Cooling tower control parameter determination method and system based on big data deep learning

Abstract

The invention discloses a method and a system for determining control parameters of a cooling tower based on big data deep learning, and relates to the technical field of energy consumption control. The method comprises the following steps: periodically obtaining an environment data set; wherein the environment data set comprises data center load, outdoor wet bulb temperature and cooling tower water inlet temperature. And the first refrigerating capacity needed by the data center is predicted based on the data center load. And on the basis of the first refrigerating capacity, the outdoor wet bulb temperature and the cooling tower water inlet temperature, the cooling tower fan frequencies needed by different starting number of the cooling tower fans are determined when the cooling tower refrigerating capacity is the first refrigerating capacity, and the starting number of the cooling tower fans and the cooling tower power corresponding to the cooling tower fan frequencies are determined. And finally, when the power of the cooling tower is the lowest, the number of started cooling tower fans and the frequency of the cooling tower fans serve as the control parameters of the cooling tower, the control parameters of the cooling tower are accurately determined, and the flexibility of determining the control parameters of the cooling tower is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of energy consumption control technology, and in particular to a method and system for determining control parameters of a cooling tower based on big data deep learning. Background Art

[0002] There are many servers and other equipment in the data center, which will generate a lot of heat when running. In the cooling system of the data center, the cooling tower is an important component, which works in coordination with the chiller, cooling water pump and other equipment.

[0003] The cooling tower relies on the circulation of water to remove heat. It first transfers the heat generated by the data center equipment to the cooling water, and then dissipates the heat into the atmosphere through the heat exchange between the air and water in the cooling tower. By adjusting the frequency and number of cooling tower fans, the operating efficiency of the cooling tower can be changed. In the prior art, the frequency and number of cooling tower fans are usually adjusted based on manual adjustment or rule-based adjustment. However, the manual adjustment and rule-based adjustment methods have the problem of inaccurate control parameter adjustment and lack of flexibility.

[0004] Therefore, how to accurately determine the cooling tower control parameters and ensure flexibility is an urgent problem that needs to be solved. Summary of the invention

[0005] The embodiments of the present invention provide a method and system for determining cooling tower control parameters based on big data deep learning, which is used to calculate the number of fan starts and the fan frequency according to the current data center load and outdoor environmental information, and finally obtain the control parameters of the cooling tower based on the cooling tower power, so as to realize dynamic adjustment of the control parameters and ensure the accuracy of the control parameters based on refined data analysis.

[0006] In a first aspect, an embodiment of the present invention provides a method for determining cooling tower control parameters based on big data deep learning, comprising:

[0007] Periodically acquiring an environmental data set, wherein the environmental data set includes a data center load, an outdoor wet bulb temperature, and a cooling tower inlet water temperature;

[0008] Predicting a first cooling capacity required by the data center based on the data center load;

[0009] Based on the first cooling capacity, the outdoor wet-bulb temperature and the cooling tower inlet water temperature, determining the cooling tower fan frequency required for different numbers of cooling tower fans started when the cooling tower cooling capacity is the first cooling capacity, and determining the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency;

[0010] When the cooling tower power is the lowest, the corresponding number of cooling tower fans started and the cooling tower fan frequency are used as the control parameters of the cooling tower. The control parameters of the cooling tower are used by the user to adjust the number of cooling tower fans started and the cooling tower fan frequency.

[0011] In the above technical scheme, the server obtains the environmental data set periodically, and calculates the cooling tower control parameters after obtaining the environmental data set. Among them, the cooling system of the data center includes at least one cooling tower, and each cooling tower includes two fans. When calculating the cooling tower control parameters in this scheme, if there are multiple cooling towers, the multiple cooling towers are regarded as one main body, and only the number of fans started and the fan frequency are considered. First, the first cooling capacity required by the data center is predicted based on the data center load to ensure that the cooling tower control parameters determined subsequently can meet the cooling requirements. Then, based on the first cooling capacity, the outdoor wet bulb temperature and the cooling tower inlet water temperature, the cooling tower fan frequency required for different cooling tower fan start-up numbers when the cooling tower cooling capacity is the first cooling capacity is determined, that is, the cooling tower fan frequency required when the number of cooling tower start-up units is i. i is an integer greater than zero and less than or equal to n. n is the total number of fans. Among them, the cooling tower fan frequency refers to the frequency of each cooling tower fan. Further determine the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency. Finally, when the cooling tower power is lowest, the corresponding number of cooling tower fans started and the cooling tower fan frequency are used as the control parameters of the cooling tower, so as to accurately determine the cooling tower control parameters and ensure the flexibility of determining the cooling tower control parameters.

[0012] Optionally, predicting a first cooling capacity required by the data center based on the data center load includes:

[0013] The data center load is input into a first model to obtain a first cooling capacity output by the first model, wherein the first model is obtained by machine learning based on historical data center loads and cooling capacities corresponding to the historical data center loads.

[0014] Optionally, the first model is obtained by machine learning based on a historical data center load and a cooling capacity corresponding to the historical data center load, and includes:

[0015] Obtain the historical data center load and the historical cooling tower inlet water temperature, historical cooling tower return water temperature, and historical cooling tower water flow corresponding to each historical data center load;

[0016] For any historical data center load, calculate the difference between the historical cooling tower inlet water temperature and the historical cooling tower return water temperature corresponding to the historical data center load, and calculate the product of the difference, the historical cooling tower water flow corresponding to the historical data center load and the specific heat capacity of water as the cooling capacity corresponding to the historical data center load;

[0017] A first model is obtained based on the historical data center loads and the cooling capacity corresponding to the historical data center loads and based on a machine learning algorithm.

[0018] In the above technical solution, machine learning is performed based on the historical data center loads and the cooling capacity corresponding to each historical data center load, so as to learn the relationship between the data center load and the cooling capacity and improve the accuracy of the subsequent determination of the control parameters.

[0019] Optionally, based on the first cooling capacity, the outdoor wet-bulb temperature and the cooling tower inlet water temperature, determining a cooling tower fan frequency required for starting different numbers of cooling tower fans when the cooling capacity of the cooling tower is the first cooling capacity, comprises:

[0020] calculating a second difference between the cooling tower inlet water temperature and the outdoor wet bulb temperature;

[0021] calculating a first product between the second difference and the first coefficient;

[0022] calculating a first ratio between the first cooling capacity and the first product, the first coefficient being used to indicate the structure and size of a cooling tower;

[0023] The ratio between the first ratio and the number of fans started in each cooling tower is calculated to obtain the cooling tower fan frequencies required for different numbers of fans started in the cooling tower.

[0024] Optionally, determining the cooling tower power corresponding to the number of fans started in each cooling tower and the fan frequency of the cooling tower includes:

[0025] For any number of cooling tower fans started, the number of cooling tower fans started and the cooling tower fan frequency corresponding to the number of cooling tower fans started are input into the second model to obtain the cooling tower power output by the second model. The second model is obtained by machine learning based on different numbers of cooling tower fans started and the corresponding cooling tower fan frequencies and historical cooling tower powers.

[0026] Optionally, the second model is obtained by machine learning based on the number of different cooling tower fans started and the corresponding cooling tower fan frequencies and historical cooling tower powers, and includes:

[0027] For any number of cooling tower fans started, obtain the historical cooling tower powers corresponding to different historical cooling tower fan frequencies when the number of cooling tower fans started is the number of cooling tower fans started;

[0028] According to the different historical cooling tower fan frequencies corresponding to the number of fans started in each cooling tower and the historical cooling tower powers corresponding to the different historical cooling tower fan frequencies, a second model is obtained according to a machine learning algorithm.

[0029] Optionally, when the cooling tower power is the lowest, the number of cooling tower fans started and the cooling tower fan frequency corresponding to the cooling tower power are used as the control parameters of the cooling tower, including:

[0030] The number of cooling tower fans started and the required cooling tower fan frequency are arranged in ascending order based on the corresponding cooling tower power, and the number of cooling tower fans started and the cooling tower fan frequency that are ranked first are used as control parameters of the cooling tower.

[0031] Optionally, after using the number of cooling tower fans started and the cooling tower fan frequency corresponding to when the cooling tower power is the lowest as the control parameters of the cooling tower, the method further includes:

[0032] The control parameters of the cooling tower and the corresponding cooling tower power are displayed to the user, so that the user can adjust the number of cooling tower fans started and the cooling tower fan frequency based on the control parameters of the cooling tower.

[0033] In a second aspect, an embodiment of the present invention provides a system for determining cooling tower control parameters based on big data deep learning, comprising:

[0034] An acquisition module, used for periodically acquiring an environmental data set, wherein the environmental data set includes a data center load, an outdoor wet-bulb temperature, and a cooling tower inlet water temperature;

[0035] A processing module, configured to predict a first cooling capacity required by a data center based on the data center load;

[0036] Based on the first cooling capacity, the outdoor wet-bulb temperature and the cooling tower inlet water temperature, determining the cooling tower fan frequency required for different numbers of cooling tower fans started when the cooling tower cooling capacity is the first cooling capacity, and determining the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency;

[0037] When the cooling tower power is the lowest, the corresponding number of cooling tower fans started and the cooling tower fan frequency are used as the control parameters of the cooling tower. The control parameters of the cooling tower are used by the user to adjust the number of cooling tower fans started and the cooling tower fan frequency.

[0038] Optionally, the processing module is specifically used for:

[0039] The data center load is input into a first model to obtain a first cooling capacity output by the first model, wherein the first model is obtained by machine learning based on historical data center loads and cooling capacities corresponding to the historical data center loads.

[0040] Optionally, the processing module is specifically used for:

[0041] Obtain the historical data center load and the historical cooling tower inlet water temperature, historical cooling tower return water temperature, and historical cooling tower water flow corresponding to each historical data center load;

[0042] For any historical data center load, calculate the difference between the historical cooling tower inlet water temperature and the historical cooling tower return water temperature corresponding to the historical data center load, and calculate the product of the difference, the historical cooling tower water flow corresponding to the historical data center load and the specific heat capacity of water as the cooling capacity corresponding to the historical data center load;

[0043] A first model is obtained based on the historical data center loads and the cooling capacity corresponding to the historical data center loads and based on a machine learning algorithm.

[0044] Optionally, the processing module is specifically used for:

[0045] calculating a second difference between the cooling tower inlet water temperature and the outdoor wet bulb temperature;

[0046] calculating a first product between the second difference and the first coefficient;

[0047] calculating a first ratio between the first cooling capacity and the first product, the first coefficient being used to indicate the structure and size of a cooling tower;

[0048] The ratio between the first ratio and the number of fans started in each cooling tower is calculated to obtain the cooling tower fan frequencies required for different numbers of fans started in the cooling tower.

[0049] Optionally, the processing module is specifically used for:

[0050] For any number of cooling tower fans started, the number of cooling tower fans started and the cooling tower fan frequency corresponding to the number of cooling tower fans started are input into the second model to obtain the cooling tower power output by the second model. The second model is obtained by machine learning based on different numbers of cooling tower fans started and the corresponding cooling tower fan frequencies and historical cooling tower powers.

[0051] Optionally, the processing module is specifically used for:

[0052] For any number of cooling tower fans started, obtain the historical cooling tower powers corresponding to different historical cooling tower fan frequencies when the number of cooling tower fans started is the number of cooling tower fans started;

[0053] According to the different historical cooling tower fan frequencies corresponding to the number of fans started in each cooling tower and the historical cooling tower powers corresponding to the different historical cooling tower fan frequencies, a second model is obtained according to a machine learning algorithm.

[0054] Optionally, the processing module is specifically used for:

[0055] The number of cooling tower fans started and the required cooling tower fan frequency are arranged in ascending order based on the corresponding cooling tower power, and the number of cooling tower fans started and the cooling tower fan frequency that are ranked first are used as control parameters of the cooling tower.

[0056] Optionally, the processing module is further used for:

[0057] The control parameters of the cooling tower and the corresponding cooling tower power are displayed to the user, so that the user can adjust the number of cooling tower fans started and the cooling tower fan frequency based on the control parameters of the cooling tower.

[0058] In a third aspect, an embodiment of the present invention further provides a computer device, including:

[0059] A memory for storing program instructions;

[0060] The processor is used to call the program instructions stored in the memory and execute the above-mentioned method for determining the cooling tower control parameters based on big data deep learning according to the obtained program.

[0061] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned method for determining cooling tower control parameters based on big data deep learning.

[0062] In a fifth aspect, an embodiment of the present invention further provides a computer program product, which includes an executable program, and the executable program is executed by a processor to implement the above-mentioned method for determining cooling tower control parameters based on big data deep learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 A schematic diagram of a system architecture provided by an embodiment of the present invention;

[0065] Figure 2A schematic flow chart of a method for determining cooling tower control parameters based on big data deep learning provided by an embodiment of the present invention;

[0066] Figure 3 A schematic diagram of a cooling tower in a refrigeration system provided by an embodiment of the present invention;

[0067] Figure 4 A schematic diagram of the structure of a system for determining cooling tower control parameters based on big data deep learning provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] The application scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions protected by the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. It is known to those of ordinary skill in the art that with the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems. The terms "first" and "second" in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In the description of the present application, unless otherwise specified, the meaning of "multiple" is two or more.

[0070] Before introducing the method for determining cooling tower control parameters based on big data deep learning provided in the embodiment of the present application, in order to facilitate understanding, the background technology and terms involved in the embodiment of the present application are first introduced below.

[0071] IT load refers to the pressure and burden generated in the IT system due to user requests, data processing and system operation.

[0072] Cooling tower: a device that uses water as a circulating coolant to absorb heat from a system and discharge it into the atmosphere to reduce the water temperature. It uses the heat exchange between water and air to produce steam, and the steam evaporates and takes away the heat to achieve evaporative heat dissipation, convection heat transfer and radiation heat transfer. It is an evaporative heat dissipation device that dissipates the waste heat generated in industry or refrigeration and air conditioning to reduce the water temperature to ensure the normal operation of the system. The device is generally barrel-shaped, hence the name cooling tower.

[0073] Outdoor wet bulb temperature: refers to the temperature when the gauze on the wet bulb thermometer cools down due to water evaporation during the isenthalpic process and reaches a stable state. This temperature reflects the cooling effect under the combined influence of air temperature and humidity, and is often used in air conditioning design and environmental assessment.

[0074] Data centers carry many servers and other heat dissipation equipment, which generate a lot of heat when running. In the refrigeration system of a data center, the cooling tower is an important component, which works in conjunction with equipment such as chillers and cooling water pumps. The effective operation of the cooling tower can reduce the condensation temperature of the chiller, thereby improving the refrigeration efficiency of the chiller. The cooling tower relies on the circulation of water to remove heat. It first transfers the heat generated by the data center equipment to the cooling water, and then dissipates the heat into the atmosphere through heat exchange between the air and water in the cooling tower.

[0075] In the cooling system of the data center, except for the terminal, the energy consumption of the cooling tower in the cooling machine mode and pre-cooling mode is second only to the energy consumption of the cooling machine; in the pure plate exchange mode, the cooling tower has the highest energy consumption, which is higher than the operating energy consumption of the refrigeration pump and the cooling pump. This shows that the cooling tower accounts for a considerable proportion of the energy consumption structure of the entire refrigeration equipment. Traditional cooling tower control methods are often extensive and have a lot of room for optimization. The frequency and number of cooling tower fans are not flexible enough to be adjusted accurately according to actual cooling needs. Therefore, achieving cooling tower energy saving based on accurate recommendation of cooling tower control parameters can significantly improve the operating efficiency of the cooling tower and achieve energy saving.

[0076] Existing cooling tower control methods include:

[0077] Manual experience method: refers to relying on the long-term practical experience accumulated by operation and maintenance personnel to recommend cooling tower frequency and number strategies. These experiences are usually based on familiarity with the cooling tower system, previous operation experience of similar projects, and intuitive judgment of the cooling needs of the data center. Existing problems: Due to differences in experience and knowledge level, operation and maintenance personnel will give different strategies under similar weather conditions and IT loads. In actual situations, the operation of cooling towers is not only affected by seasons and server loads, but also by many factors such as air humidity, water quality, and the degree of aging of the cooling tower itself. It is difficult to comprehensively consider the interaction of complex factors based on experience alone, and some important influencing factors may be ignored, resulting in inaccurate control strategies. Manual experience may not be able to quickly and accurately adjust the frequency and number strategies of cooling towers. In addition, under complex working conditions, if the fan frequency is too large, the vibration and stress generated may exceed the bearing limit of the structure, resulting in cracks in the tower body or deformation of the supporting structure. If the fan frequency is too small, it may lead to insufficient cooling effect, which in turn triggers a series of temperature-related equipment failures.

[0078] Based on the machine learning simulation method: collect a large amount of data related to the operation of the cooling tower, including but not limited to outdoor wet-bulb temperature, cooling water flow, cooling tower inlet and outlet water temperature and other data, train the machine learning model, and obtain the model. The model only needs to input the current external environment to give the optimal frequency and number of cooling towers, thereby achieving precise control of the cooling tower. Existing problems: The performance of the machine learning model depends largely on the quality and quantity of the data. If the collected data is biased or the amount of data is insufficient, the model may not be able to accurately learn the laws of cooling tower operation. With the update of data center equipment and changes in the environment, the machine learning model needs to be continuously updated. This requires re-collecting data and retraining the model, which involves high costs, including time costs, computing resource costs, etc. Moreover, during the model update process, if not handled properly, the model performance may deteriorate.

[0079] Rule-based control method: adjust the frequency and number of cooling towers according to pre-set rules. For example, different rules are set according to parameters such as the outdoor wet-bulb temperature and the load of the chiller. When the outdoor wet-bulb temperature is in a certain range and the load of the chiller is in a specific range, the frequency of the cooling tower fan is adjusted or the number of units turned on is determined according to fixed rules. Problems: Lack of flexibility, the rules are pre-set and difficult to adapt to the complex and changeable actual operating environment. For example, in extreme weather or when the layout of equipment inside the data center changes, fixed rules may not provide the optimal control strategy. It is difficult for rule-based methods to fully consider the dynamic changes of these factors, which may lead to low operating efficiency of the cooling tower, either excessive cooling causing energy waste, or insufficient cooling affecting the normal operation of the equipment.

[0080] Accurately recommending the control parameters of cooling towers can not only improve the performance of the cooling towers themselves, but also optimize the coordinated operation of the entire refrigeration system. In the data center refrigeration system, the cooling tower is an important part of the refrigeration system, and its operation consumes a lot of electricity. According to the changes in the wet-bulb temperature of the external environment, accurate recommendations on frequency and number of units can reduce the energy consumption of cooling towers and directly reduce electricity bills.

[0081] Figure 1 A system architecture applicable to an embodiment of the present invention is exemplarily shown. The system architecture includes a server 100 . The server 100 may include a processor 110 , a communication interface 120 , and a memory 130 .

[0082] The communication interface 120 is used to transmit data with the cooling tower.

[0083] The processor 110 is the control center of the server 100, and uses various interfaces and routes to connect various parts of the entire server 100, and executes various functions of the server 100 and processes data by running or executing software programs and / or modules stored in the memory 130, and calling data stored in the memory 130. Optionally, the processor 110 may include one or more processing units.

[0084] The memory 130 can be used to store software programs and modules. The processor 110 executes various functional applications and data processing by running the software programs and modules stored in the memory 130. The memory 130 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc.; the data storage area can store data created according to business processing, etc. In addition, the memory 130 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0085] It should be noted that the above Figure 1 The structure shown is only an example and is not limited to this embodiment of the present invention.

[0086] Based on the above description, Figure 2 The flowchart of the method for determining cooling tower control parameters based on big data deep learning provided by an embodiment of the present invention is exemplarily shown, and the flowchart can be executed by a system for determining cooling tower control parameters based on big data deep learning.

[0087] like Figure 2 As shown in the figure, the process specifically includes:

[0088] Step 210: periodically obtain an environmental data set, wherein the environmental data set includes data center load, outdoor wet bulb temperature, and cooling tower inlet water temperature.

[0089] In an embodiment of the present invention, the server periodically acquires an environmental data set. The period may be preset based on experience, such as acquiring an environmental data set every half an hour, and the period is not specifically limited here. For example, using Python language, Pandas reads csv data related to cooling tower energy saving, which mainly includes the current data center load, outdoor wet bulb temperature, and cooling tower inlet water temperature. The data center load is the IT load.

[0090] The heat dissipation system of the data center includes at least one cooling tower, and each cooling tower includes two fans. Figure 3 As shown, Figure 3A schematic diagram of a cooling tower in a refrigeration system provided in an embodiment of the present invention. The figure includes multiple cooling towers, which are connected to a data center through two main pipes, one main pipe is responsible for water inlet, and the other main pipe is responsible for water return. In the present invention, if there are multiple cooling towers, the multiple cooling towers are regarded as one main body, and only the number of fans and fan frequency therein are considered. It can be understood that since the multiple cooling towers are regarded as one main body, the cooling tower inlet water temperature and the cooling tower return water temperature mentioned in the present invention refer to the inlet water temperature of the cooling tower main pipe and the return water temperature of the cooling tower main pipe.

[0091] Step 220: predicting a first cooling capacity required by the data center based on the data center load.

[0092] In an embodiment of the present invention, the server predicts the first cooling capacity required by the data center based on the data center load in the environmental data set. Specifically, the data center load is input into the first model to obtain the first cooling capacity output by the first model. The first model is obtained by machine learning based on the historical data center load and the cooling capacity corresponding to the historical data center load, and is used to predict the cooling capacity required by the data center according to the data center load.

[0093] The process of establishing the first model specifically includes:

[0094] Obtain the historical data center load and the historical cooling tower inlet water temperature, historical cooling tower return water temperature, and historical cooling tower water flow corresponding to each historical data center load. For any historical data center load, calculate the difference between the historical cooling tower inlet water temperature and the historical cooling tower return water temperature corresponding to the historical data center load, and calculate the product of the difference, the historical cooling tower water flow corresponding to the historical data center load, and the specific heat capacity of water as the cooling capacity corresponding to the historical data center load. Exemplarily, the cooling capacity is calculated using the python language according to the following formula:

[0095] Cooling capacity = (historical cooling tower inlet water temperature - historical cooling tower return water temperature) * historical cooling tower water flow * specific heat capacity of water

[0096] After obtaining the cooling capacity corresponding to each historical data center load, a first model is obtained based on the machine learning algorithm according to each historical data center load and the cooling capacity corresponding to each historical data center load. The first model is a regression model with data center load as input and cooling capacity as output.

[0097] In some possible embodiments, when predicting the first cooling capacity required by the data center based on the data center load, it can also be determined based on the historical data center load and the corresponding cooling capacity. Specifically, after obtaining the historical data center load and the historical cooling tower inlet water temperature, historical cooling tower return water temperature, and historical cooling tower water flow corresponding to each historical data center load, the cooling capacity corresponding to each historical data center load is calculated based on the above method. Then, a correlation relationship between each historical data center load and the corresponding cooling capacity is established. When performing cooling capacity prediction, the first cooling capacity required by the data center can be determined based on the correlation relationship between the data center load and the cooling capacity. However, in the above method, if the number of historical data center loads is insufficient, there may be a problem of being unable to determine the cooling capacity. Therefore, in the present invention, machine learning is performed based on the historical data center load and the cooling capacity corresponding to each historical data center load, the relationship between the data center load and the cooling capacity is learned, and a first model is established.

[0098] Step 230, based on the first cooling capacity, the outdoor wet-bulb temperature and the cooling tower inlet water temperature, determine the cooling tower fan frequency required for different numbers of cooling tower fans started when the cooling capacity of the cooling tower is the first cooling capacity, and determine the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency.

[0099] In an embodiment of the present invention, the server will perform optimal recommendation based on the first cooling capacity, the outdoor wet bulb temperature and the cooling tower inlet water temperature. First, determine the cooling tower fan frequency required for different numbers of cooling tower fans to be started when the cooling capacity of the cooling tower is the first cooling capacity, that is, the cooling tower fan frequency required when the number of cooling tower fans started is i. i is an integer greater than zero and less than or equal to n. n is the total number of fans. Among them, the cooling tower fan frequency refers to the frequency of each cooling tower fan. Exemplarily, assuming that the number of cooling tower fans is 4, then determine the cooling tower fan frequency required when the number of cooling tower fans started is 1 when the cooling capacity of the cooling tower is the first cooling capacity, the cooling tower fan frequency required when the number of cooling tower fans started is 2, the cooling tower fan frequency required when the number of cooling tower fans started is 3, and the cooling tower fan frequency required when the number of cooling tower fans started is 4.

[0100] The specific determination method includes: first, calculating the second difference between the cooling tower inlet water temperature and the outdoor wet bulb temperature. Then, calculating the first product between the second difference and the first coefficient. Calculating the first ratio between the first cooling capacity and the first product. Among them, the first coefficient is used to indicate the structure and size of the cooling tower. Finally, calculating the ratio between the first ratio and the number of fans started in each cooling tower to obtain the cooling tower fan frequency required for different numbers of cooling tower fans started. Exemplarily, using Python language, the cooling tower fan frequency required for different numbers of cooling tower fans started is calculated according to the following formula:

[0101] Cooling capacity = cooling tower cooling capacity = number of cooling tower fans * cooling tower operation frequency * (cooling main water inlet temperature – outdoor wet bulb temperature) * k

[0102] Among them, k is the first coefficient, which is a coefficient related to the structure and size of the cooling tower.

[0103] Then, the server determines the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency. For example, continuing with the above example, the cooling tower power corresponding to the cooling tower fan frequency required when the number of cooling tower fans started is 1, the cooling tower power corresponding to the cooling tower fan frequency required when the number of cooling tower fans started is 2, the cooling tower power corresponding to the cooling tower fan frequency required when the number of cooling tower fans started is 3, and the cooling tower power corresponding to the cooling tower fan frequency required when the number of cooling tower fans started is 4 are determined.

[0104] The specific determination method includes: for any number of cooling tower fans started, the number of cooling tower fans started and the cooling tower fan frequency corresponding to the number of cooling tower fans started are input into the second model to obtain the cooling tower power output by the second model. The second model is obtained by machine learning based on different numbers of cooling tower fans started and the corresponding cooling tower fan frequencies and historical cooling tower powers.

[0105] The establishment process of the second model specifically includes:

[0106] For any number of cooling tower fans started, obtain the historical cooling tower powers corresponding to different historical cooling tower fan frequencies when the number of cooling tower fans started is the number of cooling tower fans started. According to the different historical cooling tower fan frequencies corresponding to the number of cooling tower fans started and the historical cooling tower powers corresponding to different historical cooling tower fan frequencies, a second model is obtained according to the machine learning algorithm. The second model is a regression model with the cooling tower fan frequency and the number of cooling tower fans started as input and the cooling tower power as output.

[0107] In a possible implementation, when the server determines the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency, it can also be determined based on the number of cooling tower fans started and the corresponding different historical cooling tower fan frequencies and historical cooling tower powers. Specifically, for any number of cooling tower fans started, the historical cooling tower power corresponding to different historical cooling tower fan frequencies when the number of cooling tower fans started is the number of cooling tower fans started is obtained. Then, the correlation between different historical cooling tower fan frequencies and historical cooling tower powers under different numbers of cooling tower fans started is established. When determining the cooling tower power, the cooling tower fairness rate can be determined according to the correlation between the number of cooling tower fans started and the cooling tower fan frequency and the cooling tower power. However, in the above method, if the number of historical data is insufficient, there may be a problem that the cooling tower frequency cannot be determined. Therefore, in the present invention, the different historical cooling tower fan frequencies corresponding to the number of cooling tower fans started and the historical cooling tower powers corresponding to the different historical cooling tower fan frequencies are machine-learned to learn the relationship between the number of cooling tower fans started and the cooling tower fan frequency and the cooling tower power, and establish a second model.

[0108] Step 240, when the cooling tower power is the lowest, the corresponding number of cooling tower fans started and the cooling tower fan frequency are used as control parameters of the cooling tower, and the control parameters of the cooling tower are used by the user to adjust the number of cooling tower fans started and the cooling tower fan frequency.

[0109] In an embodiment of the present invention, the server uses the number of cooling tower fans started and the cooling tower fan frequency corresponding to the lowest cooling tower power as the control parameters of the cooling tower. The control parameters of the cooling tower are used by the user to adjust the number of cooling tower fans started and the cooling tower fan frequency. Specifically, the number of cooling tower fans started and the required cooling tower fan frequency are arranged in ascending order based on the corresponding cooling tower power, and the number of cooling tower fans started and the cooling tower fan frequency ranked first are used as the control parameters of the cooling tower.

[0110] Finally, the server displays the control parameters of the cooling tower and the corresponding cooling tower power to the user, so that the user can adjust the number of cooling tower fans started and the cooling tower fan frequency based on the control parameters of the cooling tower.

[0111] In the embodiment of the present invention, the control parameters of the cooling tower can be dynamically adjusted and optimized in real time according to the current data center load and outdoor environment, and the most energy-saving cooling tower recommendation strategy can be given on the basis of meeting the current environment. And dynamically determining the control parameters of the cooling tower can reduce equipment wear, and the equipment can avoid operating under unreasonable conditions, stabilizing system operation. In scenarios such as data centers, IT loads and outdoor environments are constantly changing. By dynamically adjusting the control parameters of the cooling tower, the needs of the entire cooling system can be better matched.

[0112] Based on the same technical concept, Figure 4 The structural schematic diagram of the system for determining the control parameters of a cooling tower based on big data deep learning provided by an embodiment of the present invention is shown by way of example. The device can execute the process of the method for determining the control parameters of a cooling tower based on big data deep learning.

[0113] like Figure 4 As shown, the system specifically includes:

[0114] An acquisition module 410 is used to periodically acquire an environmental data set, wherein the environmental data set includes a data center load, an outdoor wet bulb temperature, and a cooling tower inlet water temperature;

[0115] A processing module 420, configured to predict a first cooling capacity required by the data center based on the data center load;

[0116] Based on the first cooling capacity, the outdoor wet-bulb temperature and the cooling tower inlet water temperature, determining the cooling tower fan frequency required for different numbers of cooling tower fans started when the cooling tower cooling capacity is the first cooling capacity, and determining the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency;

[0117] When the cooling tower power is the lowest, the corresponding number of cooling tower fans started and the cooling tower fan frequency are used as the control parameters of the cooling tower. The control parameters of the cooling tower are used by the user to adjust the number of cooling tower fans started and the cooling tower fan frequency.

[0118] Optionally, the processing module 420 is specifically used for:

[0119] The data center load is input into a first model to obtain a first cooling capacity output by the first model, wherein the first model is obtained by machine learning based on historical data center loads and cooling capacities corresponding to the historical data center loads.

[0120] Optionally, the processing module 420 is specifically used for:

[0121] Obtain the historical data center load and the historical cooling tower inlet water temperature, historical cooling tower return water temperature, and historical cooling tower water flow corresponding to each historical data center load;

[0122] For any historical data center load, calculate the difference between the historical cooling tower inlet water temperature and the historical cooling tower return water temperature corresponding to the historical data center load, and calculate the product of the difference, the historical cooling tower water flow corresponding to the historical data center load and the specific heat capacity of water as the cooling capacity corresponding to the historical data center load;

[0123] A first model is obtained based on the historical data center loads and the cooling capacity corresponding to the historical data center loads and based on a machine learning algorithm.

[0124] Optionally, the processing module 420 is specifically used for:

[0125] calculating a second difference between the cooling tower inlet water temperature and the outdoor wet bulb temperature;

[0126] calculating a first product between the second difference and the first coefficient;

[0127] calculating a first ratio between the first cooling capacity and the first product, the first coefficient being used to indicate the structure and size of a cooling tower;

[0128] The ratio between the first ratio and the number of fans started in each cooling tower is calculated to obtain the cooling tower fan frequencies required for different numbers of fans started in the cooling tower.

[0129] Optionally, the processing module 420 is specifically used for:

[0130] For any number of cooling tower fans started, the number of cooling tower fans started and the cooling tower fan frequency corresponding to the number of cooling tower fans started are input into the second model to obtain the cooling tower power output by the second model. The second model is obtained by machine learning based on different numbers of cooling tower fans started and the corresponding cooling tower fan frequencies and historical cooling tower powers.

[0131] Optionally, the processing module 420 is specifically used for:

[0132] For any number of cooling tower fans started, obtain the historical cooling tower powers corresponding to different historical cooling tower fan frequencies when the number of cooling tower fans started is the number of cooling tower fans started;

[0133] According to the different historical cooling tower fan frequencies corresponding to the number of fans started in each cooling tower and the historical cooling tower powers corresponding to the different historical cooling tower fan frequencies, a second model is obtained according to a machine learning algorithm.

[0134] Optionally, the processing module 420 is specifically used for:

[0135] The number of cooling tower fans started and the required cooling tower fan frequency are arranged in ascending order based on the corresponding cooling tower power, and the number of cooling tower fans started and the cooling tower fan frequency that are ranked first are used as control parameters of the cooling tower.

[0136] Optionally, the processing module 420 is further configured to:

[0137] The control parameters of the cooling tower and the corresponding cooling tower power are displayed to the user, so that the user can adjust the number of cooling tower fans started and the cooling tower fan frequency based on the control parameters of the cooling tower.

[0138] Based on the same technical concept, an embodiment of the present invention further provides a computer device, including:

[0139] A memory for storing program instructions;

[0140] The processor is used to call the program instructions stored in the memory and execute the above-mentioned method for determining the cooling tower control parameters based on big data deep learning according to the obtained program.

[0141] Based on the same technical concept, an embodiment of the present invention also provides a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned method for determining cooling tower control parameters based on big data deep learning.

[0142] Based on the same technical concept, an embodiment of the present invention also provides a computer program product, characterized in that the computer program product includes an executable program, which is executed by a processor to implement the above-mentioned method for determining cooling tower control parameters based on big data deep learning.

[0143] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0144] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0145] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0146] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0147] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for determining cooling tower control parameters based on big data deep learning, characterized in that: include: Periodically acquiring an environmental data set, wherein the environmental data set includes a data center load, an outdoor wet bulb temperature, and a cooling tower inlet water temperature; Predicting a first cooling capacity required by the data center based on the data center load; Based on the first cooling capacity, the outdoor wet-bulb temperature and the cooling tower inlet water temperature, determining the cooling tower fan frequency required for different numbers of cooling tower fans started when the cooling tower cooling capacity is the first cooling capacity, and determining the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency; When the cooling tower power is the lowest, the corresponding number of cooling tower fans started and the cooling tower fan frequency are used as the control parameters of the cooling tower. The control parameters of the cooling tower are used by the user to adjust the number of cooling tower fans started and the cooling tower fan frequency.

2. The method according to claim 1, characterized in that Predicting a first cooling capacity required by the data center based on the data center load includes: The data center load is input into a first model to obtain a first cooling capacity output by the first model, wherein the first model is obtained by machine learning based on historical data center loads and cooling capacities corresponding to the historical data center loads.

3. The method according to claim 2, characterized in that The first model is obtained by machine learning based on the historical data center load and the cooling capacity corresponding to the historical data center load, and includes: Obtain the historical data center load and the historical cooling tower inlet water temperature, historical cooling tower return water temperature, and historical cooling tower water flow corresponding to each historical data center load; For any historical data center load, calculate the difference between the historical cooling tower inlet water temperature and the historical cooling tower return water temperature corresponding to the historical data center load, and calculate the product of the difference, the historical cooling tower water flow corresponding to the historical data center load and the specific heat capacity of water as the cooling capacity corresponding to the historical data center load; A first model is obtained based on the historical data center loads and the cooling capacity corresponding to the historical data center loads and based on a machine learning algorithm.

4. The method according to claim 1, characterized in that Determining, based on the first cooling capacity, the outdoor wet-bulb temperature, and the cooling tower inlet water temperature, when the cooling tower cooling capacity is the first cooling capacity, a cooling tower fan frequency required for different numbers of cooling tower fans to be started, including: calculating a second difference between the cooling tower inlet water temperature and the outdoor wet bulb temperature; calculating a first product between the second difference and the first coefficient; calculating a first ratio between the first cooling capacity and the first product, the first coefficient being used to indicate the structure and size of a cooling tower; The ratio between the first ratio and the number of fans started in each cooling tower is calculated to obtain the cooling tower fan frequencies required for different numbers of fans started in the cooling tower.

5. The method according to claim 1, characterized in that Determine the cooling tower power corresponding to the number of fans started and the cooling tower fan frequency, including: For any number of cooling tower fans started, the number of cooling tower fans started and the cooling tower fan frequency corresponding to the number of cooling tower fans started are input into the second model to obtain the cooling tower power output by the second model. The second model is obtained by machine learning based on different numbers of cooling tower fans started and the corresponding cooling tower fan frequencies and historical cooling tower powers.

6. The method according to claim 5, characterized in that The second model is obtained by machine learning based on the number of different cooling tower fans started and the corresponding cooling tower fan frequency and historical cooling tower power, including: For any number of cooling tower fans started, obtain the historical cooling tower powers corresponding to different historical cooling tower fan frequencies when the number of cooling tower fans started is the number of cooling tower fans started; According to the different historical cooling tower fan frequencies corresponding to the number of fans started in each cooling tower and the historical cooling tower powers corresponding to the different historical cooling tower fan frequencies, a second model is obtained according to a machine learning algorithm.

7. The method according to claim 1, characterized in that When the cooling tower power is the lowest, the corresponding number of cooling tower fans started and the cooling tower fan frequency are used as the control parameters of the cooling tower, including: The number of cooling tower fans started and the required cooling tower fan frequency are arranged in ascending order based on the corresponding cooling tower power, and the number of cooling tower fans started and the cooling tower fan frequency that are ranked first are used as control parameters of the cooling tower.

8. The method according to claim 1, characterized in that After using the number of cooling tower fans started and the cooling tower fan frequency corresponding to the lowest cooling tower power as control parameters of the cooling tower, the method further includes: The control parameters of the cooling tower and the corresponding cooling tower power are displayed to the user, so that the user can adjust the number of cooling tower fans started and the cooling tower fan frequency based on the control parameters of the cooling tower.

9. A system for determining cooling tower control parameters based on big data deep learning, characterized in that: include: An acquisition module, used for periodically acquiring an environmental data set, wherein the environmental data set includes a data center load, an outdoor wet-bulb temperature, and a cooling tower inlet water temperature; A processing module, configured to predict a first cooling capacity required by a data center based on the data center load; Based on the first cooling capacity, the outdoor wet-bulb temperature and the cooling tower inlet water temperature, determining the cooling tower fan frequency required for different numbers of cooling tower fans started when the cooling tower cooling capacity is the first cooling capacity, and determining the cooling tower power corresponding to the number of cooling tower fans started and the cooling tower fan frequency; When the cooling tower power is the lowest, the corresponding number of cooling tower fans started and the cooling tower fan frequency are used as the control parameters of the cooling tower. The control parameters of the cooling tower are used by the user to adjust the number of cooling tower fans started and the cooling tower fan frequency.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the method according to any one of claims 1 to 8.

Images