An energy-saving method, device, equipment and medium of a data center
By combining big data and artificial intelligence technologies, the control parameters of data center cooling equipment are accurately recommended, solving the problem of energy saving and consumption reduction in data centers, achieving high efficiency and energy saving, and improving the accuracy and applicability of the model.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-01-13
- Publication Date
- 2026-03-27
Smart Images
Figure CN116225187B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of data center energy saving, and particularly relates to a data center energy saving method, device, equipment and medium. BACKGROUND
[0002] A data center is a network of specific devices for global cooperation, used to deliver, accelerate, display, calculate and store data information on the Internet network infrastructure. Under the premise of ensuring the safe operation of IT (Internet Technology) services, data center refrigeration technology is very important, and at the same time, excessive cooling capacity also causes serious energy consumption. Energy consumption is a relatively important problem at present, and energy saving is also a requirement for the development of future data centers.
[0003] At present, data centers cannot meet the requirements of energy saving and consumption reduction by relying on hardware energy saving or manual experience optimization. Therefore, how to realize energy saving and consumption reduction of data centers is a problem to be solved. SUMMARY
[0004] The present application provides a data center energy saving method, device, equipment and medium, which accurately recommends parameter values of controllable parameters of each refrigeration device of the data center, so as to realize energy saving and consumption reduction of the data center.
[0005] The specific technical solutions provided by the embodiments of the present application are as follows:
[0006] In a first aspect, the embodiments of the present application provide a data center energy saving method, comprising:
[0007] For each controllable parameter in each refrigeration device of the data center, the parameter values in the parameter value limit range of the controllable parameter are traversed respectively, wherein each time the traversal is performed, the following operations are performed:
[0008] Based on a series connection model composed of function prediction models of the refrigeration devices, and in combination with the parameter values of the controllable parameters in this traversal, parameter values of function parameters of the refrigeration devices are obtained;
[0009] If the parameter values of the function parameters of the refrigeration devices meet the safety condition, then through an energy consumption prediction model of any refrigeration device, based on the parameter values of the function parameters of the refrigeration device, a predicted energy consumption of the refrigeration device is obtained;
[0010] Based on the predicted energy consumptions of the refrigeration devices, a predicted total energy consumption is obtained;
[0011] Based on the predicted total energy consumption obtained through multiple iterations, a predicted total energy consumption satisfying a preset condition is selected, and the parameter value of each controllable parameter corresponding to the selected predicted total energy consumption is taken as a recommended parameter value combination.
[0012] In an optional embodiment, the method further comprises:
[0013] Through the business safety prediction model, a safety measurement point predicted value is obtained based on the parameter value of each functional parameter of each refrigeration device.
[0014] If the safety measurement point predicted value is within a safety value range, it is determined that the parameter value of each functional parameter of each refrigeration device satisfies a safety condition.
[0015] In an optional embodiment, the method further comprises:
[0016] For the functional prediction model of any refrigeration device, a first prediction index of the functional prediction model is monitored, and if the first prediction index does not satisfy a first index condition, an alarm information is sent out.
[0017] For the energy consumption prediction model of any refrigeration device, a second prediction index of the energy consumption prediction model is monitored, and if the second prediction index does not satisfy a second index condition, an alarm information is sent out.
[0018] In an optional embodiment, the functional prediction model of any refrigeration device in the plurality of refrigeration devices is obtained in the following manner:
[0019] Based on historical operation data of the refrigeration device, or factory equipment performance data and historical operation data of the refrigeration device, a first training data set is obtained, each first training data including a historical parameter value of a functional parameter of the refrigeration device, and an initial functional prediction model of the refrigeration device is trained based on the first training data set to obtain the functional prediction model of the refrigeration device; or
[0020] Based on factory equipment performance data of the refrigeration device, a reference parameter value of a functional parameter of the refrigeration device is obtained, and based on the reference parameter value of the functional parameter, a functional prediction function of the refrigeration device is constructed, and the functional prediction function is taken as the functional prediction model.
[0021] In an optional embodiment, the energy consumption prediction model of any refrigeration device in the plurality of refrigeration devices is obtained in the following manner:
[0022] obtaining a second training data set based on historical operation data of the refrigeration equipment, or factory equipment performance data and historical operation data of the refrigeration equipment, each second training data comprising historical energy consumption of the refrigeration equipment and historical parameter values of corresponding function parameters, and training the initial energy consumption prediction model of the refrigeration equipment based on the second training data set to obtain a function prediction model of the refrigeration equipment; or
[0023] obtaining reference energy consumption of the refrigeration equipment and reference parameter values of corresponding function parameters based on factory equipment performance data of the refrigeration equipment, and constructing an energy consumption prediction function of the refrigeration equipment based on the reference energy consumption and the reference parameter values of corresponding function parameters, and taking the energy consumption prediction function as the energy consumption prediction model.
[0024] In an optional implementation, the business safety prediction model is obtained by the following manner:
[0025] obtaining a third training data set based on historical operation data of the refrigeration equipment, each third training data comprising historical safety measurement point values and historical parameter values of function parameters of the refrigeration equipment, and training an initial business safety prediction model based on the third training data set to obtain the business safety prediction model.
[0026] In an optional implementation, the method further comprises:
[0027] monitoring a third prediction index of the business safety prediction model, and issuing an alarm information if the third prediction index does not satisfy a third index condition.
[0028] In a second aspect, an energy saving device of a data center is provided, comprising:
[0029] an energy consumption prediction module, configured to traverse parameter values of each controllable parameter in each refrigeration equipment of the data center in a parameter value limit range of the controllable parameter, wherein each time the traversal is performed, the following operations are performed:
[0030] obtaining parameter values of function parameters of each refrigeration equipment based on a series connection model composed of function prediction models of the refrigeration equipment and the parameter values of the controllable parameters obtained in the current traversal;
[0031] if the parameter values of the function parameters of each refrigeration equipment satisfy a safety condition, obtaining predicted energy consumption of any refrigeration equipment based on the parameter values of the function parameters of the any refrigeration equipment through an energy consumption prediction model of the any refrigeration equipment;
[0032] obtaining predicted total energy consumption based on the predicted energy consumption of each refrigeration equipment;
[0033] The parameter recommendation module is configured to select a predicted total energy consumption meeting a preset condition based on the predicted total energy consumptions obtained through multiple iterations, and combine parameter values of the controllable parameters corresponding to the selected predicted total energy consumption as a recommended parameter value group.
[0034] In an alternative embodiment, the device further comprises a safety determination module configured to:
[0035] obtain a safety measurement point predicted value based on the parameter values of the respective functional parameters of the respective refrigeration equipment through the business safety prediction model;
[0036] If the safety measurement point predicted value is within a safety value range, it is determined that the parameter values of the respective functional parameters of the respective refrigeration equipment meet a safety condition.
[0037] In an alternative embodiment, the device further comprises:
[0038] The first alarm module is configured to monitor a first prediction index of the functional prediction model of any refrigeration equipment, and if the first prediction index does not meet a first index condition, an alarm information is sent out.
[0039] The second alarm module is configured to monitor a second prediction index of the energy consumption prediction model of any refrigeration equipment, and if the second prediction index does not meet a second index condition, an alarm information is sent out.
[0040] In an alternative embodiment, the device further comprises a functional model construction module configured to obtain a functional prediction model of any refrigeration equipment in the refrigeration equipment through the following manner:
[0041] obtain a first training data set based on historical operation data of the refrigeration equipment, or factory equipment performance data and historical operation data of the refrigeration equipment, each first training data comprising a historical parameter value of a functional parameter of the refrigeration equipment, and train an initial functional prediction model of the refrigeration equipment based on the first training data set to obtain the functional prediction model of the refrigeration equipment; or
[0042] obtain a reference parameter value of a functional parameter of the refrigeration equipment based on factory equipment performance data of the refrigeration equipment, and construct a functional prediction function of the refrigeration equipment based on the reference parameter value of the functional parameter, and take the functional prediction function as the functional prediction model.
[0043] In an alternative embodiment, the device further comprises an energy consumption model construction module configured to obtain an energy consumption prediction model of any refrigeration equipment in the refrigeration equipment through the following manner:
[0044] obtaining a second training data set based on historical operation data of the refrigeration equipment, or factory equipment performance data and the historical operation data of the refrigeration equipment, each second training data comprising historical energy consumption of the refrigeration equipment and historical parameter values of corresponding function parameters, and training the initial energy consumption prediction model of the refrigeration equipment based on the second training data set to obtain a function prediction model of the refrigeration equipment; or
[0045] obtaining reference energy consumption of the refrigeration equipment and reference parameter values of corresponding function parameters based on factory equipment performance data of the refrigeration equipment, and constructing an energy consumption prediction function of the refrigeration equipment based on the reference energy consumption and the reference parameter values of corresponding function parameters, and taking the energy consumption prediction function as the energy consumption prediction model.
[0046] In an optional implementation, the apparatus further comprises a security model construction module configured to obtain a service security prediction model by:
[0047] obtaining a third training data set based on historical operation data of the refrigeration equipment, each third training data comprising historical security measurement point values and historical parameter values of function parameters of the refrigeration equipment, and training an initial service security prediction model based on the third training data set to obtain the service security prediction model.
[0048] In an optional implementation, the apparatus further comprises a third alarm module configured to:
[0049] monitoring a third prediction index of the service security prediction model, and issuing an alarm information if the third prediction index does not satisfy a third index condition.
[0050] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes steps of any method in the first aspect.
[0051] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and when the computer program is executed by a processor, steps of any method in the first aspect are implemented.
[0052] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising a computer program stored in a computer readable storage medium, and when a processor of an electronic device reads the computer program from the computer readable storage medium, the processor executes the computer program, so that the electronic device executes steps of any method in the first aspect.
[0053] The embodiments of the present application have at least the following beneficial effects:
[0054] In the scheme of the embodiments of the present application, in order to realize energy saving of the data center, suitable parameter value combinations are recommended for each controllable parameter in each refrigeration device of the data center, so as to reduce energy consumption as much as possible. In order to accurately recommend the parameter value combinations, the parameter values of each controllable parameter are traversed within the parameter value limit range of each controllable parameter. Each time the traversal is performed, the following operations are performed based on the parameter value combination obtained by the traversal: based on a series connection model composed of function prediction models of each refrigeration device, the parameter values of the function parameters of each refrigeration device are obtained in combination with the parameter value combination of this time, if the parameter values of these function parameters meet the safety condition, the predicted energy consumption of each refrigeration device can be obtained based on the parameter values of the function parameters of each refrigeration device through the corresponding energy consumption prediction model, and then the predicted total energy consumption is obtained based on the predicted energy consumption of each refrigeration device; then, for the predicted total energy consumption obtained by multiple traversals, the predicted total energy consumption meeting the preset condition is selected, and the parameter values of each controllable parameter corresponding to the selected predicted total energy consumption are taken as the recommended parameter value combination. BRIEF DESCRIPTION OF DRAWINGS
[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0056] Figure 1 A flowchart of an energy saving method of a data center provided in the embodiments of the present application;
[0057] Figure 2 An implementation process diagram of an energy saving method of a data center provided in the embodiments of the present application;
[0058] Figure 3 A structural block diagram of an energy saving device of a data center provided in the embodiments of the present application;
[0059] Figure 4 A structural block diagram of another energy saving device of a data center provided in the embodiments of the present application;
[0060] Figure 5 A structural block diagram of an electronic device provided in the embodiments of the present application. DETAILED DESCRIPTION
[0061] In order to enable personnel in the technical field to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described in the following in combination with the drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor should fall within the scope of protection of the present application.
[0062] In the description of the present application, it should be understood that the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.
[0063] The energy consumption problem of data center refrigeration technology is a relatively important problem at present, and energy saving and consumption reduction is also the development requirement of future data center. At present, data center cannot meet the requirement of energy saving and consumption reduction by relying on hardware energy saving or manual experience optimization. Therefore, how to realize energy saving and consumption reduction of data center is a problem to be solved.
[0064] Therefore, the present application provides a data center energy saving method, device, equipment and medium, which combines big data and AI (Artificial Intelligence) technology on the premise of having valuable data resources and rich business expert experience of each refrigeration equipment of the data center, simulates the heating process of the data center, combines the performance of the refrigeration equipment itself, accurately recommends the control parameters of each refrigeration equipment, meets the normal operation of IT service on the premise, enables energy consumption to meet the demand, so as to realize energy saving and consumption reduction of the data center.
[0065] The preferred embodiments of the present application are described below in combination with the drawings of the specification. It should be understood that the preferred embodiments described here are only used to illustrate and explain the present application, and are not used to limit the present application, and the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.
[0066] The data center energy saving method and data decryption method of the present application will be described in detail below in combination with the drawings and specific embodiments.
[0067] The energy saving method of the data center of the application can be executed by the control device of each refrigeration equipment of the data center. For example, the control device can be a server, which can be a physical server, a server cluster or a distributed system composed of multiple physical servers, a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN (Content Delivery Network), and big data and artificial intelligence platform, etc.
[0068] As shown in the energy saving method of the data center provided by the embodiment of the application, the following steps S101-S202 are included: Figure 1
[0069] Step S101, for each controllable parameter in each refrigeration equipment of the data center, each parameter value in the parameter value limit range of the controllable parameter is traversed, wherein each time the following steps S1011-S1013 are executed.
[0070] The device type of each refrigeration equipment can be determined according to the specific application scenario, and each controllable parameter in each refrigeration equipment can also be determined according to the device type, which is not limited. For example, each refrigeration equipment includes but is not limited to a refrigeration pump, a cooling machine, a plate heat exchanger, a cooling pump, a cooling tower, etc.; the controllable parameters involved include but are not limited to the refrigeration pump frequency, the cooling pump frequency, the number of cooling towers, the cooling water outlet temperature of the cooling machine, etc.
[0071] In order to ensure the safe operation of each refrigeration equipment, each controllable parameter corresponds to a parameter value limit range, and the parameter value can be adjusted within the parameter value limit range. For example, the cooling water outlet temperature of the cooling machine is in the range of 10℃≤T≤14℃, and the cooling water outlet temperature of the cooling machine can be adjusted between 10℃ and 14Hz℃.
[0072] Step S1011, based on the series connection model composed of the function prediction models of each refrigeration equipment, the parameter values of the function parameters of each refrigeration equipment are obtained in combination with the parameter values of each controllable parameter in this iteration.
[0073] Each refrigeration equipment can correspond to a function prediction model, which is used to predict the corresponding output function parameter according to the input function parameter of the refrigeration equipment, and the input function parameter and the output function parameter of each refrigeration equipment are both regarded as function parameters. For example, the input function parameter of the cooling pump is the cooling pump frequency, and the output function parameter is the cooling pump flow. It should be noted that the input function parameter of a refrigeration equipment may or may not contain a controllable parameter.
[0074] The function prediction model of each refrigeration device is connected in series according to the data center heating and ventilation process to obtain a series connection model. For example, the data center heating and ventilation process can be: a refrigeration pump, a refrigeration machine, a plate heat exchanger, and a cooling pump are connected in series, and finally connected to a cooling tower. In this way, the output function parameter of one refrigeration device can be used as the input function parameter of the next refrigeration device in series, and the input function parameter of one refrigeration device can include one or more.
[0075] Specifically, the series connection model can include a chilled water instantaneous flow model, a cooling water instantaneous flow model, a refrigeration machine cooling water return water temperature model, and the like, which are not limited.
[0076] In step S1012, if the parameter values of the function parameters of each refrigeration device meet the safety condition, the predicted energy consumption of any refrigeration device is obtained based on the parameter values of the function parameters of any refrigeration device through the energy consumption prediction model of any refrigeration device.
[0077] After obtaining the parameter values of the function parameters of each refrigeration device, it can be determined whether these parameter values meet the safety condition. Specifically, based on these parameter values, the value of the safety measurement point can be predicted, and then it is determined whether the value of the safety measurement point is within the safety value range. The safety measurement point can be one or more, for example, the safety measurement point includes but is not limited to the most unfavorable end pressure difference of the chilled side mother pipe, the water inlet temperature of the cold storage tank, and the temperature of the lower tower, which are not limited.
[0078] In an optional embodiment, taking one safety measurement point as an example, when determining whether the parameter values of the function parameters meet the safety condition, the following operations A1-A2 can be performed:
[0079] A1, obtaining a safety measurement point prediction value based on the parameter values of the function parameters of each refrigeration device through a business safety prediction model.
[0080] Specifically, by inputting the parameter values of the function parameters of each refrigeration device into the business safety prediction model, the corresponding safety measurement point prediction value can be obtained.
[0081] A2, if the safety measurement point prediction value is within the safety value range, it is determined that the parameter values of the function parameters of each refrigeration device meet the safety condition.
[0082] The safety value range of the safety measurement point can be set as needed, which is not limited. Taking the most unfavorable end pressure difference of the chilled side mother pipe as an example, assuming that its safety value range is 80Kpa≤P≤150Kpa, when the prediction value of the most unfavorable end pressure difference of the chilled side mother pipe is between 80-150Kpa, it is considered that the parameter values of the function parameters of each refrigeration device meet the safety condition.
[0083] When it is determined that the parameter values of the function parameters of each refrigeration device satisfy the safety condition, the energy consumption, for example, power, of each refrigeration device can be predicted based on the parameter values of the function parameters of each refrigeration device. Specifically, each refrigeration device corresponds to an energy consumption prediction model, and the parameter values of the function parameters of each refrigeration device are input into the corresponding energy consumption prediction model to obtain the predicted energy consumption of the refrigeration device.
[0084] In step S1013, the predicted total energy consumption is obtained based on the predicted energy consumptions of the refrigeration devices.
[0085] Specifically, the predicted energy consumptions of the refrigeration devices are added to obtain the predicted total energy consumption.
[0086] In step S102, the predicted total energy consumption that satisfies the preset condition is selected based on the predicted total energy consumptions obtained through multiple iterations, and the parameter values of the controllable parameters corresponding to the selected predicted total energy consumption are taken as the recommended parameter value combination.
[0087] Among the predicted total energy consumptions obtained through multiple iterations, the smallest predicted total energy consumption is selected, and the parameter values of the controllable parameters corresponding to the smallest predicted total energy consumption are taken as the recommended parameter value combination.
[0088] In the embodiments of the present application, when the parameter values of the controllable parameters are recommended, the parameter value combinations of the controllable parameters are iterated, and the parameter value combinations that satisfy the safety condition are selected from all the iterated parameter value combinations. Then, the energy consumptions of the refrigeration devices are predicted under each parameter value combination, and the predicted total energy consumptions are obtained. Finally, the parameter value combination whose predicted total energy consumption satisfies the preset condition is selected from the selected parameter value combinations as the final recommended parameter value combination. In this way, the parameter values of the controllable parameters of the refrigeration devices in the data center can be accurately recommended to achieve energy saving and consumption reduction of the data center.
[0089] The construction processes of the function prediction model, the energy consumption prediction model and the business safety prediction model of the refrigeration device in the above embodiments will be introduced below.
[0090] First, the construction process of the function prediction model of the refrigeration device is introduced.
[0091] In an optional embodiment, for any refrigeration device in the refrigeration devices, when the historical operation data of the refrigeration device has more working conditions, a black box model can be built by using machine learning and deep learning to obtain the function prediction model; when the historical operation data has fewer working conditions, on the one hand, a white box model can be built in combination with the factory equipment performance data of the refrigeration device to obtain a smooth curve as the function prediction model, and data of all working conditions can be obtained; on the other hand, a black box model can also be built by combining the historical operation data with the factory equipment performance data.
[0092] Therefore, the function prediction model of any refrigeration equipment can be constructed in any of the following three ways:
[0093] The first way is that when the working conditions of the historical operation data of the refrigeration equipment are more, the first training data set is obtained based on the historical operation data of the refrigeration equipment, each first training data includes the historical parameter value of the function parameter of the refrigeration equipment, and the initial function prediction model of the refrigeration equipment is trained based on the first training data set to obtain the function prediction model of the refrigeration equipment.
[0094] Among them, according to the historical operation parameters of the refrigeration equipment, the historical parameter values of the function parameters of the refrigeration equipment under various working conditions (including the historical parameter values of the input function parameters and the historical parameter values of the output function parameters) can be obtained, and the historical parameter values of the function parameters under each working condition are taken as a first training data. Then, when training the initial function prediction model based on the first training data set, the historical parameter values of the input function parameters in each first training data are taken as the input, the predicted parameter values of the output function parameters are obtained, and the historical parameter values of the output function parameters are compared with the corresponding predicted parameter values to adjust the parameters of the initial function prediction model.
[0095] The second way is that when the working conditions of the historical operation data of the refrigeration equipment are less, the first training data set is obtained based on the factory equipment performance data and the historical operation data of the refrigeration equipment, each first training data includes the historical parameter value of the function parameter of the refrigeration equipment, and the initial function prediction model of the refrigeration equipment is trained based on the first training data set to obtain the function prediction model of the refrigeration equipment.
[0096] Among them, when the working conditions of the historical operation data of the refrigeration equipment are less, but the running time of the refrigeration equipment is longer, more data under various working conditions can be obtained through the factory equipment performance data of the refrigeration equipment, so as to obtain the first training data set.
[0097] The third way is that when the working conditions of the historical operation data of the refrigeration equipment are less, the reference parameter values of the function parameters of the refrigeration equipment are obtained based on the factory equipment performance data of the refrigeration equipment, and the function prediction function of the refrigeration equipment is constructed based on the reference parameter values of the function parameters, and the function prediction function is taken as the function prediction model.
[0098] When the historical operation data of the refrigeration equipment has few working conditions and the operation time of the refrigeration equipment is short, the baseline parameter value of the function parameter of the refrigeration equipment (including the baseline parameter value of the input function parameter and the baseline parameter value of the output function parameter) can be directly obtained based on the factory equipment performance data of the refrigeration equipment, and the function prediction function (i.e., a white box model) is constructed based on the baseline parameter value of the function parameter, similar to the function of f(x1, x2) = 1.5x1 + 2.6x2, as the function prediction model.
[0099] In the embodiments of the present application, during the early stage of operation of the refrigeration equipment, a white box model can be built by using the factory equipment performance data to obtain relatively continuous working conditions. After the refrigeration equipment has been operated for a period of time, a black box model can be built by using the historical operation data (or the historical operation data and the factory equipment performance data) to accurately predict the output function parameter of the refrigeration equipment and accurately recommend the combination of parameter values of each controllable parameter.
[0100] Next, the construction process of the energy consumption prediction model of the refrigeration equipment is introduced.
[0101] In an optional implementation, similar to the construction process of the function prediction model described above, for any refrigeration equipment in each refrigeration equipment, when the historical operation data of the refrigeration equipment has many working conditions, a black box model can be built to obtain an energy consumption prediction model; when the historical operation data has few working conditions, on the one hand, a white box model can be built in combination with the factory equipment performance data of the refrigeration equipment as an energy consumption prediction model; on the other hand, a black box model can be built in combination of the historical operation data and the factory equipment performance data.
[0102] Therefore, the energy consumption prediction model of any refrigeration equipment in each refrigeration equipment can be obtained in any one of the following three ways:
[0103] The first way is to obtain a second training data set based on the historical operation data of the refrigeration equipment, each second training data including the historical energy consumption of the refrigeration equipment and the corresponding historical parameter value of the function parameter, and training the initial energy consumption prediction model of the refrigeration equipment based on the second training data set to obtain the function prediction model of the refrigeration equipment.
[0104] The historical parameter values of the functional parameters and the historical energy consumption of the refrigeration equipment under various working conditions can be obtained according to the historical operating parameters of the refrigeration equipment, and the historical parameter values of the functional parameters and the historical energy consumption under each working condition are taken as a second training data. In addition, the second training data can include external environment parameters such as IT load, outdoor temperature, and wet-bulb temperature in addition to the historical energy consumption and the historical parameter values of the corresponding functional parameters. That is, the input of the functional prediction model of the refrigeration equipment includes controllable parameters and uncontrollable parameters. When the power real-time prediction is performed, the uncontrollable parameters are obtained through real-time external environment parameters and the prediction values of the series connection model composed of the functional prediction models.
[0105] Then, when the initial functional prediction model is trained based on the second training data set, the historical parameter values of the functional parameters in each second training data are taken as the input (which can also include external environment parameters), the predicted energy consumption is obtained, and the predicted energy consumption and the historical energy consumption are compared to adjust the parameters of the initial energy consumption prediction model.
[0106] The second way is to obtain a second training data set based on the historical operating data of the refrigeration equipment, each second training data including the historical energy consumption of the refrigeration equipment and the historical parameter values of the corresponding functional parameters, and train the initial energy consumption prediction model of the refrigeration equipment based on the second training data set to obtain the functional prediction model of the refrigeration equipment.
[0107] The third way is to obtain the reference energy consumption of the refrigeration equipment and the reference parameter values of the corresponding functional parameters based on the factory equipment performance data of the refrigeration equipment, and construct an energy consumption prediction function of the refrigeration equipment based on the reference energy consumption and the reference parameter values of the corresponding functional parameters, and take the energy consumption prediction function as the energy consumption prediction model.
[0108] The three construction methods of the above-mentioned energy consumption prediction model are similar to the three construction methods of the aforementioned functional prediction model, which will not be described here.
[0109] The energy consumption prediction model obtained through the above-mentioned embodiments can accurately predict the energy consumption of the corresponding refrigeration equipment, and further accurately predict the total energy consumption to accurately recommend the parameter value combination of each controllable parameter.
[0110] The construction process of the business security prediction model will be introduced below.
[0111] In an optional embodiment, the business security prediction model can be obtained by any one of the following two ways:
[0112] Based on historical operation data of each refrigeration equipment, a third training data set is obtained, each third training data including historical safety measurement point values and historical parameter values of the functional parameters of each refrigeration equipment, and the initial business safety prediction model is trained based on the third training data set to obtain the business safety prediction model.
[0113] In the above embodiment, the historical parameter values of the functional parameters of each refrigeration equipment under various working conditions and the historical safety measurement point values can be obtained based on the historical operation data of each refrigeration equipment, and the historical safety measurement point values and the historical parameter values of the functional parameters of each refrigeration equipment under each working condition are taken as a third training data to obtain the third training data set. Then, when the initial business safety prediction model is trained based on the third training data set, the historical parameter values of the functional parameters in each third training data are taken as inputs to obtain safety measurement point prediction values, and the safety measurement point prediction values are compared with the historical safety measurement point values to adjust the parameters of the initial business safety prediction model.
[0114] The business safety prediction model obtained in the above embodiment can accurately predict the safety measurement point values, and then select the parameter value combination that meets the safety condition from all parameter value combinations of the controllable parameters to recommend the final parameter value combination.
[0115] In an optional embodiment, in order to ensure the accuracy of the functional prediction model and the energy consumption prediction model of each refrigeration equipment, the functional prediction model and the energy consumption prediction model of each refrigeration equipment can also be monitored, and the following steps B1-B2 can be performed:
[0116] B1, for the functional prediction model of any refrigeration equipment, monitoring the first prediction index of the functional prediction model, and if the first prediction index does not meet the first index condition, an alarm information is sent.
[0117] Specifically, the prediction value of the functional prediction model can be compared with the corresponding actual value to obtain the corresponding first prediction index, and the calculation method of the first prediction index can be set according to needs, which is not limited. When the first prediction index does not reach the first preset index value (which can be set according to needs), it is considered that the first index condition is not met, and an alarm information can be sent to improve the functional prediction model in time and ensure the prediction accuracy.
[0118] B2, for the energy consumption prediction model of any refrigeration equipment, monitoring the second prediction index of the energy consumption prediction model, and if the second prediction index does not meet the second index condition, an alarm information is sent.
[0119] Specifically, the predicted values of the energy consumption prediction model can be compared with the corresponding actual values to obtain a second prediction index. The calculation method of this second prediction index can be set as needed and is not limited thereto. When the second prediction index fails to reach the second preset index value (which can be set as needed), it is considered that the second index condition is not met, and an alarm message can be issued so that the energy consumption prediction model can be improved in a timely manner to ensure the accuracy of the prediction.
[0120] Similarly, to ensure the accuracy of the business security prediction model, the third prediction indicator of the business security prediction model can also be monitored. If the third prediction indicator does not meet the conditions of the third indicator, an alarm message will be issued.
[0121] Specifically, the predicted values of the business security prediction model can be compared with the corresponding actual values to obtain a third prediction indicator. The calculation method of this third prediction indicator can be set as needed and is not limited thereto. When the third prediction indicator fails to reach the third preset indicator value (which can be set as needed), it is considered that the third indicator condition is not met, and an alarm message can be issued so that the business security prediction model can be improved in a timely manner to ensure the accuracy of the prediction.
[0122] The following is combined Figure 2 The specific implementation process of the energy-saving method for data centers according to embodiments of this application is described.
[0123] like Figure 2 As shown, the implementation process of the data center energy-saving method in this application embodiment mainly includes:
[0124] Step 1: Using historical operating data and factory performance data of each refrigeration unit, build a series model consisting of functional prediction models of each refrigeration unit according to the data center HVAC process, and check the accuracy of the functional prediction model to ensure the accuracy of the model.
[0125] Specifically, by preprocessing the historical operating data and factory performance data of each refrigeration unit, we can obtain the operating data of each refrigeration unit under various operating conditions, including functional parameters and energy consumption. Functional parameters include controllable parameters (i.e.,...) Figure 2 Adjustable features and uncontrollable parameters (i.e., Figure 2 (The unadjustable features in it).
[0126] For any refrigeration equipment, if there is a large amount of historical operating data, a black-box model (i.e., a functional prediction model) can be built using the historical operating data (which can also be combined with factory performance data) and machine learning and deep learning. This includes models such as instantaneous flow rate of chilled water, instantaneous flow rate of cooling water, and return water temperature of chiller cooling water. If there is a small amount of historical operating data, a white-box model can be built by combining the equipment's factory data, which can produce smooth curves and obtain all operating data.
[0127] After obtaining the functional prediction models of each refrigeration unit, a series model is constructed according to the data center HVAC process, namely... Figure 2 The equipment unit process series model in the middle.
[0128] Step 2: Based on the above series model, build an energy consumption prediction model for each refrigeration device to obtain the total energy consumption model, and check the accuracy of the energy consumption prediction model to ensure the accuracy of the model.
[0129] Among them, the energy consumption prediction model for each refrigeration unit is Figure 2 The energy consumption model for each individual refrigeration unit is defined in the model. The total energy consumption model is the sum of the energy consumption prediction models for each refrigeration unit. The input to the energy consumption prediction model for each refrigeration unit includes the functional parameters of that unit, and may also include external environmental parameters (such as IT load, outdoor temperature, and wet-bulb temperature). When performing real-time power prediction, the energy consumption prediction model makes predictions based on the real-time functional parameters of the refrigeration unit and the real-time external environmental parameters.
[0130] Step 3: The business security prediction model mainly considers the security of safety measurement points in HVAC processes to ensure the safe operation of IT services.
[0131] Among them, the business security prediction model is Figure 2 The operational (assistance) prediction model is used. For example, the safety monitoring point can be the pressure differential at the most unfavorable end of the chilled water supply pipe. Additionally, the inlet water temperature of the cold storage tank and the lower tower temperature can also be used as safety monitoring points, and predictions can be made using corresponding operational safety prediction models.
[0132] Step 4: Parameter value combination optimization. By traversing the parameter value limits of all controllable parameters of each refrigeration device, the parameter value combination corresponding to the minimum total energy consumption under the premise of business security is selected for precise recommendation to achieve the effect of energy saving.
[0133] As shown in Table 1 below, for safety considerations, the functional parameters in each refrigeration device have a parameter value limit range on the heating and ventilation process. By traversing the limit value range of the controllable parameters in each refrigeration device, the change of the uncontrollable parameters is affected, and the parameter value combination of each controllable parameter corresponding to the minimum total energy consumption is selected, which is not lower than the lower limit and not higher than the upper limit of the safety measurement point value in the limit value range of the controllable parameters and the uncontrollable parameters, so as to achieve the effect of energy saving.
[0134] Table 1
[0135]
[0136]
[0137] The energy saving method of the data center of the embodiment of the present application has at least the following advantages:
[0138] 1. The embodiment of the present application can accurately recommend device control parameters to achieve precise and efficient energy saving of the data center under the premise that data center hardware energy saving and manual experience optimization cannot meet the requirement of energy consumption reduction. The overall energy saving idea needs to be protected.
[0139] 2. The embodiment of the present application builds a single device series connection model and an energy consumption model according to the data center process. When a new data center energy saving project is started, the model has a certain reusability, which can shorten the model development cost of the premise.
[0140] 3. The embodiment of the present application does not completely abandon the traditional machine learning energy saving scheme in the early stage, combines the factory equipment performance data, builds a white box model, obtains a relatively continuous working condition, and performs preliminary device control parameter recommendation. After running for a period of time, combined with the historical data of the device, an AI model (black box model) can be built for accurate recommendation.
[0141] 4. The embodiment of the present application contains a model accuracy real-time monitoring function. When the model evaluation index does not meet the standard, the model is alarmed and prompted.
[0142] Based on the same inventive concept, the embodiment of the present application also provides an energy saving device for a data center. Since the principle of solving the problem of the device is similar to the above method, the implementation of the device can be referred to the embodiments of the method, and the repeated parts will not be described here.
[0143] As shown in Table 1, the embodiment of the present application provides an energy saving device for a data center, which comprises: Figure 3 An energy consumption prediction module 31 is configured to traverse the parameter value limit range of each controllable parameter in each refrigeration device of the data center, wherein each time the parameter value limit range of each controllable parameter is traversed, the following operations are performed:
[0144]
[0145] Based on the series model composed of the function prediction model of each refrigeration device, in combination with the parameter value of each controllable parameter in this iteration, the parameter value of the function parameter of each refrigeration device is obtained;
[0146] If the parameter value of the function parameter of each refrigeration device meets the safety condition, the predicted energy consumption of any refrigeration device is obtained through the energy consumption prediction model of any refrigeration device based on the parameter value of the function parameter of any refrigeration device;
[0147] Based on the predicted energy consumption of each refrigeration device, the predicted total energy consumption is obtained;
[0148] The parameter recommendation module 32 is configured to select the predicted total energy consumption meeting the preset condition based on the predicted total energy consumption obtained through multiple iterations, and combine the parameter value of each controllable parameter corresponding to the selected predicted total energy consumption as the recommended parameter value combination.
[0149] In the embodiments of the present application, when the parameter value of each controllable parameter is recommended, the parameter value combinations of each controllable parameter are iterated, and each parameter value combination meeting the safety condition is selected from all the iterated parameter value combinations. Then, the energy consumption of each refrigeration device is predicted under each parameter value combination, and the predicted total energy consumption is obtained. Finally, the parameter value combination meeting the preset condition in the predicted total energy consumption is selected from the selected parameter value combinations as the final recommended parameter value combination. In this way, the parameter value of the controllable parameter of each refrigeration device in the data center can be accurately recommended to achieve energy saving and consumption reduction of the data center.
[0150] In an optional embodiment, as shown in Figure 4 The device further comprises a safety determination module 33 configured to:
[0151] Obtain the safety measurement point prediction value based on the parameter value of the function parameter of each refrigeration device through the business safety prediction model;
[0152] If the safety measurement point prediction value is within the safety value range, it is determined that the parameter value of the function parameter of each refrigeration device meets the safety condition.
[0153] In an optional embodiment, as shown in Figure 4 The device further comprises:
[0154] The first alarm module 34 is configured to monitor the first prediction index of the function prediction model of any refrigeration device, and send an alarm information if the first prediction index does not meet the first index condition;
[0155] The second alarm module 35 is configured to monitor the second prediction index of the energy consumption prediction model of any refrigeration device, and send an alarm information if the second prediction index does not meet the second index condition.
[0156] In an optional implementation, as shown in Figure 4 The apparatus further comprises a function model construction module 36, configured to obtain a function prediction model of any of the refrigeration devices by:
[0157] obtaining a first training data set based on historical operation data of the refrigeration device, or factory device performance data and the historical operation data of the refrigeration device, each first training data comprising historical parameter values of function parameters of the refrigeration device, and training an initial function prediction model of the refrigeration device based on the first training data set to obtain the function prediction model of the refrigeration device; or
[0158] obtaining reference parameter values of the function parameters of the refrigeration device based on factory device performance data of the refrigeration device, and constructing a function prediction function of the refrigeration device based on the reference parameter values of the function parameters, taking the function prediction function as the function prediction model.
[0159] In an optional implementation, as shown in Figure 4 The apparatus further comprises an energy consumption model construction module 37, configured to obtain an energy consumption prediction model of any of the refrigeration devices by:
[0160] obtaining a second training data set based on historical operation data of the refrigeration device, or factory device performance data and the historical operation data of the refrigeration device, each second training data comprising historical energy consumption of the refrigeration device and historical parameter values of corresponding function parameters, and training an initial energy consumption prediction model of the refrigeration device based on the second training data set to obtain the function prediction model of the refrigeration device; or
[0161] obtaining reference energy consumption of the refrigeration device and reference parameter values of corresponding function parameters based on factory device performance data of the refrigeration device, and constructing an energy consumption prediction function of the refrigeration device based on the reference energy consumption and the reference parameter values of the corresponding function parameters, taking the energy consumption prediction function as the energy consumption prediction model.
[0162] In an optional implementation, as shown in Figure 4 The apparatus further comprises a security model construction module 38, configured to obtain a business security prediction model by:
[0163] obtaining a third training data set based on historical operation data of the refrigeration devices, each third training data comprising historical security measurement point values and historical parameter values of function parameters of the refrigeration devices, and training an initial business security prediction model based on the third training data set to obtain the business security prediction model.
[0164] In an optional implementation, as shown in Figure 4 The apparatus further comprises a third alarm module 39, configured to:
[0165] The third prediction index of the business security prediction model is monitored, and if the third prediction index does not satisfy a third index condition, an alarm information is sent.
[0166] Based on the same inventive concept, the embodiments of the present application also provide an electronic device. Since the principle of the electronic device to solve the problem is similar to the method, the implementation of the electronic device can refer to the embodiments of the method, and the repeated parts will not be described here.
[0167] Referring to Figure 5 As shown in the figure, the electronic device can include a processor 52 and a memory 51. The memory 51 provides the processor 52 with program instructions and data stored in the memory 51. In the embodiments of the present disclosure, the memory 51 can be used to store the programs of the multimedia resource processing in the embodiments of the present disclosure.
[0168] The processor 52 processes the program instructions stored in the memory 51 by calling, and the processor 52 is used to execute the method in any method embodiment described above, for example Figure 1 The embodiment shown provides a power saving method of a data center.
[0169] The specific connection medium between the memory 51 and the processor 52 in the embodiments of the present disclosure is not limited. In the embodiments of the present disclosure Figure 5 The memory 51 and the processor 52 are connected through a bus 53, and the bus 53 is represented by a thick line in Figure 5 The connection mode between other components is only schematically illustrated and is not limited. The bus 53 can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 5 In the figure, only one thick line is used to represent, but it does not mean that there is only one bus or only one type of bus.
[0170] The memory can include a read-only memory (ROM) and a random access memory (RAM), and can also include a non-volatile memory (NVM), such as at least one disk memory. Alternatively, the memory can also be at least one storage device located away from the aforementioned processor.
[0171] The processor described above can be a general-purpose processor, including a central processing unit, a network processor (NP), etc.; can also be a digital signal processing (DSP) processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc.
[0172] The embodiment of the present disclosure further provides a computer storage medium, in which a computer program is stored, and a processor of an electronic device reads the computer program from the computer storage medium. The processor executes the computer program, so that the electronic device executes the energy-saving method of the data center in any method embodiment.
[0173] In the specific implementation process, the computer storage medium can include a universal serial bus flash drive (USB), a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various storage media that can store program codes.
[0174] Based on the same inventive concept as the above method embodiments, the embodiment of the present disclosure provides a computer program product, which includes computer instructions stored in a computer readable storage medium. A processor of an electronic device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the steps of the energy-saving method of the data center.
[0175] The computer program product can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0176] Those skilled in the art should understand that the embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, the present disclosure can adopt a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure can adopt the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.
[0177] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0178] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0179] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0180] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.
Claims
1. An energy-saving method for a data center, characterized in that, include: For each controllable parameter in each cooling device of the data center, the parameter values within the parameter value limit range of each controllable parameter are traversed. During each traversal, the following operations are performed: Based on the series model composed of the functional prediction models of each refrigeration device, and combined with the parameter values of each controllable parameter traversed in this iteration, the parameter values of the functional parameters of each refrigeration device are obtained. The series model is obtained by building the functional prediction model of each refrigeration device using the historical operating data and factory performance data of each refrigeration device, according to the data center HVAC process. The functional parameters include input functional parameters and output functional parameters. Each functional prediction model is used to predict the corresponding output functional parameters based on the input functional parameters of the corresponding refrigeration device. If the parameter values of each functional parameter of each refrigeration device meet the safety conditions, then the predicted energy consumption of any refrigeration device can be obtained based on the parameter values of the functional parameters of any refrigeration device through the energy consumption prediction model of any refrigeration device. Based on the predicted energy consumption of each of the aforementioned refrigeration devices, the predicted total energy consumption is obtained; Based on the predicted total energy consumption obtained through multiple traversals, a predicted total energy consumption that meets preset conditions is selected, and the parameter values of each controllable parameter corresponding to the selected predicted total energy consumption are used as a recommended parameter value combination.
2. The method according to claim 1, characterized in that, The method further includes: Based on the parameter values of the functional parameters of each refrigeration device, the predicted values of the safety measurement points are obtained through the business security prediction model. If the predicted value of the safety measurement point is within the safe range, then the parameter values of the functional parameters of each refrigeration device are determined to meet the safety conditions.
3. The method according to claim 1, characterized in that, The method further includes: For any refrigeration equipment's functional prediction model, monitor the first prediction index of the functional prediction model. If the first prediction index does not meet the first index condition, issue an alarm message. For any energy consumption prediction model of a refrigeration device, the second prediction index of the energy consumption prediction model is monitored. If the second prediction index does not meet the second index condition, an alarm message is issued.
4. The method according to any one of claims 1 to 3, characterized in that, The functional prediction model for any of the refrigeration devices is obtained through the following method: Based on the historical operating data of the refrigeration equipment, or the factory performance data and historical operating data of the refrigeration equipment, a first training dataset is obtained. Each first training dataset includes historical parameter values of the functional parameters of the refrigeration equipment. An initial functional prediction model for the refrigeration equipment is trained based on the first training dataset to obtain the functional prediction model of the refrigeration equipment; or Based on the factory performance data of the refrigeration equipment, the baseline parameter values of the functional parameters of the refrigeration equipment are obtained, and based on the baseline parameter values of the functional parameters, a functional prediction function of the refrigeration equipment is constructed, and the functional prediction function is used as the functional prediction model.
5. The method according to any one of claims 1 to 3, characterized in that, The energy consumption prediction model for any of the refrigeration devices is obtained through the following method: Based on the historical operating data of the refrigeration equipment, or the factory performance data and historical operating data of the refrigeration equipment, a second training dataset is obtained. Each second training dataset includes the historical energy consumption of the refrigeration equipment and the historical parameter values of the corresponding functional parameters. The initial energy consumption prediction model of the refrigeration equipment is then trained based on the second training dataset to obtain the energy consumption prediction model of the refrigeration equipment; or Based on the factory performance data of the refrigeration equipment, the baseline energy consumption and the baseline parameter values of the corresponding functional parameters of the refrigeration equipment are obtained. Based on the baseline energy consumption and the baseline parameter values of the corresponding functional parameters, the energy consumption prediction function of the refrigeration equipment is constructed, and the energy consumption prediction function is used as the energy consumption prediction model.
6. The method according to claim 2, characterized in that, The business security prediction model is obtained through the following methods: Based on the historical operating data of each refrigeration device, a third training dataset is obtained. Each third training dataset includes historical safety measurement point values and historical parameter values of the functional parameters of each refrigeration device. The initial business security prediction model is trained based on the third training dataset to obtain the business security prediction model.
7. The method according to claim 2 or 6, characterized in that, The method further includes: Monitor the third prediction indicator of the business security prediction model. If the third prediction indicator does not meet the third indicator conditions, issue an alarm message.
8. An energy-saving device for a data center, characterized in that, include: The energy consumption prediction module is used to iterate through the parameter values within the limit range of each controllable parameter in each cooling device of the data center. Each iteration performs the following operations: Based on the series model composed of the functional prediction models of each refrigeration device, and combined with the parameter values of each controllable parameter traversed in this iteration, the parameter values of the functional parameters of each refrigeration device are obtained. The series model is obtained by building the functional prediction model of each refrigeration device using the historical operating data and factory performance data of each refrigeration device, according to the data center HVAC process. The functional parameters include input functional parameters and output functional parameters. Each functional prediction model is used to predict the corresponding output functional parameters based on the input functional parameters of the corresponding refrigeration device. If the parameter values of each functional parameter of each refrigeration device meet the safety conditions, then the predicted energy consumption of any refrigeration device can be obtained based on the parameter values of the functional parameters of any refrigeration device through the energy consumption prediction model of any refrigeration device. Based on the predicted energy consumption of each of the aforementioned refrigeration devices, the predicted total energy consumption is obtained; The parameter recommendation module is used to select the predicted total energy consumption that meets the preset conditions based on the predicted total energy consumption obtained through multiple traversals, and use the parameter values of each controllable parameter corresponding to the selected predicted total energy consumption as the recommended parameter value combination.
9. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, It includes a computer program that, when run on an electronic device, causes the electronic device to perform the steps of the method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Air conditioning energy consumption model training method and air conditioning system control method
CN112577161A