Method and device for predicting return water temperature of data center cooling system

Abstract

The invention relates to the technical field of time sequence prediction, in particular to a data center cooling system return water temperature prediction method and device, and the method comprises the steps: collecting the initial operation data of a cooling system in a data center; preprocessing the initial operation data to obtain final operation data meeting certain conditions; and inputting the final operation data into a pre-trained time series prediction model to obtain a prediction result of the return water temperature of the cooling system, the time series prediction model having a Kronecker lamination layer, a kernel matrix layer and a multi-layer perceptron layer. Therefore, the problems that in related technologies, the model parameter quantity is huge, black box is caused by expression of nonlinear characteristics through a complex activation function, the cross-variable relation is difficult to directly express, the effectiveness and interpretability of the model in practical application are limited, and understanding and trust of a user on a model prediction result are affected are solved.

Description

Technical Field

[0001] The present application relates to the technical field of time series prediction, and in particular to a method and device for predicting the return water temperature of a cooling system in a data center. Background Art

[0002] Data centers are infrastructure that ensure the operation of servers and network equipment. The cooling system is an important part of data center operations. If it is not properly controlled, it will lead to an increase in server failure rates and even service interruptions.

[0003] In the related technology, the original database can be constructed by acquiring the thermodynamic parameters of the data center cooling system in real time, and preprocessing can be performed to obtain the power consumption data set and the chip temperature data set, and then the power consumption prediction neural network can be trained to maintain the optimization control results within the critical temperature of the chip, thereby maximizing energy saving in the data center; data-driven model predictive control methods can also be used to apply data-driven control strategies based on behavioral theory to the data center rack air cooling control system, and the control algorithm can be improved according to the characteristics of the rack system, thereby directly realizing temperature control of the rack-mounted data center.

[0004] However, in related technologies, these models usually have a large number of parameters, which not only leads to huge consumption of computing resources and prolonged training time, but also increases the cost of model deployment and operation. In addition, the nonlinear characteristics of the model are mainly expressed through activation functions, making the internal mechanism of the model black-box, difficult to explain and understand intuitively, lacking transparency and credibility in practical applications, and urgently needs to be improved. Summary of the invention

[0005] The present application provides a method and device for predicting the return water temperature of a data center cooling system, in order to solve the problems in the related technology, such as the large number of model parameters, the black box caused by the expression of nonlinear characteristics through complex activation functions, and the difficulty in intuitively expressing cross-variable relationships, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results.

[0006] The first aspect of the present application provides a method for predicting the return water temperature of a data center cooling system, comprising the following steps: collecting initial operating data of a cooling system in a data center; preprocessing the initial operating data to obtain final operating data that meets preset conditions; inputting the final operating data into a pre-trained time series prediction model to obtain a prediction result of the return water temperature of the cooling system, wherein the time series prediction model has a Kronecker product layer, a kernel matrix layer, and a multi-layer perceptron layer.

[0007] Optionally, in one embodiment of the present application, before inputting the operating data into a pre-trained time series prediction model, it also includes: constructing a Kronecker product layer of the time series prediction model using a Kronecker product and a rotation mask, so as to obtain the input matrix of the final operating data using the Kronecker product layer; constructing a kernel matrix layer of the time series prediction model in combination with the input matrix and a kernel function, so as to obtain an output matrix corresponding to the input matrix according to the kernel matrix layer; and constructing a multi-layer perceptron layer of the time series prediction model using the output matrix and a linear multi-layer perceptron, so as to construct the pre-trained time series prediction model.

[0008] Optionally, in one embodiment of the present application, constructing the pre-trained time series prediction model includes: obtaining a training data set and a test data set according to the final operation data; generating the pre-trained time series prediction model based on the training data set, the test data set, MSE (Mean Squared Error) and MAE (Mean Absolute Error).

[0009] Optionally, in one embodiment of the present application, the Kronecker product layer of the time series prediction model is constructed using the Kronecker product and the rotating mask to obtain the input matrix of the final operating data using the Kronecker product layer, including: obtaining an initial input matrix corresponding to the final operating data based on the final operating data; calculating the Kronecker product matrix of the final operating data using the initial input matrix and the Kronecker product; and obtaining the input matrix based on the Kronecker product matrix and the rotating mask.

[0010] Optionally, in one embodiment of the present application, the mathematical expression of the multi-layer perceptron layer may be, but is not limited to:

[0011] y(k)=b+k1×y (1) (k)+…+k n ×y (n) (k)

[0012] Among them, y n (k) is the n-order output result of the model, b is the bias term of the multilayer perceptron, k i is the weight of the corresponding order result.

[0013] The second aspect of the present application provides a device for predicting the return water temperature of a data center cooling system, comprising: an acquisition module for acquiring initial operating data of a cooling system in a data center; a preprocessing module for preprocessing the initial operating data to obtain final operating data that meets preset conditions; a prediction module for inputting the final operating data into a pre-trained time series prediction model to obtain a prediction result of the return water temperature of the cooling system, wherein the time series prediction model has a Kronecker product layer, a kernel matrix layer and a multi-layer perceptron layer.

[0014] Optionally, in one embodiment of the present application, it also includes: a first construction module, which is used to construct the Kronecker product layer of the time series prediction model using the Kronecker product and the rotation mask before inputting the operating data into the pre-trained time series prediction model, so as to obtain the input matrix of the final operating data using the Kronecker product layer; a second construction module, which is used to construct the kernel matrix layer of the time series prediction model in combination with the input matrix and the kernel function, so as to obtain the output matrix corresponding to the input matrix according to the kernel matrix layer; a third construction module, which is used to construct the multi-layer perceptron layer of the time series prediction model using the output matrix and the linear multi-layer perceptron, so as to construct the pre-trained time series prediction model.

[0015] Optionally, in one embodiment of the present application, the third building module includes: an acquisition unit, used to acquire a training data set and a test data set according to the final operating data; and a first generation unit, used to generate the pre-trained time series prediction model based on the training data set, the test data set, MSE and MAE.

[0016] Optionally, in one embodiment of the present application, the first construction module includes: a second generation unit, used to obtain an initial input matrix corresponding to the final operating data based on the final operating data; a calculation unit, used to calculate the Kronecker product matrix of the final operating data using the initial input matrix and the Kronecker product; and a third generation unit, used to obtain the input matrix based on the Kronecker product matrix and the rotation mask.

[0017] Optionally, in one embodiment of the present application, the mathematical expression of the multi-layer perceptron layer may be, but is not limited to:

[0018] y(k)=b+k1×y (1) (k)+…+k n ×y (n) (k)

[0019] Among them, y n (k) is the n-order output result of the model, b is the bias term of the multilayer perceptron, ki is the weight of the corresponding order result.

[0020] A third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for predicting the return water temperature of a data center cooling system as described in the above embodiment.

[0021] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above method for predicting the return water temperature of a data center cooling system.

[0022] The fifth aspect of the present application provides a computer program product, including a computer program, which, when executed, implements the above method for predicting the return water temperature of a data center cooling system.

[0023] The embodiment of the present application can pre-process the collected initial operation data of the cooling system in the data center, and then obtain the final operation data that meets certain conditions, and input the final operation data into the pre-trained time series prediction model to obtain the prediction result of the return water temperature of the cooling system, so as to accurately predict the return water temperature of the cooling system and the average temperature and humidity of the return air, intuitively express the variable relationship, conduct detailed variable impact analysis, and provide effective system optimization strategies, thereby improving the overall performance of the cooling system and ensuring the safety and efficient operation of the server room. In this way, the problems in the related technology such as the large number of model parameters, the black box caused by the expression of nonlinear characteristics through complex activation functions, and the difficulty in intuitively expressing the cross-variable relationship are solved, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results.

[0024] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0026] Figure 1 A flowchart of a method for predicting return water temperature of a data center cooling system provided according to an embodiment of the present application;

[0027] Figure 2 A schematic diagram of the structure of a time series prediction model provided according to an embodiment of the present application;

[0028] Figure 3A block diagram of a data center cooling system prediction provided according to an embodiment of the present application;

[0029] Figure 4 A block diagram of a predicted indirect evaporative cooling system according to one embodiment of the present application;

[0030] Figure 5 A schematic block diagram of an overall algorithm of a time series prediction model provided according to an embodiment of the present application;

[0031] Figure 6 A block diagram of a device for predicting return water temperature of a data center cooling system provided according to an embodiment of the present application;

[0032] Figure 7 It is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0033] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0034] The following describes the prediction method and device of the return water temperature of the data center cooling system of the embodiment of the present application with reference to the accompanying drawings. In view of the huge number of model parameters mentioned in the above background technology, the black box caused by the expression of nonlinear characteristics through complex activation functions, and the difficulty in intuitively expressing the cross-variable relationship, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results, the present application provides a prediction method for the return water temperature of the data center cooling system. In this method, the initial operation data of the cooling system in the collected data center can be preprocessed to obtain the final operation data that meets certain conditions, and the final operation data is input into the pre-trained time series prediction model to obtain the prediction result of the return water temperature of the cooling system, so as to achieve accurate prediction of the return water temperature of the cooling system and the average temperature and humidity of the return air, intuitively express the variable relationship, conduct detailed variable impact analysis, and provide effective system optimization strategies, thereby improving the overall performance of the cooling system and ensuring the safe and efficient operation of the server room. As a result, problems in related technologies such as the large number of model parameters, black boxing caused by expressing nonlinear characteristics through complex activation functions, and difficulty in intuitively expressing cross-variable relationships were solved, which limited the effectiveness and interpretability of the model in practical applications and affected users' understanding and trust in the model's prediction results.

[0035] Specifically, Figure 1The present invention is a flowchart of a method for predicting the return water temperature of a data center cooling system provided according to an embodiment of the present application.

[0036] like Figure 1 As shown, the method for predicting the return water temperature of the data center cooling system includes the following steps:

[0037] In step S101, initial operation data of a cooling system in a data center is collected.

[0038] It can be understood that the initial operating data of the embodiment of the present application may include but is not limited to inlet water temperature, return water temperature, power consumption of IT (Information Technology) equipment, water supply pressure difference, average return air temperature, average return air humidity, etc., which can be specifically set by technicians in this field according to actual conditions, and this application does not impose specific restrictions.

[0039] In addition, it should be noted that the embodiments of the present application may utilize sampling sensors to collect initial operating data of the cooling system of the data center, or may utilize other methods, and the present application does not impose any specific limitations thereto.

[0040] As a possible implementation method, the embodiment of the present application can collect initial operation data of the cooling system.

[0041] For example, the embodiments of the present application can use sampling sensors to collect initial operating data of the cooling center, such as inlet water temperature, return water temperature, IT equipment power consumption, water supply pressure difference, return air average temperature, return air average humidity, etc., and the present application does not impose specific restrictions.

[0042] In step S102, the initial operation data is preprocessed to obtain final operation data that meets preset conditions.

[0043] It should be noted that the preprocessing in the embodiments of the present application may include but is not limited to data cleaning, data standardization, and processing of missing data values, etc., which can be specifically configured by technicians in this field according to actual conditions, and this application does not impose any specific restrictions.

[0044] As a possible implementation method, the embodiment of the present application can perform data cleaning, data standardization, and processing of missing data values ​​on the initial operation data to ensure the consistency and accuracy of the data, and then obtain the final operation data that meets certain conditions. Among them, the certain conditions can be set by technicians in this field according to actual conditions, and this application does not make specific restrictions.

[0045] Optionally, in one embodiment of the present application, before inputting the operating data into the pre-trained time series prediction model, it also includes: constructing the Kronecker product layer of the time series prediction model using the Kronecker product and the rotation mask, so as to obtain the input matrix of the final operating data using the Kronecker product layer; constructing the kernel matrix layer of the time series prediction model in combination with the input matrix and the kernel function, so as to obtain the output matrix corresponding to the input matrix according to the kernel matrix layer; constructing the multi-layer perceptron layer of the time series prediction model using the output matrix and the linear multi-layer perceptron, so as to construct the pre-trained time series prediction model. Among them, the mathematical expression of the multi-layer perceptron layer can be, but is not limited to, as follows:

[0046] y(k)=b+k1×y (1) (k)+…+k n ×y (n) (k)

[0047] Among them, y n (k) is the n-order output result of the model, b is the bias term of the multilayer perceptron, k i is the weight of the corresponding order result.

[0048] In some embodiments, the Kronecker product layer of the present application may include, but is not limited to, a rotation mask. When two identical matrices are subjected to a Kronecker product to generate repeated high-order terms, an N-order rotation mask is formed in the form of an N-order input matrix X. N Has exactly the same form, can be used for X after N-order Kronecker product N Tidy up to remove duplicates in higher order cases.

[0049] It can be understood that the embodiments of the present application can use the Kronecker product layer to perform multi-order Kronecker product operations on the input matrix to form an N-order input matrix to express the high-order nonlinear relationship of the system, and use a rotating mask to mask the N-order input matrix output by the Kronecker product layer to eliminate repeated high-order terms to reduce the number of parameters.

[0050] In some embodiments, the kernel matrix layer of the embodiment of the present application may, but is not limited to, include N kernel functions, and the kernel functions are applied to the input matrix of the corresponding order to obtain the output matrix of the corresponding order, that is, the embodiment of the present application can obtain the output matrix of the corresponding order through the N-order input matrix.

[0051] In some embodiments, the multilayer perceptron layer of the embodiment of the present application can splice the output matrix and output it through a linear multilayer perceptron, so that the obtained prediction result is mathematically representable, and it can clearly know how each input variable affects the prediction result, which has transparency and credibility in practical applications.

[0052] It can be understood that the multi-layer perceptron layer of the embodiment of the present application abandons the multi-layer perceptron of the activation function, concatenates the output matrix of the kernel matrix layer, and inputs it into the multi-layer perceptron layer for linear combination to obtain the prediction result, which has stronger interpretability. Among them, the mathematical expression of the multi-layer perceptron layer can be but is not limited to: y(k) = b + k1 × y (1) (k)+…+k n ×y (n) (k).

[0053] Optionally, in one embodiment of the present application, a Kronecker product layer of a time series prediction model is constructed using Kronecker products and a rotation mask, so as to use the Kronecker product layer to obtain an input matrix of final operating data, including: obtaining an initial input matrix corresponding to the final operating data based on the final operating data; calculating the Kronecker product matrix of the final operating data using the initial input matrix and the Kronecker product; and obtaining the input matrix based on the Kronecker product matrix and the rotation mask.

[0054] As a possible implementation method, the embodiment of the present application can obtain an initial input matrix based on the final operation data, calculate the Kronecker product matrix of the final operation data using the initial input matrix and the Kronecker product, and then obtain the input matrix using a rotation mask.

[0055] Exemplarily, the embodiment of the present application obtains an initial input matrix based on the final operating data, performs multi-order Kronecker product operations on the initial input matrix to form a Kronecker product matrix to intuitively express the relationship between the variables, making it easier for users to understand and trust their prediction results. When two identical matrices produce repeated high-order terms when performing Kronecker product, a rotating mask is used to sort them out to ignore the high-order repeated terms, thereby obtaining the input matrix.

[0056] Optionally, in one embodiment of the present application, a pre-trained time series prediction model is constructed, including: obtaining a training data set and a test data set according to the final operation data; and generating a pre-trained time series prediction model based on the training data set, the test data set, MSE and MAE.

[0057] It can be understood that after obtaining the final operating data, the embodiment of the present application can divide the final operating data into a training data set and a test data set, and normalize the training data set and the test data set so that the input data has a more stable gradient, so as to reduce the magnitude difference between the features and improve the accuracy of the model.

[0058] For example, the embodiment of the present application can divide the final operating data according to the following method: select the first eighty percent of the final operating data as the training data set, and the last twenty percent as the test data set, so as to fully capture the dynamic characteristics of the cooling system and evaluate the predictive performance of the model on unseen data sets, providing a reliable basis for model training and verification.

[0059] Furthermore, the embodiments of the present application can also use MSE and MAE to evaluate the prediction accuracy of the model, and can perform interpretable analysis, so as to utilize the variable relationship and impact analysis provided by the model to formulate a specific optimization plan.

[0060] In the embodiment of the present application, the calculation formula of MSE can be, but is not limited to, expressed as:

[0061]

[0062] The calculation formula of MAE can be expressed as but not limited to:

[0063]

[0064] Among them, y i For the prediction target, is the predicted value.

[0065] In step S103, the final operation data is input into a pre-trained time series prediction model to obtain a prediction result of the return water temperature of the cooling system, wherein the time series prediction model has a Kronecker product layer, a kernel matrix layer and a multi-layer perceptron layer.

[0066] It can be understood that the time series prediction model of the embodiment of the present application may include a Kronecker product layer, a kernel matrix layer, a multi-layer perceptron layer, etc., and may also include a data set loading module, a model configuration module, a tool set module, a test module and a parser module, etc., which can be specifically configured by technicians in this field according to actual conditions, and this application does not impose any specific restrictions.

[0067] In the actual execution process, the embodiment of the present application can input the final operation data into the pre-trained time series prediction model to obtain the prediction result of the return water temperature of the cooling system.

[0068] For example, Figure 2 As shown, the time series prediction model of the embodiment of the present application includes a three-layer structure: a Kronecker product layer, a kernel matrix layer, and a multi-layer perception machine layer, which aims to accurately predict the return water temperature of the cooling system and the average temperature and humidity of the return air, intuitively express the relationship between variables, conduct detailed variable impact analysis, and provide effective system optimization strategies, thereby improving the overall performance of the cooling system and ensuring the safe and efficient operation of the server room.

[0069] In general, if Figure 3-Figure 5 As shown, the embodiment of the present application pre-processes the initial operating data of the cooling system collected by the sampling sensor, such as the inlet water temperature, the return water temperature, the power consumption of IT equipment, the water supply pressure difference, the average return air temperature, the average return air humidity, etc., to obtain the final operating data that meets certain conditions, and then inputs these final operating data into the pre-trained time series prediction model, so as to predict and output important parameters such as the east side return water temperature, the west side return water temperature, and the average return air temperature and humidity.

[0070] Specifically, after the final operation data of the embodiment of the present application is converted into the initial input matrix through the input layer, a multi-order Kronecker product operation is performed to generate an N-order Kronecker product matrix. For any Kronecker product matrix X, the mapping relationship between the numbers before and after expand is as follows: X(i, j) = X e (m), where: m = (i-1) * N + j. The Kronecker product matrix X after the N-order Kronecker product is denoted as X N , where X N The shape of can be, but is not limited to, represented as (length,…,length), and its dimension is N, then there is the following relationship: N (i,j,…,n)=X e (i)×X e (j)×…×X e (n). Where X N The matrix is ​​composed of N X e The result obtained from the matrix Kronecker product has a large number of repeated high-order terms.

[0071] Furthermore, in order to solve the problem of repeated high-order terms, the embodiment of the present application creates an N-order kernel matrix layer, which is formally an N-order Kronecker product matrix X N has exactly the same form, each element of which can be derived from the following formula, which can be expressed as but not limited to:

[0072]

[0073] In addition, the kernel matrix layer of the embodiment of the present application includes N kernel functions, and the kernel functions are applied to the input matrix of the corresponding order, so as to obtain the output matrix of the corresponding order of the N-order input matrix.

[0074] Then, the embodiment of the present application further processes the output matrix in the multi-layer perceptron layer after splicing. Unlike the traditional multi-layer perceptron layer, the multi-layer perceptron layer of the embodiment of the present application does not contain a nonlinear activation function, that is, the output matrix obtained in the previous stage is directly input into the multi-layer perceptron layer for linear combination, and the prediction result is obtained through linear superposition of corresponding orders. The mathematical expression of the multi-layer perceptron layer can be, but is not limited to, y(k)=b+k1×y (1) (k)+…+k n ×y (n) (k), where the output result y for any order (m) (k), which can be expressed as, but not limited to: (m) (k) = h m ×X N , where, in the embodiment of the present application, × is redefined through the einsum operation, and its einsum can be described as: a…, b…→ab.

[0075] In addition, it should be noted that the embodiment of the present application captures the nonlinear characteristics of the system and the correlation between covariates through the kernel function, abandons the traditional activation function to capture nonlinear characteristics, and is proven to be effective through experiments.

[0076] The embodiments of the present application can surpass the best existing models in multiple data sets, significantly reduce the computational requirements required for fitting, and have the following advantages:

[0077] (1) Accurate prediction of the return water temperature of the cooling system: The embodiments of the present application require that the pre-trained time series prediction model constructed can process a large amount of data, which may include but is not limited to inlet water temperature, return water temperature, IT equipment power consumption, water supply pressure difference, return air average temperature, return air average humidity, etc., comprehensively capture the dynamic characteristics of the cooling system, and can effectively utilize these data to improve the accuracy of the model prediction effect, ensuring that reliable prediction results can be provided under various operating conditions.

[0078] (2) Model interpretability: The pre-trained time series prediction model constructed in the embodiment of the present application can not only accurately predict the return water temperature of the cooling system, but also intuitively express the relationship between the variables. The model parameters can not only reflect the specific contribution and mutual influence of each variable on the prediction results, helping users understand the internal mechanism of the model, but also conduct detailed impact analysis of the variables, quantify the degree of influence of different variables on the return water temperature of the cooling system, and clarify the role and function of each variable in the prediction through techniques such as feature importance analysis and partial dependence diagram.

[0079] (3) Proposing and verifying cooling system optimization strategies: The embodiments of the present application can utilize the variable relationships and impact analysis provided by the pre-trained time series prediction model to formulate specific optimization plans, such as adjusting the supply and return water temperature and pressure, and verify the effectiveness of these optimization strategies through simulation and actual application, thereby improving the efficiency and stability of the cooling system and ensuring that the equipment in the server room operates stably at the optimal temperature.

[0080] Combine the following Figure 2-Figure 5 As shown, a specific embodiment is used to introduce in detail the method for predicting the return water temperature of the data center cooling system proposed in the embodiment of the present application.

[0081] like Figure 2 As shown, the time series prediction model of the embodiment of the present application includes a three-layer structure: a Kronecker product layer, a kernel matrix layer and a multi-layer perceptron layer.

[0082] Further, in the embodiments of the present application, Figure 3-Figure 5 As shown, the initial operation data of the cooling system collected by the sampling sensor is preprocessed to obtain the final operation data that meets certain conditions, and then these final operation data are input into the pre-trained time series prediction model to predict and output important parameters such as the east return water temperature, the west return water temperature, and the average temperature and humidity of the return air.

[0083] Specifically, after the final operation data of the embodiment of the present application is converted into an initial input matrix through the input layer, a multi-order Kronecker product operation is performed to generate an N-order Kronecker product matrix. N It is composed of N X e The result obtained from the matrix Kronecker product has a large number of repeated high-order terms.

[0084] Furthermore, in order to solve the problem of repeated high-order terms, the embodiment of the present application creates an N-order kernel matrix layer, applies the kernel function to the input matrix of the corresponding order, and obtains the output matrix of the corresponding order.

[0085] Furthermore, the embodiment of the present application concatenates the output matrices and further processes them in a multi-layer perceptron layer, and obtains a prediction result through linear superposition of corresponding orders.

[0086] According to the prediction method of the return water temperature of the cooling system of the data center proposed in the embodiment of the present application, the initial operation data of the cooling system in the collected data center can be preprocessed to obtain the final operation data that meets certain conditions, and the final operation data is input into the pre-trained time series prediction model to obtain the prediction result of the return water temperature of the cooling system, so as to accurately predict the return water temperature of the cooling system and the average temperature and humidity of the return air, intuitively express the relationship between variables, conduct detailed variable impact analysis, and provide effective system optimization strategies, thereby improving the overall performance of the cooling system and ensuring the safety and efficient operation of the server room. In this way, the problems in the related technology such as the large number of model parameters, the black box caused by the expression of nonlinear characteristics through complex activation functions, and the difficulty in intuitively expressing the relationship between variables are solved, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results.

[0087] Next, a device for predicting the return water temperature of a data center cooling system proposed in an embodiment of the present application will be described with reference to the accompanying drawings.

[0088] Figure 6 A block diagram of a device for predicting return water temperature of a data center cooling system provided according to an embodiment of the present application.

[0089] like Figure 6 As shown, the prediction device 10 for the return water temperature of the data center cooling system includes: a collection module 100 , a preprocessing module 200 and a prediction module 300 .

[0090] The acquisition module 100 is used to acquire initial operation data of the cooling system in the data center.

[0091] The preprocessing module 200 is used to preprocess the initial operation data to obtain final operation data that meets preset conditions.

[0092] The prediction module 300 is used to input the final operation data into a pre-trained time series prediction model to obtain a prediction result of the return water temperature of the cooling system, wherein the time series prediction model has a Kronecker product layer, a kernel matrix layer and a multi-layer perceptron layer.

[0093] Optionally, in one embodiment of the present application, it further includes: a first building module, a second building module and a third building module.

[0094] Among them, the first construction module is used to construct the Kronecker product layer of the time series prediction model using the Kronecker product and the rotation mask before inputting the operating data into the pre-trained time series prediction model, so as to obtain the input matrix of the final operating data using the Kronecker product layer.

[0095] The second construction module is used to construct a kernel matrix layer of the time series prediction model by combining the input matrix and the kernel function, so as to obtain an output matrix corresponding to the input matrix according to the kernel matrix layer.

[0096] The third building module is used to build a multi-layer perceptron layer of the time series prediction model using the output matrix and the linear multi-layer perceptron to build a pre-trained time series prediction model.

[0097] Optionally, in one embodiment of the present application, the third building module includes: an acquisition unit and a first generation unit.

[0098] The acquisition unit is used to acquire a training data set and a test data set according to the final operation data.

[0099] The first generating unit is used to generate a pre-trained time series prediction model based on a training data set, a test data set, MSE and MAE.

[0100] Optionally, in one embodiment of the present application, the first building module includes: a second generating unit, a calculating unit and a third generating unit.

[0101] The second generating unit is used to obtain an initial input matrix corresponding to the final operating data based on the final operating data.

[0102] A calculation unit is used to calculate a Kronecker product matrix of final running data using an initial input matrix and the Kronecker product.

[0103] The third generating unit is used to obtain an input matrix based on the Kronecker product matrix and the rotation mask.

[0104] Optionally, in one embodiment of the present application, the mathematical expression of the multi-layer perceptron layer may be, but is not limited to:

[0105] y(k)=b+k1×y (1) (k)+…+k n ×y (n) (k)

[0106] Among them, y n (k) is the n-order output result of the model, b is the bias term of the multilayer perceptron, k i is the weight of the corresponding order result.

[0107] It should be noted that the above explanation of the embodiment of the method for predicting the return water temperature of a data center cooling system is also applicable to the device for predicting the return water temperature of a data center cooling system of this embodiment, and will not be repeated here.

[0108] According to the prediction device for the return water temperature of the cooling system of the data center proposed in the embodiment of the present application, the initial operation data of the cooling system in the collected data center can be preprocessed to obtain the final operation data that meets certain conditions, and the final operation data can be input into the pre-trained time series prediction model to obtain the prediction result of the return water temperature of the cooling system, so as to accurately predict the return water temperature of the cooling system and the average temperature and humidity of the return air, intuitively express the relationship between variables, conduct detailed variable impact analysis, and provide effective system optimization strategies, thereby improving the overall performance of the cooling system and ensuring the safety and efficient operation of the server room. In this way, the problems in the related technology such as the large number of model parameters, the black box caused by the expression of nonlinear characteristics through complex activation functions, and the difficulty in intuitively expressing the relationship between variables are solved, which limits the effectiveness and interpretability of the model in practical applications and affects the user's understanding and trust in the model prediction results.

[0109] Figure 7 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. The electronic device may include:

[0110] A memory 701 , a processor 702 , and a computer program stored in the memory 701 and executable on the processor 702 .

[0111] When the processor 702 executes the program, the method for predicting the return water temperature of the data center cooling system provided in the above embodiment is implemented.

[0112] Furthermore, the electronic device further comprises:

[0113] The communication interface 703 is used for communication between the memory 701 and the processor 702 .

[0114] The memory 701 is used to store computer programs that can be executed on the processor 702 .

[0115] The memory 701 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0116] If the memory 701, the processor 702 and the communication interface 703 are implemented independently, the communication interface 703, the memory 701 and the processor 702 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0117] Optionally, in a specific implementation, if the memory 701, the processor 702 and the communication interface 703 are integrated on a chip, the memory 701, the processor 702 and the communication interface 703 can communicate with each other through an internal interface.

[0118] The processor 702 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0119] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above method for predicting the return water temperature of a data center cooling system.

[0120] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, implements the above method for predicting the return water temperature of a data center cooling system.

[0121] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0122] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0123] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0124] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0125] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of multiple of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0126] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0127] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0128] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for predicting the return water temperature of a data center cooling system, characterized in that: The following steps are involved: Collect initial operating data of the cooling system in the data center; Preprocessing the initial operation data to obtain final operation data that meets preset conditions; The final operation data is input into a pre-trained time series prediction model to obtain a prediction result of the return water temperature of the cooling system, wherein the time series prediction model has a Kronecker product layer, a kernel matrix layer and a multi-layer perceptron layer.

2. The method according to claim 1, characterized in that Before inputting the operating data into the pre-trained time series prediction model, the method further includes: Constructing a Kronecker product layer of the time series prediction model by using the Kronecker product and the rotation mask, so as to obtain an input matrix of the final operation data by using the Kronecker product layer; Combining the input matrix and the kernel function to construct a kernel matrix layer of the time series prediction model, so as to obtain an output matrix corresponding to the input matrix according to the kernel matrix layer; The multi-layer perceptron layer of the time series prediction model is constructed using the output matrix and a linear multi-layer perceptron to construct the pre-trained time series prediction model.

3. The method according to claim 2, characterized in that The constructing the pre-trained time series prediction model comprises: Acquire a training data set and a test data set according to the final running data; The pre-trained time series prediction model is generated based on the training data set, the test data set, the mean square error (MSE) and the mean absolute error (MAE).

4. The method according to claim 2, characterized in that: The step of constructing the Kronecker product layer of the time series prediction model by using the Kronecker product and the rotation mask, so as to obtain the input matrix of the final operation data by using the Kronecker product layer, comprises: Obtaining an initial input matrix corresponding to the final operation data based on the final operation data; Calculating a Kronecker product matrix of the final operation data using the initial input matrix and the Kronecker product; The input matrix is ​​obtained based on the Kronecker product matrix and the rotation mask.

5. The method according to claim 2, characterized in that: The mathematical expression of the multi-layer perceptron layer is: y(k)=b+k1×y (1) (k)+…+k n ×y (n) (k), Among them, y n (k) is the n-order output result of the model, b is the bias term of the multilayer perceptron, k i is the weight of the corresponding order result.

6. A device for predicting return water temperature of a data center cooling system, characterized in that: include: A collection module, used to collect initial operation data of the cooling system in the data center; A preprocessing module, used for preprocessing the initial operation data to obtain final operation data that meets preset conditions; A prediction module is used to input the final operation data into a pre-trained time series prediction model to obtain a prediction result of the return water temperature of the cooling system, wherein the time series prediction model has a Kronecker product layer, a kernel matrix layer and a multi-layer perceptron layer.

7. The device according to claim 6, characterized in that Also includes: A first construction module is used for constructing a Kronecker product layer of the time series prediction model by using a Kronecker product and a rotation mask before inputting the operation data into a pre-trained time series prediction model, so as to obtain an input matrix of the final operation data by using the Kronecker product layer; A second construction module is used to construct a kernel matrix layer of the time series prediction model by combining the input matrix and the kernel function, so as to obtain an output matrix corresponding to the input matrix according to the kernel matrix layer; The third construction module is used to construct the multi-layer perceptron layer of the time series prediction model by using the output matrix and the linear multi-layer perceptron to construct the pre-trained time series prediction model.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for predicting the return water temperature of a data center cooling system according to any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method for predicting the return water temperature of a data center cooling system as described in any one of claims 1 to 5.

10. A computer program product, characterized in that It comprises a computer program, which, when executed, is used to implement the method for predicting the return water temperature of a data center cooling system as claimed in any one of claims 1 to 5.

Images