new population. Then, the process proceeds from step 712 to step 714.

[0102] At step 714, new individuals are obtained based on the new individual group. Then, the process proceeds from step 714 to step 716.

[0103] At step 716, a new operating cost (i.e., system operating cost) for the indirect evaporative cooling system 100 is calculated based on the new individual obtained in step 714. Then, the process proceeds from step 716 to step 718.

[0104] At step 718, it is determined whether the maximum number of iterations has been reached. If the maximum number of iterations has not been reached, the process proceeds from step 718 to step 704. If the maximum number of iterations has been reached, the process proceeds from step 718 to step 720.

[0105] At step 720, optimized control parameter data for system operating costs is obtained, for example, control parameter data when the total operating cost of the indirect evaporative cooling system 100 is minimized. Then, the process proceeds from step 720 to step 312 in Figure 3. In one embodiment, the operation of the indirect evaporative cooling system 100 is controlled based on the obtained optimized control parameter data for system operating costs, and after a period of time (e.g., a predetermined time period), the process proceeds to step 312 in Figure 3.

[0106] When obtaining the control parameter data of the indirect evaporative cooling system 100 to optimize its operating cost, the limits of the control parameters need to be considered. Control parameter data exceeding the limits may allow the indirect evaporative cooling system 100 to operate normally, but it may increase its operating cost. Therefore, limiting the control parameter data of the indirect evaporative cooling system 100 within its limits can optimize its operating cost. For example, after outdoor air is sprayed by the water mist system 103, once the air's moisture content is saturated, the air temperature will no longer decrease. Since the nozzles 13 in the water mist system 103 are driven by the nozzle moving motor 11 to reciprocate, the humidity level in the air depends on the reciprocating speed of the nozzles 13. Therefore, after outdoor air is sprayed by the water mist system 103, when the air's moisture content is saturated, the nozzle moving motor 11 has a corresponding maximum speed limit, i.e., a speed limit. Even when the nozzle moving motor 11 operates beyond its speed limit, the air remains saturated with moisture and the air temperature no longer decreases. Therefore, operating the nozzle moving motor 11 beyond its speed limit generates unnecessary power consumption, resulting in wasted operating costs for the indirect evaporative cooling system 100. In one embodiment, to save on the operating costs of the indirect evaporative cooling system 100, this application obtains the speed limit of the nozzle moving motor 11 after the outdoor air is sprayed by the water mist system 103 at the current outdoor dry-bulb temperature and current outdoor dry air moisture content. The speed of the nozzle moving motor 11 is then limited to within the corresponding speed limit to optimize the operating costs of the indirect evaporative cooling system 100. In one embodiment, this application obtains the speed limit of the nozzle moving motor 11 after the outdoor air is sprayed by the water mist system 103 using a motor speed limit model. The motor speed limit model is configured to obtain the speed limit of the nozzle moving motor 11 after the outdoor air is sprayed by the water mist system 103 based on the outdoor air dry bulb temperature and outdoor air dry air moisture content of the cooled unit 106.

[0107] Figure 8 shows a flowchart of one embodiment of a method for obtaining a motor speed limit model.

[0108] As shown in Figure 8, the method 800 for obtaining the motor speed limit model begins execution at step 802. Then, the process proceeds from step 802 to step 804.

[0109] At step 804, a temperature-humidity model is created and obtained between the humidity content and air temperature of the outdoor air after being sprayed by the water mist system 103. Then, the process proceeds from step 804 to step 806. The temperature-humidity model uses the current outdoor dry-bulb temperature and the current outdoor dry air humidity content as constants. In one embodiment, the temperature-humidity model between the humidity content and air temperature of the outdoor air after being sprayed by the water mist system 103 is as follows: T = K × (RR) 初始 )+T 初始 ,

[0110] Where T is the temperature of the outdoor air after it has been sprayed. 初始 Let K be the current outdoor dry-bulb temperature, K be the variation coefficient, and R be the moisture content of the outdoor air after spraying. 初始 This refers to the current dry-bulb humidity content of the outdoor air. For example, the current outdoor dry-bulb temperature T. 初始 And the current outdoor air dry air moisture content R 初始 Detected by sensor A (see Figure 2). This application uses a detection device (not shown) to detect and acquire the outdoor air temperature T and the dry air moisture content R after spraying within a predetermined time period. Assume the current outdoor air dry-bulb temperature T... 初始 And the current outdoor air dry air moisture content R 初始 The outdoor air temperature T and the dry air humidity R obtained during this predetermined time period remain unchanged. These values ​​are then input into the aforementioned temperature and humidity model, along with the current outdoor dry-bulb temperature T. 初始 And the current outdoor air dry air moisture content R 初始 This allows us to obtain the variation coefficient K in the above temperature and humidity model, and thus obtain the current outdoor dry-bulb temperature T. 初始 And the current outdoor air dry air moisture content R 初始 A defined temperature and humidity model is used.

[0111] In step 806, a temperature and humidity limit model is created and obtained between the air humidity limit and the air temperature limit when the outdoor air is saturated after being sprayed by the water mist system 103. Then, the process proceeds from step 806 to step 808. According to the air enthalpy-humidity diagram (see Figure 9), when water is sprayed onto the air, the outdoor air is in contact with the water for a long time. The temperature of the water and the saturated air layer on its surface is the wet-bulb temperature of the humid air, i.e., the lowest temperature under isoenthalpy. To lower the air temperature using the spraying method, humidity needs to be increased. When the air humidity is saturated, the air temperature drops to its lowest point. After the outdoor air is sprayed and the air humidity is saturated, the air temperature no longer decreases. The temperature and humidity limit model can be trained using the boundary curve (saturation boundary line) data at the lower right edge of the air enthalpy-humidity diagram (see Figure 9). In one embodiment, the temperature and humidity limit model uses an Nth-order polynomial model, as follows:

[0112] Where T is the air temperature limit at air humidity saturation, R is the air moisture content limit at air humidity saturation, and a1 is the variation coefficient of the polynomial model. Multiple data points are collected from the boundary curve (saturation boundary line) at the lower right edge of the air enthalpy-humidity diagram (see Figure 9), and these collected data are input into the aforementioned polynomial model to obtain a definite temperature and humidity limit model. Therefore, the current outdoor dry-bulb temperature T can be obtained from the temperature and humidity model and the temperature and humidity limit model. 初始 And the current outdoor air dry air moisture content R 初始 The extreme humidity and extreme temperature of the outdoor air after it has been sprayed.

[0113] At step 808, a speed-moisture content limit model is created and obtained between the speed limit of the nozzle moving motor 11 and the air moisture content limit when the air is saturated. Then, the process proceeds from step 808 to step 810. Since the nozzle 13 in the water mist system 103 is driven by the nozzle moving motor 11 to reciprocate, the humidity level in the air depends on the reciprocating speed of the nozzle 13. Therefore, after the outdoor air is sprayed by the water mist system 103, when the air moisture content is saturated, the nozzle moving motor 11 has a corresponding maximum speed limit, i.e., a speed limit. For example, based on the speed limit of the nozzle moving motor 11 and the air moisture content limit when the air is saturated, provided by the component manufacturer, a speed-moisture content limit model can be fitted, for example, using a polynomial.

[0114] In step 810, the acquired temperature and humidity model, temperature and humidity limit model, and speed-to-humidity limit model are combined to obtain the motor speed limit model. Then, the process proceeds from step 810 to step 812, ending the execution of the method 800 for obtaining the motor speed limit model. In one embodiment, the combination of the above-mentioned temperature and humidity model, temperature and humidity limit model, and speed-to-humidity limit model is called the motor speed limit model. The motor speed limit model, which combines the temperature and humidity model, temperature and humidity limit model, and speed-to-humidity limit model, can obtain the current outdoor dry-bulb temperature T. 初始 And the current outdoor air dry air moisture content R 初始 The speed limit of the nozzle moving motor 11 corresponding to the saturated moisture content of the outdoor air after it is sprayed.

[0115] Figure 9 shows the enthalpy-humidity diagram of air. As shown in Figure 9, the horizontal axis represents the moisture content of the air, and the vertical axis represents the air temperature. Several parallel diagonal lines sloping relative to the horizontal and vertical axes represent isenthalpic lines. The boundary curve (saturation boundary line) at the lower right edge of the enthalpy-humidity diagram shows the relationship between the air moisture content (i.e., the air moisture content limit) and the air temperature (i.e., the air temperature limit) when the air humidity is saturated (100%).

[0116] Figure 10 shows the block diagram structure of the PID control of the controller 101 shown in Figure 1. The control parameter data corresponding to the optimization of system operating cost, such as the output parameter data of the power consumption model G(y) of the indirect evaporative cooling system (e.g., the power consumption of each part of the system), is used to minimize the total system operating cost. The input parameter y = ((N) of the power consumption model G(y) of the indirect evaporative cooling system is then calculated. P ,n 冷凝 ,n 喷头 D PF The input parameters are fed into the controller 101, and the PID algorithm is used to quickly make the actuator of the indirect evaporative cooling system 100 run to the specified input parameter value.

[0117] As shown in Figure 10, the input parameter data / control parameter data y = ((N P ,n 冷凝 ,n 喷头 D PF The input parameters (i.e., target data) are used as input to the PID algorithm. After passing through proportional, derivative, and integral control modules, the data is superimposed, and the result is input to the actuator of the indirect evaporative cooling system 100 to operate it. Measuring elements, such as the detection device 112 (see Figure 1), detect and acquire the operating parameter data of the indirect evaporative cooling system 100 and use it as feedback data. This feedback data is superimposed with the input parameter data / control parameter data and then passed through the proportional, derivative, and integral control modules. This process is repeated until the indirect evaporative cooling system 100 operates to the required input parameter data / control parameter data (i.e., target data).

[0118] Figure 11 shows a block diagram of the computing system 110 and control system 111 shown in Figure 1.

[0119] As shown in Figure 11, the computing system 110 includes a memory 1101, a processor 1102, an input interface 1103, an output interface 1104, and a bus 1105. The memory 1101, processor 1102, input interface 1103, and output interface 1104 are connected to the bus 1105. The processor 1102 can read programs (or instructions) from the memory 1101 and execute the programs (or instructions) to perform data processing. The processor 1102 can also write data or programs (or instructions) into the memory 1101. The memory 1101 can store programs (instructions) or data. By executing the instructions in the memory 1101, the processor 1102 can control the memory 1101, the input interface 1103, and the output interface 1104.

[0120] Input interface 1103 is configured to receive training data for an indirect evaporative cooling system refrigeration model and an indirect evaporative cooling system operating power consumption model via connection line 115. Input interface 1103 is also configured to convert the received data into data recognizable by processor 1102 and output the data to processor 1102. Processor 1102 is configured to process the received data (e.g., train the model) to obtain a determined indirect evaporative cooling system refrigeration model and an indirect evaporative cooling system operating power consumption model, and generate a program for the determined indirect evaporative cooling system refrigeration model and indirect evaporative cooling system operating power consumption model. Processor 1102 is configured to create and acquire an indirect evaporative cooling system operating cost optimization model, and generate a program for the determined indirect evaporative cooling system operating cost optimization model. Processor 1102 is also configured to combine the programs for the determined indirect evaporative cooling system refrigeration model and indirect evaporative cooling system operating power consumption model and the program for the determined indirect evaporative cooling system operating cost optimization model into a module package. Output interface 1104 is configured to receive the module package from processor 1102 and send the module package to control system 111 via connection line 113.

[0121] The control system 111 is configured to receive module packages from the computing system 110 via connection line 113 and store the module packages in memory 1111. The control system 111 includes memory 1111, processor 1112, input interface 1113, output interface 1114, and bus 1115. Memory 1111, processor 1112, input interface 1113, and output interface 1114 are connected to bus 1115. Processor 1112 can read programs (or instructions) from memory 1111 and execute the programs (or instructions) to perform data processing. Processor 1112 can also write data or programs (or instructions) into memory 1111. Memory 1111 can store programs (instructions) or data. By executing instructions in memory 1111, processor 1112 can control memory 1111, input interface 1113, and output interface 1114. The memory 1111 includes a data preprocessing module 1118, an indirect evaporative cooling system module 1117, and an operating cost optimization module 1116.

[0122] Input interface 1113 is configured to receive a module package from computing system 110 via connection line 113, convert the module package into a module package recognizable by memory 1111, store the determined indirect evaporative cooling system refrigeration model and indirect evaporative cooling system operating power consumption model programs in the module package into indirect evaporative cooling system module 1117 in memory 1111, and store the determined indirect evaporative cooling system operating cost optimization model programs in the module package into operating cost optimization module 1116 in memory 1111. In one embodiment, the above module package is copied from computing system 110 to memory 1111 of control system 111 via USB interface.

[0123] Input interface 1113 is also configured to receive field data of the indirect evaporative cooling system 100 detected by detection device 112 via connection line 116, convert the data into data recognizable by processor 1112, and output the data to processor 1112. Processor 1112 is configured to invoke data preprocessing module 1118 to read its data and / or program (or instructions), and execute the program (or instructions) to process (e.g., preprocess) the received field data of indirect evaporative cooling system 100. Processor 1112 is also configured to invoke indirect evaporative cooling system module 1117 to read its data and / or program (or instructions), and execute the program (or instructions) to process the preprocessed data generated by data preprocessing module 1118 (e.g., run indirect evaporative cooling system refrigeration model) to obtain the output parameter dataset of indirect evaporative cooling system refrigeration model. The processor 1112 is also configured to invoke the operating cost optimization module 1116 to read its data and / or program (or instructions), and execute the program (or instructions) to process the output parameter dataset generated by the indirect evaporative cooling system module 1117 to generate output parameter data (i.e., control parameter data) of the indirect evaporative cooling system refrigeration model corresponding to the operating cost optimization of the indirect evaporative cooling system 100. In one embodiment, the operating cost optimization module 1116, when executed, invokes the program of the determined indirect evaporative cooling system operating power consumption model in the indirect evaporative cooling system module 1117, and combines it with the output parameter dataset generated by the indirect evaporative cooling system module 1117 to generate control parameter data for the operating cost optimization of the indirect evaporative cooling system 100. In one embodiment, the processor 1112 obtains an operating power set consisting of the operating power consumption of each part of the indirect evaporative cooling system 100 based on the output parameter dataset generated by the indirect evaporative cooling system module 1117 and the corresponding operating power consumption model of the indirect evaporative cooling system. Based on the obtained operating power set, the processor obtains an operating cost set of the indirect evaporative cooling system 100 and optimizes the obtained operating cost set to obtain the output parameter data of the indirect evaporative cooling system refrigeration model corresponding to the operating cost optimization of the indirect evaporative cooling system 100.

[0124] In one embodiment, the program for the determined refrigeration model of the indirect evaporative cooling system is stored in the indirect evaporative cooling system module 1117 in memory 1111, and the programs for the determined operating power consumption model and the determined operating cost optimization model of the indirect evaporative cooling system are stored in the operating cost optimization module 1116 in memory 1111. When executed, the operating cost optimization module 1116 calls the program for the determined operating power consumption model of the indirect evaporative cooling system and combines it with the output parameter dataset generated by the indirect evaporative cooling system module 1117 to generate control parameter data for optimizing the operating cost of the indirect evaporative cooling system 100.

[0125] Output interface 1114 is configured to receive control parameter data from processor 1112 when optimizing the operating cost of indirect evaporative cooling system 100, convert the data into a control signal suitable for indirect evaporative cooling system 100, and send the control signal to indirect evaporative cooling system 100 (e.g., its controller 101) via connection line 114 to control the operation of indirect evaporative cooling system 100.

[0126] All features and / or steps of any method or process so disclosed in this specification (including any appended claims, abstract, and drawings) can be combined in any suitable combination, except for at least some mutually exclusive combinations of such features and / or steps. This application is not limited to the specific sequence of steps described in the specification, but includes other suitable sequences of steps for carrying out this application.

[0127] Although this application has been described with reference to examples of the embodiments outlined above, various alternatives, modifications, variations, improvements, and / or substantially equivalents, whether known or currently or soon to be foreseen, will likely be apparent to those skilled in the art. Furthermore, the technical effects and / or technical problems described herein are exemplary and not limiting; therefore, the disclosures herein may be used to solve other technical problems and have other technical effects and / or can solve other technical problems. Thus, the examples of embodiments of this application as set forth above are intended to be illustrative and not limiting. Various changes can be made without departing from the spirit or scope of this application. Therefore, this application is intended to include all known or previously developed alternatives, modifications, variations, improvements, and / or substantially equivalents.

Claims

1. A control method (300) for an indirect evaporative cooling system, the control method (300) comprising: S1: Create a refrigeration model and an operating power consumption model for an indirect evaporative cooling system. The refrigeration model includes input parameters and output parameters, and the operating power consumption model includes input parameters and output parameters. The output parameters of the refrigeration model include the control parameters of the indirect evaporative cooling system (100), and the output parameters of the operating power consumption model include the operating power consumption of each part of the indirect evaporative cooling system (100). S2: Obtain training data, which includes the input and output parameters of the refrigeration model of the indirect evaporative cooling system and the input and output parameters of the power consumption model of the indirect evaporative cooling system. S3: Based on the acquired training data, train the refrigeration model of the indirect evaporative cooling system and the operating power consumption model of the indirect evaporative cooling system to obtain a determined refrigeration model and an determined operating power consumption model of the indirect evaporative cooling system; and S4: At least based on the determined indirect evaporative cooling system refrigeration model and the determined indirect evaporative cooling system operating power consumption model, optimize the operating cost of the indirect evaporative cooling system (100) to obtain the output parameter data of the indirect evaporative cooling system refrigeration model corresponding to the optimization of the operating cost of the indirect evaporative cooling system (100), thereby controlling the operation of the indirect evaporative cooling system (100).

2. The control method (300) for an indirect evaporative cooling system according to claim 1, wherein, The output parameters of the refrigeration model of the indirect evaporative cooling system are used as the input parameters of the power consumption model of the indirect evaporative cooling system.

3. The control method (300) for an indirect evaporative cooling system according to claim 1, wherein, The input parameters for the refrigeration model of the indirect evaporative cooling system include the indoor supply air temperature and the indoor return air temperature.

4. The control method (300) for an indirect evaporative cooling system according to claim 1, wherein, In step S4, the following operations are performed to obtain data on the output parameters of the refrigeration model of the indirect evaporative cooling system (100) when optimizing the operating cost: S4.1: Obtain field data of the input parameters for the refrigeration model of the indirect evaporative cooling system; S4.2: Input the acquired field data into the determined indirect evaporative cooling system refrigeration model to obtain the dataset of output parameters of the indirect evaporative cooling system refrigeration model; S4.3: Input the dataset of output parameters of the obtained indirect evaporative cooling system refrigeration model into the indirect evaporative cooling system operating cost optimization model to obtain the data of output parameters of the indirect evaporative cooling system refrigeration model corresponding to the operating cost optimization of the indirect evaporative cooling system (100).

5. The control method (300) for an indirect evaporative cooling system according to claim 4, wherein, The operating cost optimization model for the indirect evaporative cooling system is configured as follows: Based on the dataset of output parameters of the refrigeration model of the indirect evaporative cooling system and the power consumption model of the indirect evaporative cooling system, the power consumption of each part of the indirect evaporative cooling system (100) is obtained accordingly. The operating cost set of the indirect evaporative cooling system (100) is obtained based on the acquired operating power consumption set; as well as The obtained operating cost set is optimized to obtain the output parameter data of the refrigeration model of the indirect evaporative cooling system (100) when optimizing the operating cost of the indirect evaporative cooling system (100).

6. The control method (300) for an indirect evaporative cooling system according to claim 1, wherein, The operating power consumption of each part of the indirect evaporative cooling system (100) includes electrical power consumption and water consumption, and the operating cost of the indirect evaporative cooling system (100) includes the electrical cost generated by the electrical power consumption and the water cost generated by the water consumption.

7. The control method (300) for an indirect evaporative cooling system according to claim 1, wherein: The indirect evaporative cooling system (100) is configured to cool a unit (106). The indirect evaporative cooling system (100) includes a water mist system (103). The output parameters of the refrigeration model of the indirect evaporative cooling system include the rotational speed of the nozzle moving motor (11) of the water mist system (103). The nozzle moving motor (11) is configured to control the movement of the nozzles (13) of the water mist system (103). The control method (300) further includes: obtaining a motor speed limit model, wherein the motor speed limit model is configured to obtain the speed limit of the nozzle moving motor (11) after the outdoor air is sprayed by the water mist system (103) based on the outdoor air dry bulb temperature and outdoor air dry air moisture content of the cooled unit (106); In step S4, the speed limit of the nozzle moving motor (11) after the outdoor air is sprayed by the water mist system (103) under the current outdoor air dry bulb temperature and the current outdoor air dry air humidity is obtained, and the speed of the nozzle moving motor (11) when the outdoor air is sprayed by the water mist system (103) is limited to the corresponding speed limit of the nozzle moving motor (11) in order to optimize the operating cost of the indirect evaporative cooling system (100).

8. The control method (300) for an indirect evaporative cooling system according to claim 7, wherein, The motor speed limit model is obtained through the following operations: Create and obtain a temperature and humidity model between the humidity of outdoor air after it has been sprayed by the water mist system (103) and the air temperature, wherein the temperature and humidity model uses the current outdoor air dry-bulb temperature and the current outdoor air dry air humidity as constants; Create and obtain a temperature and humidity limit model between the air humidity limit and the air temperature limit when the outdoor air is saturated after being sprayed by the water mist system (103); Create and obtain a speed-moisture content limit model between the speed limit of the nozzle moving motor (11) and the air moisture content limit when the air is saturated; and The obtained temperature and humidity model, temperature and humidity limit model, and speed and humidity content limit model are combined to obtain the motor speed limit model.

9. The control method (300) for an indirect evaporative cooling system according to claim 7, wherein: The indirect evaporative cooling system (100) also includes a direct expansion refrigeration system (104) and a fan system (105); The input parameters of the refrigeration model of the indirect evaporative cooling system include the target parameters of the cooled unit (106) and the operating parameters of the indirect evaporative cooling system (100). The output parameters of the refrigeration model of the indirect evaporative cooling system include the control parameters that enable the operation of the water spray system (103), the direct expansion refrigeration system (104) and the fan system (105). The output parameters of the power consumption model of the indirect evaporative cooling system include the power consumption and water consumption of the water mist system (103) of the indirect evaporative cooling system (100), the power consumption of the direct expansion refrigeration system (104), and the power consumption of the fan system (105); and The operating cost of the indirect evaporative cooling system (100) includes the sum of the electricity and water costs of the water spray system (103) of the indirect evaporative cooling system (100), the electricity costs of the direct expansion refrigeration system (104), and the electricity costs of the fan system (105).

10. The control method (300) for an indirect evaporative cooling system according to claim 9, wherein, The direct expansion refrigeration system (104) includes a compressor (6), a condenser (4), an evaporator (3), and a flow regulating valve (9). The fan system (105) includes an indoor fan (18) and an outdoor fan (19). The target parameter of the cooled unit (106) is the target temperature of the cooled unit (106), and the operating parameters of the indirect evaporative cooling system (100) include the outdoor air temperature, outdoor air humidity, indoor supply air temperature, indoor supply air humidity, indoor return air temperature and indoor return air humidity of the cooled unit (106). The control parameters for operating the water spray system (103) include the rotational speed of the nozzle moving motor (11), the control parameters for operating the direct expansion refrigeration system (104) include the rotational speed of the compressor (6) and the opening degree of the flow regulating valve (9), and the control parameters for operating the fan system (105) include the fan speed of the outdoor fan (19). The power consumption of the water spray system (103) includes the power consumption of the nozzle moving motor (11), the water consumption of the water spray system (103) includes the water volume of the nozzle (13), the power consumption of the direct expansion refrigeration system (104) includes the power consumption of the compressor (6), and the power consumption of the fan system (105) includes the power consumption of the indoor fan (18) and the power consumption of the outdoor fan (19). The power cost of the water spray system (103) includes the power consumption of the nozzle moving motor (11), the water cost of the water spray system (103) includes the water consumption of the nozzle (13), the power cost of the direct expansion refrigeration system (104) includes the power consumption of the compressor (6), and the power cost of the fan system (105) includes the power consumption of the indoor fan (18) and the power consumption of the outdoor fan (19).

11. The control method (300) for an indirect evaporative cooling system according to claim 1, wherein, Both the refrigeration model and the power consumption model of the indirect evaporative cooling system are fully connected deep neural network models.

12. The control method (300) for an indirect evaporative cooling system according to claim 1, wherein the backpropagation algorithm and the gradient descent optimization algorithm are used to train the refrigeration model of the indirect evaporative cooling system and the power consumption model of the indirect evaporative cooling system.

13. The control method (300) for an indirect evaporative cooling system according to claim 4, wherein, The indirect evaporative cooling system operating cost optimization model is configured to use a genetic algorithm to optimize the operating cost of the indirect evaporative cooling system (100), wherein the fitness function in the genetic algorithm is obtained based on the determined indirect evaporative cooling system operating power consumption model.

14. The control method (300) for an indirect evaporative cooling system according to claim 1, wherein: The indirect evaporative cooling system (100) is applied to a test bench, and data obtained from the test bench is used as training data to train the refrigeration model and the power consumption model of the indirect evaporative cooling system, so as to obtain a determined refrigeration model and the determined power consumption model of the indirect evaporative cooling system; and The indirect evaporative cooling system (100) is applied to a cooled unit (106) in the field, and the data obtained from the cooled unit (106) in the field is used as further training data to further train the determined indirect evaporative cooling system refrigeration model and the determined indirect evaporative cooling system operating power consumption model, so as to adjust the determined indirect evaporative cooling system refrigeration model and the determined indirect evaporative cooling system operating power consumption model, thereby obtaining the final determined indirect evaporative cooling system refrigeration model and the indirect evaporative cooling system operating power consumption model.

15. A computing system (110), the computing system (110) comprising: A processor (1102) configured to execute the control method (300) according to any one of claims 1-14 to obtain a determined refrigeration model of the indirect evaporative cooling system and a determined power consumption model of the indirect evaporative cooling system.

16. A control system (111), the control system (111) comprising: Indirect evaporative cooling system module (1117); as well as An indirect evaporative cooling system operating cost optimization module (1116) is connected to the indirect evaporative cooling system module (1117). The indirect evaporative cooling system operating cost optimization module (1116) is configured to perform the control method (300) according to any one of claims 1-14 to optimize the operating cost of the indirect evaporative cooling system (100) based on the current operating condition data of the indirect evaporative cooling system (100) and the indirect evaporative cooling system module (1117), so as to obtain the control data of the indirect evaporative cooling system (100) corresponding to the optimized operating cost of the indirect evaporative cooling system (100), thereby controlling the operation of the indirect evaporative cooling system (100). The indirect evaporative cooling system module (1117) includes a determined refrigeration model of the indirect evaporative cooling system and a determined operating power consumption model of the indirect evaporative cooling system obtained by the control method (300) according to any one of claims 1-14; or The indirect evaporative cooling system module (1117) and the indirect evaporative cooling system operating cost optimization module (1116) respectively include a determined indirect evaporative cooling system refrigeration model and a determined indirect evaporative cooling system operating power consumption model obtained by the control method (300) according to any one of claims 1-14.

17. The control system (111) according to claim 16, wherein, The determined refrigeration model and the determined operating power consumption model of the indirect evaporative cooling system are deployed to the indirect evaporative cooling system module (1117) by the computing system (110) as described in claim 15; or The obtained refrigeration model of the indirect evaporative cooling system and the determined power consumption model of the indirect evaporative cooling system are deployed to the indirect evaporative cooling system module (1117) and the indirect evaporative cooling system operating cost optimization module (1116) respectively by the computing system (110) of claim 15.

18. The control system (111) according to claim 16, wherein, The control system (111) includes a processor (1112) configured to execute the control method (300) according to any one of claims 1-14 to obtain a determined refrigeration model of the indirect evaporative cooling system and a determined operating power consumption model of the indirect evaporative cooling system, and to deploy the obtained determined refrigeration model of the indirect evaporative cooling system and the determined operating power consumption model of the indirect evaporative cooling system to the indirect evaporative cooling system module (1117), or to deploy them to the indirect evaporative cooling system module (1117) and the indirect evaporative cooling system operating cost optimization module (1116), respectively.

19. An indirect evaporative cooling system (100), the indirect evaporative cooling system (100) comprising: Water mist system (103); Direct expansion refrigeration system (104); Fan system (105); A detection device (112) is connected to at least one of the water spray system (103), the direct expansion refrigeration system (104), and the fan system (105) to detect the operating condition data of the at least one. A controller (101) configured to control the operation of at least one of the water mist system (103), the direct expansion refrigeration system (104), and the fan system (105) based on a control signal received from the control system (111) of any one of claims 16-18. The control system (111) is configured to optimize the operating cost of the indirect evaporative cooling system (100) based on the operating condition data detected by the detection device (112) to obtain the control data of the indirect evaporative cooling system (100) when the operating cost of the indirect evaporative cooling system (100) is optimized, thereby generating the control signal.

Images