Data center data processing method and system for energy-saving control
By using predictive neural networks in the data center to combine energy consumption curves and ambient temperature change curves, dynamically adjusting the energy consumption of computing equipment, solving the problem of poor energy-saving control effect in the existing technology, and achieving high-efficiency management of the data center and Cost reduction.
Patent Information
- Application Number
- CN202410903367.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-08
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-07-08
AI Technical Summary
The energy-saving control methods of existing data centers are mainly focused on the improvement of hardware cooling equipment, and the equipment energy consumption curve and ambient temperature change curve are not fully considered, resulting in high cooling design cost and the inability to achieve dynamic and flexible energy-saving control.
Using a prediction neural network based on the energy consumption curve analysis model and environmental temperature analysis model, the energy consumption and temperature prediction neural network of the target data center is trained to dynamically adjust the energy consumption according to the current environmental information. To achieve energy-saving effect.
It realizes dynamic and flexible energy-saving control of the data center, reducing energy consumption and operational costs, and reducing carbon emissions.
Smart Images

Figure CN119066353A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a data center data processing method and system for energy-saving control. Background Art
[0002] With the continued rapid growth of the investment scale of the data center market and the continued expansion of the scale of data centers, the energy consumption of data centers is also rising. In addition, the development of artificial intelligence technology has increased the demand for parallel computing, and its energy consumption is even more serious. However, the resource utilization levels of various data centers are uneven. Data centers generally have problems such as extensive heat dissipation design, insufficient heat dissipation capacity, uneven heat dissipation effect, and high overall energy waste. The high energy consumption of data centers not only brings about increased power consumption and a sharp rise in operating costs, but also generates a heavy carbon emission burden. Therefore, how to use technical means and measures to reduce the energy consumption of data centers has become a top priority. Most of the existing energy-saving control methods for data centers still remain in the direction of improving hardware heat dissipation equipment, and do not fully consider the equipment energy consumption curve and environmental temperature change curve in the data center. Therefore, the hardware cost of its heat dissipation design is higher, and it is impossible to achieve dynamic and flexible energy-saving control effects at the algorithm level. It can be seen that the existing technology has defects that need to be solved urgently. Summary of the invention
[0003] The technical problem to be solved by the present invention is to provide a data center data processing method and system for energy-saving control, which can realize the training of a predictive neural network for energy consumption and temperature corresponding to a target data center, so as to achieve a dynamic and flexible energy-saving control effect.
[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a data center data processing method for energy-saving control, the method comprising: Acquire device sensor information of a plurality of computing devices in a target data center and environmental sensor information of a plurality of device areas; Based on the energy consumption curve analysis model, according to the device sensing information, analyzing the device energy consumption curve corresponding to each of the computing devices; Based on the environmental temperature analysis model, analyzing the regional temperature change curve corresponding to each of the equipment areas according to the environmental sensing information; Based on the equipment energy consumption curve and the regional temperature change curve, a prediction neural network for energy consumption and temperature corresponding to the target data center is trained; the prediction neural network is used to predict the optimal energy consumption based on the current environmental sensor information to control the computing equipment of the target data center.
[0005] As an optional implementation, in the first aspect of the present invention, the device sensing information includes device temperature, device humidity, device performance, device image and device displacement.
[0006] As an optional implementation, in the first aspect of the present invention, the environmental sensing information includes environmental temperature, environmental humidity, environmental image and environmental infrared detection information.
[0007] As an optional implementation, in the first aspect of the present invention, the energy consumption curve analysis model is based on the device sensing information, and the device energy consumption curve corresponding to each computing device is analyzed, including: For each of the computing devices, based on the time point sensing information corresponding to each time point in the device sensing information corresponding to the computing device, predicting the energy consumption information corresponding to the time point sensing information based on the neural network; According to the time points and energy consumption information corresponding to the sensing information at all the time points, an energy consumption curve of the device corresponding to the computing device with the variables of time and energy consumption being established.
[0008] As an optional implementation, in the first aspect of the present invention, the predicting the energy consumption information corresponding to the sensing information at the time point based on the neural network includes: Inputting the data of each sensor type in the sensor information at the time point into the trained multi-source information energy consumption prediction neural network model to obtain energy consumption prediction information corresponding to the data of each sensor type; the multi-source information energy consumption prediction neural network model is trained by a training data set including training data of multiple sensor types and corresponding energy consumption annotations; the sensor type is device temperature, device humidity, device performance, device image or device displacement; The weighted average value of the energy consumption prediction information corresponding to the data of all sensor types in the sensor information at the time point is calculated to obtain the energy consumption information corresponding to the sensor information at the time point.
[0009] As an optional implementation, in the first aspect of the present invention, the analyzing the regional temperature change curve corresponding to each of the equipment regions based on the environmental temperature analysis model according to the environmental sensing information includes: For each of the device areas, based on the time point environment sensor information corresponding to each time point in the environment sensor information corresponding to the device area, predicting the temperature information corresponding to the time point environment sensor information based on a neural network; According to the time points and temperature information corresponding to the environmental sensing information at all the time points, a regional temperature change curve corresponding to the device area with the variables of time and temperature is established.
[0010] As an optional implementation, in the first aspect of the present invention, the predicting the temperature information corresponding to the environmental sensor information at the time point based on the neural network includes: Input each non-temperature sensor information in the environmental sensor information at the time point into the trained multi-source information temperature prediction neural network model to obtain temperature prediction information corresponding to each non-temperature sensor information; the multi-source information energy consumption prediction neural network model is trained by a training data set including training non-temperature sensor information of multiple sensor types and corresponding energy consumption annotations; the non-temperature sensor information is environmental humidity, environmental image or environmental infrared detection information; Calculating a weighted average of the temperature prediction information corresponding to all non-temperature sensor information in the environmental sensor information at the time point to obtain the corrected temperature information corresponding to the environmental sensor information at the time point; An average value between the corrected temperature information and the ambient temperature information in the ambient sensor information at the time point is calculated to obtain temperature information corresponding to the ambient sensor information at the time point.
[0011] As an optional implementation, in the first aspect of the present invention, the step of training a prediction neural network of energy consumption and temperature corresponding to the target data center according to the equipment energy consumption curve and the regional temperature change curve includes: Calculating the overall curve positive correlation between the device energy consumption curve of any one of the computing devices and the curve temperature change curve of any one of the device regions; Determine the computing devices and the device regions whose overall curve positive correlation is greater than a first threshold as a device region corresponding group; For any of the equipment area corresponding groups, calculating the positive correlation between the equipment energy consumption curve corresponding to the equipment area corresponding group and the curve in any time period of the regional temperature change curve; Determine a curve portion in which the curve positive correlation degree is greater than a second threshold value in the device energy consumption curve and the regional temperature change curve as a training curve portion; Determine the energy consumption information and the temperature information at any same time point in any of the training curve parts and the environmental sensor information corresponding to the temperature information to obtain a plurality of training data pairs; Each of the training data pairs is input into a predictive neural network for training until convergence, thereby obtaining a predictive neural network capable of predicting energy consumption information based on environmental sensor information.
[0012] A second aspect of an embodiment of the present invention discloses a data center data processing system for energy-saving control, the system comprising: An acquisition module, used to acquire device sensor information of multiple computing devices in a target data center and environmental sensor information of multiple device areas; A first analysis module, configured to analyze a device energy consumption curve corresponding to each of the computing devices based on an energy consumption curve analysis model and according to the device sensing information; A second analysis module is used to analyze a regional temperature change curve corresponding to each of the equipment regions based on an environmental temperature analysis model and according to the environmental sensing information; A training module is used to train a prediction neural network for energy consumption and temperature corresponding to the target data center based on the equipment energy consumption curve and the regional temperature change curve; the prediction neural network is used to predict the optimal energy consumption based on the current environmental sensor information to control the computing equipment of the target data center.
[0013] As an optional implementation, in the second aspect of the present invention, the device sensing information includes device temperature, device humidity, device performance, device image and device displacement.
[0014] As an optional implementation, in the second aspect of the present invention, the environmental sensing information includes environmental temperature, environmental humidity, environmental image and environmental infrared detection information.
[0015] As an optional implementation, in the second aspect of the present invention, the first analysis module analyzes the device energy consumption curve corresponding to each computing device based on the energy consumption curve analysis model and according to the device sensing information, including: For each of the computing devices, based on the time point sensing information corresponding to each time point in the device sensing information corresponding to the computing device, predicting the energy consumption information corresponding to the time point sensing information based on the neural network; According to the time points and energy consumption information corresponding to the sensing information at all the time points, an energy consumption curve of the device corresponding to the computing device with the variables of time and energy consumption being established.
[0016] As an optional implementation, in the second aspect of the present invention, the specific manner in which the first analysis module predicts the energy consumption information corresponding to the sensing information at the time point based on the neural network includes: Inputting the data of each sensor type in the sensor information at the time point into the trained multi-source information energy consumption prediction neural network model to obtain energy consumption prediction information corresponding to the data of each sensor type; the multi-source information energy consumption prediction neural network model is trained by a training data set including training data of multiple sensor types and corresponding energy consumption annotations; the sensor type is device temperature, device humidity, device performance, device image or device displacement; The weighted average value of the energy consumption prediction information corresponding to the data of all sensor types in the sensor information at the time point is calculated to obtain the energy consumption information corresponding to the sensor information at the time point.
[0017] As an optional implementation, in the second aspect of the present invention, the second analysis module analyzes the regional temperature change curve corresponding to each of the device regions based on the environmental temperature analysis model and according to the environmental sensing information, including: For each of the device areas, based on the time point environment sensor information corresponding to each time point in the environment sensor information corresponding to the device area, predicting the temperature information corresponding to the time point environment sensor information based on a neural network; According to the time points and temperature information corresponding to the environmental sensing information at all the time points, a regional temperature change curve corresponding to the device area with the variables of time and temperature is established.
[0018] As an optional implementation, in the second aspect of the present invention, the specific manner in which the second analysis module predicts the temperature information corresponding to the environmental sensor information at the time point based on the neural network includes: Input each non-temperature sensor information in the environmental sensor information at the time point into the trained multi-source information temperature prediction neural network model to obtain temperature prediction information corresponding to each non-temperature sensor information; the multi-source information energy consumption prediction neural network model is trained by a training data set including training non-temperature sensor information of multiple sensor types and corresponding energy consumption annotations; the non-temperature sensor information is environmental humidity, environmental image or environmental infrared detection information; Calculating a weighted average of the temperature prediction information corresponding to all non-temperature sensor information in the environmental sensor information at the time point to obtain the corrected temperature information corresponding to the environmental sensor information at the time point; An average value between the corrected temperature information and the ambient temperature information in the ambient sensor information at the time point is calculated to obtain temperature information corresponding to the ambient sensor information at the time point.
[0019] As an optional implementation, in the second aspect of the present invention, the training module trains a specific method of predicting the energy consumption and temperature of the target data center according to the equipment energy consumption curve and the regional temperature change curve, including: Calculating the overall curve positive correlation between the device energy consumption curve of any one of the computing devices and the curve temperature change curve of any one of the device regions; Determine the computing devices and the device regions whose overall curve positive correlation is greater than a first threshold as a device region corresponding group; For any of the equipment area corresponding groups, calculating the positive correlation between the equipment energy consumption curve corresponding to the equipment area corresponding group and the curve in any time period of the regional temperature change curve; Determine a curve portion in which the curve positive correlation degree is greater than a second threshold value in the device energy consumption curve and the regional temperature change curve as a training curve portion; Determine the energy consumption information and the temperature information at any same time point in any of the training curve parts and the environmental sensor information corresponding to the temperature information to obtain a plurality of training data pairs; Each of the training data pairs is input into a predictive neural network for training until convergence, thereby obtaining a predictive neural network capable of predicting energy consumption information based on environmental sensor information.
[0020] A third aspect of the present invention discloses another data center data processing system for energy-saving control, the system comprising: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute part or all of the steps in the data center data processing method for energy-saving control disclosed in the first aspect of the present invention.
[0021] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the data center data processing method for energy-saving control disclosed in the first aspect of the present invention.
[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: The present invention can analyze the device energy consumption curve corresponding to each computing device based on the energy consumption curve analysis model and analyze the regional temperature change curve corresponding to each device area based on the ambient temperature analysis model, so as to train a predictive neural network corresponding to the target data center, and subsequently predict the optimal energy consumption according to the current environmental sensor information to control the computing equipment of the target data center, thereby realizing the training of the predictive neural network of energy consumption and temperature corresponding to the target data center, so as to achieve a dynamic and flexible energy-saving control effect. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.
[0024] Figure 1 It is a flow chart of a data center data processing method for energy-saving control disclosed in an embodiment of the present invention.
[0025] Figure 2 It is a structural schematic diagram of a data center data processing system for energy-saving control disclosed in an embodiment of the present invention.
[0026] Figure 3 It is a structural schematic diagram of another data center data processing system for energy-saving control disclosed in an embodiment of the present invention. DETAILED DESCRIPTION
[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of 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.
[0028] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.
[0029] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0030] The present invention discloses a data center data processing method and system for energy-saving control, which can analyze the device energy consumption curve corresponding to each computing device based on the energy consumption curve analysis model and the regional temperature change curve corresponding to each device area based on the environmental temperature analysis model, so as to train the prediction neural network corresponding to the target data center, and then predict the optimal energy consumption according to the current environmental sensor information to control the computing equipment of the target data center, so as to realize the training of the prediction neural network of energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effect. The following are detailed descriptions.
[0031] Embodiment 1 See also Figure 1 , Figure 1 1 is a flow chart of a data center data processing method for energy-saving control disclosed in an embodiment of the present invention. Figure 1 The data center data processing method for energy-saving control described herein can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the data center data processing method for energy-saving control may include the following operations: 101. Obtain device sensor information of multiple computing devices in a target data center and environmental sensor information of multiple device areas.
[0032] 102. Based on the energy consumption curve analysis model, analyze the device energy consumption curve corresponding to each computing device according to the device sensor information. 103. Based on the environmental temperature analysis model, analyze the regional temperature change curve corresponding to each equipment area according to the environmental sensor information. 104. According to the equipment energy consumption curve and the regional temperature change curve, a prediction neural network for energy consumption and temperature corresponding to the target data center is trained.
[0033] Optionally, a predictive neural network is used to predict optimal energy consumption based on current environmental sensing information to control computing equipment in a target data center.
[0034] It can be seen that the above-mentioned embodiments of the invention can analyze the device energy consumption curve corresponding to each computing device based on the energy consumption curve analysis model and analyze the regional temperature change curve corresponding to each device area based on the ambient temperature analysis model, so as to train the prediction neural network corresponding to the target data center, and then predict the optimal energy consumption according to the current environmental sensor information to control the computing equipment of the target data center, thereby realizing the training of the prediction neural network of energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0035] As an optional embodiment, in the above steps, the device sensing information includes device temperature, device humidity, device performance, device image and device displacement.
[0036] It can be seen that through the above optional embodiments, the data content of the device sensor information is limited, which can fully characterize the working condition of the device, and is subsequently used to accurately analyze the device energy consumption curve corresponding to each computing device, and assist in the training of the predictive neural network for the energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0037] As an optional embodiment, in the above steps, the environmental sensing information includes environmental temperature, environmental humidity, environmental image and environmental infrared detection information.
[0038] It can be seen that through the above optional embodiments, the data content of the environmental sensing information is limited, which can fully characterize the situation in which the environment is affected by the operation of the equipment. It is subsequently used to accurately analyze the regional temperature change curve corresponding to each equipment area, and assist in the training of the predictive neural network for the energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0039] As an optional embodiment, in the above steps, based on the energy consumption curve analysis model, analyzing the device energy consumption curve corresponding to each computing device according to the device sensing information includes: For each computing device, based on the time point sensing information corresponding to each time point in the device sensing information corresponding to the computing device, predicting the energy consumption information corresponding to the time point sensing information based on the neural network; According to the time points and energy consumption information corresponding to the sensing information at all time points, an energy consumption curve of the device corresponding to the computing device with the variables of time and energy consumption is established.
[0040] It can be seen that through the above-mentioned optional embodiments, the energy consumption information corresponding to the sensing information at a time point can be predicted through a neural network to establish an equipment energy consumption curve corresponding to the computing device with the variables of time and energy consumption, which is then used to accurately train the prediction neural network corresponding to the target data center, and assist in the training of the prediction neural network for energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0041] As an optional embodiment, in the above steps, predicting the energy consumption information corresponding to the sensing information at the time point based on the neural network includes: Input the data of each sensor type in the time point sensor information into the trained multi-source information energy consumption prediction neural network model to obtain energy consumption prediction information corresponding to the data of each sensor type; optionally, the multi-source information energy consumption prediction neural network model is trained by a training data set including training data of multiple sensor types and corresponding energy consumption annotations; the sensor type is device temperature, device humidity, device performance, device image or device displacement; The weighted sum average of the energy consumption prediction information corresponding to the data of all sensor types in the time point sensor information is calculated to obtain the energy consumption information corresponding to the time point sensor information.
[0042] It can be seen that through the above-mentioned optional embodiments, the energy consumption prediction information corresponding to each sensor type of data can be predicted through the trained multi-source information energy consumption prediction neural network model, wherein the neural network trained by multi-source information can predict prediction results that have reference value to each other for multi-source sensor information, and then determine the energy consumption information corresponding to the sensor information at the time point based on the weighted sum calculation, which is subsequently used to establish an equipment energy consumption curve corresponding to the computing device with the variables of time and energy consumption, to assist in the training of the energy consumption and temperature prediction neural network corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0043] As an optional embodiment, in the above steps, based on the environmental temperature analysis model and according to the environmental sensor information, analyzing the regional temperature change curve corresponding to each device area includes: For each device area, based on the time point environment sensor information corresponding to each time point in the environment sensor information corresponding to the device area, predict the temperature information corresponding to the time point environment sensor information based on the neural network; According to the time points and temperature information corresponding to the environmental sensor information at all time points, a regional temperature change curve with time and temperature as the variables corresponding to the device area is established.
[0044] It can be seen that through the above optional embodiments, the temperature information corresponding to the environmental sensor information at a time point can be predicted through a neural network to establish a regional temperature change curve corresponding to the equipment area with the variables of time and temperature, which is then used to accurately train the prediction neural network corresponding to the target data center, and assist in the training of the prediction neural network for energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0045] As an optional embodiment, in the above steps, predicting the temperature information corresponding to the environmental sensor information at the time point based on the neural network includes: Input each non-temperature sensor information in the environmental sensor information at a time point into the trained multi-source information temperature prediction neural network model to obtain the temperature prediction information corresponding to each non-temperature sensor information; optionally, the multi-source information energy consumption prediction neural network model is trained by a training data set including training non-temperature sensor information of multiple sensor types and corresponding energy consumption annotations; the non-temperature sensor information is environmental humidity, environmental image or environmental infrared detection information; Calculate the weighted average of the temperature prediction information corresponding to all non-temperature sensor information in the environmental sensor information at the time point to obtain the corrected temperature information corresponding to the environmental sensor information at the time point; The average value between the corrected temperature information and the ambient temperature information in the ambient sensor information at the time point is calculated to obtain the temperature information corresponding to the ambient sensor information at the time point.
[0046] It can be seen that through the above optional embodiments, the temperature prediction information corresponding to each non-temperature sensor information can be predicted through the trained multi-source information temperature prediction neural network model, wherein the neural network trained by multi-source information can predict the prediction results that have reference value to each other for the multi-source sensor information, and then determine the corrected temperature information based on the weighted sum calculation to correct the ambient temperature information in the ambient sensor information at the time point, so as to calculate more accurate temperature data, which is subsequently used to establish a regional temperature change curve corresponding to the equipment area with the variables of time and temperature, to assist in the training of the energy consumption and temperature prediction neural network corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0047] As an optional embodiment, in the above steps, training a prediction neural network of energy consumption and temperature corresponding to the target data center according to the equipment energy consumption curve and the regional temperature change curve includes: Calculate the overall curve positive correlation between the equipment energy consumption curve of any computing equipment and the curve temperature change curve of any equipment area; Determine computing devices and device regions whose overall curve positive correlation is greater than a first threshold as a device region corresponding group; For any equipment area corresponding group, calculate the positive correlation between the equipment energy consumption curve corresponding to the equipment area corresponding group and the curve in any time period of the regional temperature change curve; Determine a curve portion in which the curve positive correlation degree of the equipment energy consumption curve and the regional temperature change curve is greater than a second threshold as a training curve portion; Determine the energy consumption information and temperature information at any same time point in any training curve part and the environmental sensor information corresponding to the temperature information to obtain a plurality of training data pairs; Each training data pair is input into a predictive neural network for training until convergence, thereby obtaining a predictive neural network that can predict energy consumption information based on environmental sensor information.
[0048] It can be seen that through the above-mentioned optional embodiments, a more relevant training data set can be determined by calculating and screening the positive correlation between the curves, so as to train a predictive neural network that can predict energy consumption information based on environmental sensor information, thereby realizing the training of the predictive neural network for energy consumption and temperature corresponding to the target data center, so as to achieve a dynamic and flexible energy-saving control effect.
[0049] Embodiment 2 See also Figure 2 , Figure 2 1 is a schematic diagram of a data center data processing system for energy-saving control disclosed in an embodiment of the present invention. Figure 2The data center data processing system for energy-saving control described above can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the data center data processing system for energy-saving control may include: The acquisition module 201 is used to acquire device sensor information of multiple computing devices in a target data center and environmental sensor information of multiple device areas.
[0050] The first analysis module 202 is used to analyze the device energy consumption curve corresponding to each computing device based on the energy consumption curve analysis model and according to the device sensing information. The second analysis module 203 is used to analyze the regional temperature change curve corresponding to each device area based on the environmental temperature analysis model and according to the environmental sensing information. The training module 204 is used to train a prediction neural network of energy consumption and temperature corresponding to the target data center according to the equipment energy consumption curve and the regional temperature change curve.
[0051] Optionally, a predictive neural network is used to predict optimal energy consumption based on current environmental sensing information to control computing equipment in a target data center.
[0052] It can be seen that the above-mentioned embodiments of the invention can analyze the device energy consumption curve corresponding to each computing device based on the energy consumption curve analysis model and analyze the regional temperature change curve corresponding to each device area based on the ambient temperature analysis model, so as to train the prediction neural network corresponding to the target data center, and then predict the optimal energy consumption according to the current environmental sensor information to control the computing equipment of the target data center, thereby realizing the training of the prediction neural network of energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0053] As an optional embodiment, the device sensing information includes device temperature, device humidity, device performance, device image and device displacement.
[0054] It can be seen that through the above optional embodiments, the data content of the device sensor information is limited, which can fully characterize the working condition of the device, and is subsequently used to accurately analyze the device energy consumption curve corresponding to each computing device, and assist in the training of the predictive neural network for the energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0055] As an optional embodiment, the environmental sensing information includes environmental temperature, environmental humidity, environmental image and environmental infrared detection information.
[0056] It can be seen that through the above optional embodiments, the data content of the environmental sensing information is limited, which can fully characterize the situation in which the environment is affected by the operation of the equipment. It is subsequently used to accurately analyze the regional temperature change curve corresponding to each equipment area, and assist in the training of the predictive neural network for the energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0057] As an optional embodiment, the first analysis module analyzes the device energy consumption curve corresponding to each computing device based on the energy consumption curve analysis model and according to the device sensing information, including: For each computing device, based on the time point sensing information corresponding to each time point in the device sensing information corresponding to the computing device, predicting the energy consumption information corresponding to the time point sensing information based on the neural network; According to the time points and energy consumption information corresponding to the sensing information at all time points, an energy consumption curve of the device corresponding to the computing device with the variables of time and energy consumption is established.
[0058] It can be seen that through the above-mentioned optional embodiments, the energy consumption information corresponding to the sensing information at a time point can be predicted through a neural network to establish an equipment energy consumption curve corresponding to the computing device with the variables of time and energy consumption, which is then used to accurately train the prediction neural network corresponding to the target data center, and assist in the training of the prediction neural network for energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0059] As an optional embodiment, the specific manner in which the first analysis module predicts the energy consumption information corresponding to the time point sensing information based on the neural network includes: Input the data of each sensor type in the time point sensor information into the trained multi-source information energy consumption prediction neural network model to obtain energy consumption prediction information corresponding to the data of each sensor type; optionally, the multi-source information energy consumption prediction neural network model is trained by a training data set including training data of multiple sensor types and corresponding energy consumption annotations; the sensor type is device temperature, device humidity, device performance, device image or device displacement; The weighted sum average of the energy consumption prediction information corresponding to the data of all sensor types in the time point sensor information is calculated to obtain the energy consumption information corresponding to the time point sensor information.
[0060] It can be seen that through the above-mentioned optional embodiments, the energy consumption prediction information corresponding to each sensor type of data can be predicted through the trained multi-source information energy consumption prediction neural network model, wherein the neural network trained by multi-source information can predict prediction results that have reference value to each other for multi-source sensor information, and then determine the energy consumption information corresponding to the sensor information at the time point based on the weighted sum calculation, which is subsequently used to establish an equipment energy consumption curve corresponding to the computing device with the variables of time and energy consumption, to assist in the training of the energy consumption and temperature prediction neural network corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0061] As an optional embodiment, the second analysis module analyzes the regional temperature change curve corresponding to each device area based on the environmental temperature analysis model and according to the environmental sensor information, including: For each device area, based on the time point environment sensor information corresponding to each time point in the environment sensor information corresponding to the device area, predict the temperature information corresponding to the time point environment sensor information based on the neural network; According to the time points and temperature information corresponding to the environmental sensor information at all time points, a regional temperature change curve with time and temperature as the variables corresponding to the device area is established.
[0062] It can be seen that through the above optional embodiments, the temperature information corresponding to the environmental sensor information at a time point can be predicted through a neural network to establish a regional temperature change curve corresponding to the equipment area with the variables of time and temperature, which is then used to accurately train the prediction neural network corresponding to the target data center, and assist in the training of the prediction neural network for energy consumption and temperature corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0063] As an optional embodiment, the specific manner in which the second analysis module predicts the temperature information corresponding to the environmental sensor information at a time point based on the neural network includes: Input each non-temperature sensor information in the environmental sensor information at a time point into the trained multi-source information temperature prediction neural network model to obtain the temperature prediction information corresponding to each non-temperature sensor information; optionally, the multi-source information energy consumption prediction neural network model is trained by a training data set including training non-temperature sensor information of multiple sensor types and corresponding energy consumption annotations; the non-temperature sensor information is environmental humidity, environmental image or environmental infrared detection information; Calculate the weighted average of the temperature prediction information corresponding to all non-temperature sensor information in the environmental sensor information at the time point to obtain the corrected temperature information corresponding to the environmental sensor information at the time point; The average value between the corrected temperature information and the ambient temperature information in the ambient sensor information at the time point is calculated to obtain the temperature information corresponding to the ambient sensor information at the time point.
[0064] It can be seen that through the above optional embodiments, the temperature prediction information corresponding to each non-temperature sensor information can be predicted through the trained multi-source information temperature prediction neural network model, wherein the neural network trained by multi-source information can predict the prediction results that have reference value to each other for the multi-source sensor information, and then determine the corrected temperature information based on the weighted sum calculation to correct the ambient temperature information in the ambient sensor information at the time point, so as to calculate more accurate temperature data, which is subsequently used to establish a regional temperature change curve corresponding to the equipment area with the variables of time and temperature, to assist in the training of the energy consumption and temperature prediction neural network corresponding to the target data center, so as to achieve dynamic and flexible energy-saving control effects.
[0065] As an optional embodiment, the training module trains a specific method of predicting the energy consumption and temperature of the target data center according to the equipment energy consumption curve and the regional temperature change curve, including: Calculate the overall curve positive correlation between the equipment energy consumption curve of any computing equipment and the curve temperature change curve of any equipment area; Determine computing devices and device regions whose overall curve positive correlation is greater than a first threshold as a device region corresponding group; For any equipment area corresponding group, calculate the positive correlation between the equipment energy consumption curve corresponding to the equipment area corresponding group and the curve in any time period of the regional temperature change curve; Determine a curve portion in which the curve positive correlation degree of the equipment energy consumption curve and the regional temperature change curve is greater than a second threshold as a training curve portion; Determine the energy consumption information and temperature information at any same time point in any training curve part and the environmental sensor information corresponding to the temperature information to obtain a plurality of training data pairs; Each training data pair is input into a predictive neural network for training until convergence, thereby obtaining a predictive neural network that can predict energy consumption information based on environmental sensor information.
[0066] It can be seen that through the above-mentioned optional embodiments, a more relevant training data set can be determined by calculating and screening the positive correlation between the curves, so as to train a predictive neural network that can predict energy consumption information based on environmental sensor information, thereby realizing the training of the predictive neural network for energy consumption and temperature corresponding to the target data center, so as to achieve a dynamic and flexible energy-saving control effect.
[0067] Embodiment 3 See also Figure 3 , Figure 3 This is another data center data processing system for energy-saving control disclosed in an embodiment of the present invention. Figure 3The data center data processing system for energy-saving control described is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the data center data processing system for energy-saving control may include: A memory 301 storing executable program codes; a processor 302 coupled to the memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the data center data processing method for energy-saving control described in the first embodiment.
[0068] Embodiment 4 An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the data center data processing method for energy-saving control described in the first embodiment.
[0069] Embodiment 5 An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the data center data processing method for energy-saving control described in the first embodiment.
[0070] The above describes specific embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0071] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
[0072] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.
[0073] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in 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 contain computer-usable program code.
[0074] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. 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.
[0075] 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.
[0076] 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.
[0077] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0078] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0079] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0080] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0081] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.
[0082] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0083] Finally, it should be noted that the data center data processing method and system for energy-saving control disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A data center data processing method for energy-saving control, It is characterized in that The method comprises: Acquire device sensor information of a plurality of computing devices in a target data center and environmental sensor information of a plurality of device areas; Based on the energy consumption curve analysis model, according to the device sensing information, analyzing the device energy consumption curve corresponding to each of the computing devices; Based on the environmental temperature analysis model, analyzing the regional temperature change curve corresponding to each of the equipment areas according to the environmental sensing information; Based on the equipment energy consumption curve and the regional temperature change curve, a prediction neural network for energy consumption and temperature corresponding to the target data center is trained; the prediction neural network is used to predict the optimal energy consumption based on the current environmental sensor information to control the computing equipment of the target data center.
2. The data center data processing method for energy-saving control according to claim 1, It is characterized in that The device sensing information includes device temperature, device humidity, device performance, device image and device displacement.
3. The data center data processing method for energy-saving control according to claim 1, It is characterized in that The environmental sensing information includes environmental temperature, environmental humidity, environmental image and environmental infrared detection information.
4. The data center data processing method for energy-saving control according to claim 1, It is characterized in that The energy consumption curve analysis model is based on the device sensing information, and the device energy consumption curve corresponding to each computing device is analyzed, including: For each of the computing devices, based on the time point sensing information corresponding to each time point in the device sensing information corresponding to the computing device, predicting the energy consumption information corresponding to the time point sensing information based on the neural network; According to the time points and energy consumption information corresponding to the sensing information at all the time points, an energy consumption curve of the device corresponding to the computing device with the variables of time and energy consumption being established.
5. The data center data processing method for energy-saving control according to claim 4, It is characterized in that The predicting of the energy consumption information corresponding to the sensing information at the time point based on the neural network includes: Inputting the data of each sensor type in the sensor information at the time point into the trained multi-source information energy consumption prediction neural network model to obtain energy consumption prediction information corresponding to the data of each sensor type; the multi-source information energy consumption prediction neural network model is trained by a training data set including training data of multiple sensor types and corresponding energy consumption annotations; the sensor type is device temperature, device humidity, device performance, device image or device displacement; The weighted average value of the energy consumption prediction information corresponding to the data of all sensor types in the sensor information at the time point is calculated to obtain the energy consumption information corresponding to the sensor information at the time point.
6. The data center data processing method for energy-saving control according to claim 4, It is characterized in that The method of analyzing the regional temperature change curve corresponding to each of the equipment regions based on the environmental temperature analysis model and according to the environmental sensing information includes: For each of the device areas, based on the time point environment sensor information corresponding to each time point in the environment sensor information corresponding to the device area, predicting the temperature information corresponding to the time point environment sensor information based on a neural network; According to the time points and temperature information corresponding to the environmental sensing information at all the time points, a regional temperature change curve corresponding to the device area with the variables of time and temperature is established.
7. The data center data processing method for energy-saving control according to claim 6, It is characterized in that The predicting the temperature information corresponding to the environmental sensor information at the time point based on the neural network includes: Input each non-temperature sensor information in the environmental sensor information at the time point into the trained multi-source information temperature prediction neural network model to obtain temperature prediction information corresponding to each non-temperature sensor information; the multi-source information energy consumption prediction neural network model is trained by a training data set including training non-temperature sensor information of multiple sensor types and corresponding energy consumption annotations; the non-temperature sensor information is environmental humidity, environmental image or environmental infrared detection information; Calculating a weighted average of the temperature prediction information corresponding to all non-temperature sensor information in the environmental sensor information at the time point to obtain the corrected temperature information corresponding to the environmental sensor information at the time point; An average value between the corrected temperature information and the ambient temperature information in the ambient sensor information at the time point is calculated to obtain temperature information corresponding to the ambient sensor information at the time point.
8. The data center data processing method for energy-saving control according to claim 1, It is characterized in that The step of training a prediction neural network of energy consumption and temperature corresponding to the target data center according to the equipment energy consumption curve and the regional temperature change curve includes: Calculating the overall curve positive correlation between the device energy consumption curve of any one of the computing devices and the curve temperature change curve of any one of the device regions; Determine the computing devices and the device regions whose overall curve positive correlation is greater than a first threshold as a device region corresponding group; For any of the equipment area corresponding groups, calculating the positive correlation between the equipment energy consumption curve corresponding to the equipment area corresponding group and the curve in any time period of the regional temperature change curve; Determine a curve portion in which the curve positive correlation degree is greater than a second threshold value in the device energy consumption curve and the regional temperature change curve as a training curve portion; Determine the energy consumption information and the temperature information at any same time point in any of the training curve parts and the environmental sensor information corresponding to the temperature information to obtain a plurality of training data pairs; Each of the training data pairs is input into a predictive neural network for training until convergence, thereby obtaining a predictive neural network capable of predicting energy consumption information based on environmental sensor information.
9. A data center data processing system for energy-saving control, It is characterized in that The system comprises: An acquisition module, used to acquire device sensor information of multiple computing devices in a target data center and environmental sensor information of multiple device areas; A first analysis module, configured to analyze a device energy consumption curve corresponding to each of the computing devices based on an energy consumption curve analysis model and according to the device sensing information; A second analysis module is used to analyze a regional temperature change curve corresponding to each of the equipment regions based on an environmental temperature analysis model and according to the environmental sensing information; A training module is used to train a prediction neural network for energy consumption and temperature corresponding to the target data center based on the equipment energy consumption curve and the regional temperature change curve; the prediction neural network is used to predict the optimal energy consumption based on the current environmental sensor information to control the computing equipment of the target data center.
10. A data center data processing system for energy-saving control, It is characterized in that The system comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the data center data processing method for energy-saving control as described in any one of claims 1-8.
Citation Information
Patent Citations
Method and device for dynamic load global optimization and energy-saving control of central air conditioner
CN113028610A
Comprehensive prediction method and system for air conditioner energy consumption
CN113188243A
Indoor environment monitoring system suitable for near-zero energy consumption building
CN116734918A
Building energy consumption monitoring method and system based on sensor network
CN117077854A
Method For Predicting The Areas Of Information Needed To Be Collected
US20240160895A1
