Data center energy saving air conditioning system and regulating method

Abstract

The application provides a data center energy-saving air conditioning system and a regulating method, which has the characteristics of significantly reducing energy consumption and improving operation efficiency by integrating historical data collection, preprocessing, prediction model construction and intelligent control, wherein the system accurately predicts air conditioning energy consumption by using a multivariate linear regression model, optimizes the air conditioning set temperature by combining a gradient descent method, realizes precise adaptive regulation, updates the model online by using a recursive least square method, ensures long-term adaptation to environmental changes, reduces invalid operation of the air conditioning system, fully utilizes external natural conditions to reduce energy consumption, and provides an innovative and practical solution for data center energy saving.

Description

Technical Field

[0001] This invention relates to the field of air conditioning technology, and in particular to an energy-saving air conditioning system and adjustment method for data centers. Background Technology

[0002] With the widespread adoption of cloud computing, big data, and artificial intelligence technologies, data centers, as the core infrastructure of the digital economy, are experiencing continuous growth in scale and number, leading to increasingly prominent energy consumption issues. Air conditioning systems are a key component of data center energy consumption, accounting for approximately 30% to 50% of total energy consumption, primarily used to maintain stable operating temperatures for servers and other equipment. Traditional data center air conditioning systems often employ control strategies based on fixed setpoints, such as adjusting compressor frequency or fan speed through feedback from indoor temperature sensors. However, this method lacks adaptability to changes in the external environment, resulting in the air conditioning system operating at high energy consumption year-round. Existing technologies have introduced some improvements, such as adjusting operating parameters through programmable logic controllers (PLCs), but these remain limited to real-time reactive regulation, failing to fully utilize historical data and predictive models for forward-looking optimization.

[0003] Existing data center air conditioning systems suffer from numerous problems: First, the systems cannot adaptively adjust to ambient temperature. When the outside temperature is low, the air conditioning system continues to operate in a fixed mode, wasting natural cooling opportunities and increasing unnecessary energy consumption. Second, the control strategies are simplistic and lack intelligent predictive capabilities. Relying solely on real-time data leads to lag in adjustments, making it unable to cope with dynamic environmental changes, resulting in energy consumption fluctuations and low efficiency. Furthermore, existing systems often neglect the value of historical data and fail to utilize machine learning algorithms to uncover operational patterns, leading to unreasonable air conditioning parameter settings and long-term energy efficiency degradation. These problems result in persistently high energy consumption and increased operating costs for data center air conditioning systems, while also exacerbating carbon emissions, hindering the realization of green data centers and sustainable development.

[0004] Therefore, there is an urgent need in this field for an energy-saving air conditioning system and regulation method for data centers that can intelligently adapt to changes in the external environment, build a predictive model based on historical data and achieve dynamic control, so as to effectively reduce energy consumption and improve energy utilization efficiency. Summary of the Invention

[0005] The purpose of this invention is to provide an energy-saving air conditioning system and adjustment method for data centers to solve the problems existing in the prior art.

[0006] To achieve the above objectives, the present invention provides the following solution:

[0007] This invention provides a method for regulating an energy-saving air conditioning system in a data center, comprising:

[0008] S1. Collect historical environmental data and air conditioning operation data, and preprocess them. The historical environmental data includes ambient temperature and ambient humidity, and the air conditioning operation data includes air conditioning set temperature and actual air conditioning energy consumption.

[0009] S2. Construct a prediction model using preprocessed historical data. The prediction model is used to predict air conditioning energy consumption and future ambient temperature.

[0010] S3. Based on a predictive model, the air conditioner's operating parameters are dynamically adjusted through intelligent control algorithms to reduce energy consumption;

[0011] S4. Regularly update the prediction model to adapt to environmental changes.

[0012] Preferably, in step S1, the historical environmental data further includes the temperature and time characteristics inside the data center, and the air conditioning operating parameters further include the compressor frequency and fan speed.

[0013] Preferably, in step S1, the preprocessing includes data cleaning, outlier handling, and normalization; data cleaning uses the 3σ principle to remove outliers, and normalization uses the Min-Max normalization method to map the data to the range [0,1], with the following formula: ;in, The original data, and These are the minimum and maximum values ​​of the variable, respectively. This is the normalized value.

[0014] Preferably, step S1 further includes feature selection, using the Pearson correlation coefficient to select features that are highly correlated with energy consumption, and retaining features with an absolute value of correlation coefficient greater than 0.5.

[0015] Preferably, in step S2, the prediction model adopts a multiple linear regression model, and the model formula is:

[0016] ;in, To predict energy consumption, For the outside temperature, As for the external humidity, Set the temperature for the air conditioner. For the regression intercept term, This represents the regression coefficient of external temperature on energy consumption. The regression coefficient of external humidity on energy consumption. To set the regression coefficient of temperature on energy consumption, This is the error term.

[0017] Preferably, in step S3, the intelligent control algorithm includes optimizing the air conditioner set temperature using gradient descent to minimize predicted energy consumption, and the update formula is:

[0018] ;in, For the updated set temperature, Set the current temperature. For learning rate, To predict the partial derivative of energy consumption with respect to a set temperature; the optimization process satisfies the internal temperature constraints of the data center, the constraint formula is: ;in, The lower limit of the internal temperature of the data center is set at 22°C. The upper limit of the internal temperature of the data center is set at 24°C. This represents the actual internal temperature of the data center.

[0019] Preferably, in step S4, the periodic update of the prediction model includes updating the model parameters online using the recursive least squares method, with the update formula being:

[0020] ; ; ;in, For a moment The parameter vector, For a moment The Kalman gain matrix, For a moment The energy consumption of observation, For a moment The transpose of the eigenvectors. Forgetting factor, For a moment The error covariance matrix.

[0021] This invention also provides a data center energy-saving air conditioning system, based on a data center energy-saving air conditioning system adjustment method, comprising:

[0022] The data acquisition module is used to collect historical environmental data and air conditioning operation data;

[0023] The data preprocessing module is used to preprocess the collected data;

[0024] The prediction model module is used to build prediction models using preprocessed historical data.

[0025] The intelligent control module is used to dynamically adjust the air conditioner's operating parameters based on a predictive model.

[0026] The model update module is used to periodically update the prediction model.

[0027] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of a data center energy-saving air conditioning system adjustment method.

[0028] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of a data center energy-saving air conditioning system adjustment method.

[0029] The present invention achieves the following beneficial technical effects compared to the prior art:

[0030] This invention provides an energy-saving air conditioning system and regulation method for data centers. By integrating historical data acquisition, preprocessing, predictive model construction, and intelligent control, it features significant energy reduction and improved operating efficiency. The system uses a multiple linear regression model to accurately predict air conditioning energy consumption, combines gradient descent method to optimize air conditioning set temperature, and achieves precise adaptive regulation. At the same time, it updates the model online through recursive least squares method to ensure long-term adaptation to environmental changes. This not only reduces the ineffective operation of the air conditioning system but also makes full use of external natural conditions to reduce energy consumption, providing an innovative and practical solution for energy saving in data centers. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 Flowchart of the data center energy-saving air conditioning system adjustment method provided by the present invention;

[0033] Figure 2 This is a schematic diagram of the control relationship of an energy-saving air conditioning system for data centers provided by the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] The purpose of this invention is to provide an energy-saving air conditioning system and adjustment method for data centers, which achieves adaptive energy-saving adjustment of the air conditioning system through intelligent data acquisition, processing and predictive control.

[0036] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] Example 1:

[0038] like Figure 1 As shown, the data center energy-saving air conditioning system adjustment method provided by this invention mainly includes four core steps: historical data acquisition and preprocessing, predictive model construction, intelligent control execution, and model adaptive updating. These steps are interconnected to form a closed-loop control system, which can dynamically optimize air conditioning operating parameters according to changes in the external environment, significantly reducing energy consumption.

[0039] During the data acquisition and preprocessing phase, the system collects historical environmental data and air conditioning operation data in real time through a sensor network deployed inside and outside the data center. Historical environmental data includes ambient temperature, ambient humidity, internal data center temperature, and time characteristics (such as season and hour). Air conditioning operation data includes air conditioning set temperature, actual energy consumption, compressor frequency, and fan speed. The collected raw data undergoes preprocessing to improve data quality. Preprocessing includes data cleaning, outlier handling, and normalization. Data cleaning uses the 3σ principle to remove outliers, i.e., calculating the mean μ and standard deviation σ of each variable, removing data points that satisfy |x-μ|>3σ, and filling missing values ​​using linear interpolation. Normalization uses the Min-Max normalization method, mapping the data to the [0,1] range, with the formula:

[0040] ;in, The original data, and These are the minimum and maximum values ​​of the variable, respectively. These are the normalized values. After normalization, the system performs feature selection, using the Pearson correlation coefficient to select features highly correlated with energy consumption, retaining features with an absolute correlation coefficient greater than 0.5. The formula for calculating the Pearson correlation coefficient is:

[0041] ;

[0042] in, For the first Feature values ​​of each sample For the first Energy consumption value of each sample and The mean, This step, setting the sample size, eliminates the influence of dimensions and improves the accuracy and stability of subsequent models.

[0043] During the predictive model building phase, the system utilizes preprocessed historical data to construct a multiple linear regression model for predicting air conditioning energy consumption or future ambient temperature. The model formula is as follows:

[0044] ;in, To predict energy consumption, For the outside temperature, As for the external humidity, Set the temperature for the air conditioner. For the regression intercept term, This represents the regression coefficient of external temperature on energy consumption. The regression coefficient of external humidity on energy consumption. To set the regression coefficient of temperature on energy consumption, This represents the error term. The regression coefficients are estimated using the least squares method, with the objective of minimizing the sum of squared residuals, as shown in the formula:

[0045] ;

[0046] in, For the coefficient estimation vector, The characteristic matrix, It is the transpose of the characteristic matrix. This represents the energy consumption vector. After model training, the root mean square error (RMSE) is used for evaluation, as shown in the formula:

[0047] ;

[0048] in, To predict energy consumption. If the RMSE exceeds a threshold (e.g., 0.1), the features or model are readjusted to ensure prediction accuracy.

[0049] During the intelligent control execution phase, the system dynamically optimizes the air conditioning set temperature based on the predictive model output using gradient descent to minimize predicted energy consumption. The gradient descent update formula is:

[0050] ;in, For the updated set temperature, Set the current temperature. Let the learning rate be 0.1. The partial derivative of predicted energy consumption with respect to a set temperature; in a linear regression model, it equals... The optimization process must meet the internal temperature constraints of the data center.

[0051] ;in, The lower limit of the internal temperature of the data center is set at 22°C. The upper limit of the internal temperature of the data center is set at 24°C. The actual internal temperature of the data center can be compared with the set temperature using historical data. The intelligent control module sends the optimized parameters to the actuator module to adjust the air conditioning compressor frequency, fan speed, etc., to achieve precise control.

[0052] During the model adaptive update phase, the system periodically updates the model parameters online using the Recursive Least Squares (RLS) method to adapt to environmental changes and data drift. The RLS update formula is:

[0053] ; ; ;in, For a moment The parameter vector, For a moment The Kalman gain matrix, For a moment The energy consumption of observation, For a moment The transpose of the eigenvectors. Forgetting factor, For a moment The error covariance matrix. The update frequency can be set to every 24 hours or triggered when the prediction error increases significantly, ensuring the long-term adaptability of the model.

[0054] Example 2:

[0055] like Figure 2 As shown, the data center energy-saving air conditioning system includes a data acquisition module, a data preprocessing module, a prediction model module, an intelligent control module, and a model update module. The data acquisition module collects environmental and operational data through a sensor network and outputs it to the data preprocessing module. After cleaning and normalizing, the data preprocessing module sends the processed data to the prediction model module. The prediction model module builds and trains a prediction model and outputs the prediction results to the intelligent control module. The intelligent control module optimizes control parameters based on the prediction results and adjusts the air conditioning equipment through the actuator module. The model update module periodically updates the prediction model, forming a closed-loop control. All modules of the system are connected via data flow to achieve efficient collaboration.

[0056] Through the above-described embodiments, this invention fully utilizes historical data and machine learning algorithms to achieve intelligent energy-saving regulation of data center air conditioning systems, effectively reducing energy consumption and improving operational efficiency.

[0057] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] It should be noted that the components mentioned in the above embodiments are all general standard parts or components known to those skilled in the art. Their structures and principles can be learned by those skilled in the art through technical manuals or conventional experimental methods.

[0059] This invention has illustrated its principles and implementation methods using specific examples. The descriptions of these embodiments are merely illustrative of the method and its core ideas; furthermore, those skilled in the art will recognize that modifications may be made to the specific implementation methods and application scope based on the principles of this invention. Therefore, the content of this specification should not be construed as limiting the invention.

Claims

1. A method for regulating an energy-saving air conditioning system in a data center, characterized in that, include: S1. Collect historical environmental data and air conditioning operation data, and preprocess them. The historical environmental data includes ambient temperature and ambient humidity, and the air conditioning operation data includes air conditioning set temperature and actual air conditioning energy consumption. S2. Construct a prediction model using preprocessed historical data. The prediction model is used to predict air conditioning energy consumption and future ambient temperature. S3. Based on a predictive model, the air conditioner's operating parameters are dynamically adjusted through intelligent control algorithms to reduce energy consumption; S4. Regularly update the prediction model to adapt to environmental changes.

2. The data center energy-saving air conditioning system adjustment method as described in claim 1, characterized in that, In step S1, the historical environmental data also includes the temperature and time characteristics inside the data center, and the air conditioning operating parameters also include the compressor frequency and fan speed.

3. The data center energy-saving air conditioning system adjustment method as described in claim 2, characterized in that, In step S1, the preprocessing includes data cleaning, outlier handling, and normalization; Data cleaning uses the 3σ principle to remove outliers, and normalization uses the Min-Max normalization method to map the data to the range [0,1]. The formula is as follows: ;in, The original data, and These are the minimum and maximum values ​​of the variable, respectively. This is the normalized value.

4. The data center energy-saving air conditioning system adjustment method as described in claim 3, characterized in that, Step S1 also includes feature selection, using the Pearson correlation coefficient to select features that are highly correlated with energy consumption, and retaining features with an absolute value of correlation coefficient greater than 0.

5.

5. The data center energy-saving air conditioning system adjustment method as described in claim 1, characterized in that, In step S2, the prediction model adopts a multiple linear regression model, and the model formula is: ;in, To predict energy consumption, For the outside temperature, As for the external humidity, Set the temperature for the air conditioner. For the regression intercept term, This represents the regression coefficient of external temperature on energy consumption. The regression coefficient of external humidity on energy consumption. To set the regression coefficient of temperature on energy consumption, This is the error term.

6. The data center energy-saving air conditioning system adjustment method as described in claim 1, characterized in that, In step S3, the intelligent control algorithm includes using gradient descent to optimize the air conditioner set temperature to minimize predicted energy consumption. The update formula is as follows: ;in, For the updated set temperature, Set the current temperature. For learning rate, To predict the partial derivative of energy consumption with respect to a set temperature; the optimization process satisfies the internal temperature constraints of the data center, the constraint formula is: ;in, The lower limit of the internal temperature of the data center is set at 22°C. The upper limit of the internal temperature of the data center is set at 24°C. This represents the actual internal temperature of the data center.

7. The data center energy-saving air conditioning system adjustment method as described in claim 1, characterized in that, In step S4, the periodic update of the prediction model includes updating the model parameters online using the recursive least squares method, with the update formula as follows: ; ; ;in, For a moment The parameter vector, For a moment The Kalman gain matrix, For a moment The energy consumption of observation, For a moment The transpose of the eigenvectors. Forgetting factor, For a moment The error covariance matrix.

8. A data center energy-saving air conditioning system, based on the data center energy-saving air conditioning system adjustment method according to any one of claims 1-7, characterized in that, include: The data acquisition module is used to collect historical environmental data and air conditioning operation data; The data preprocessing module is used to preprocess the collected data; The prediction model module is used to build prediction models using preprocessed historical data. The intelligent control module is used to dynamically adjust the air conditioner's operating parameters based on a predictive model. The model update module is used to periodically update the prediction model.

9. A computer device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the data center energy-saving air conditioning system adjustment method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the data center energy-saving air conditioning system adjustment method as described in any one of claims 1-7.

Images