Green data center energy consumption monitoring management system and method

Abstract

The invention provides a green data center energy consumption monitoring management system and method, and the system comprises a data prediction module which obtains the power consumption value of a prediction time point and the temperature of each temperature measurement point based on the power consumption values of a plurality of historical time points and the temperature of each temperature measurement point of the plurality of historical time points; the model construction module is used for constructing an equipment temperature model; the air cooling power consumption determination module is used for determining the environment temperature based on the environment temperature model, determining the air cooling demand intensity based on the environment temperature and determining the external air cooling power consumption and the internal air cooling power consumption based on the air cooling demand intensity; the liquid cooling power consumption determination module is used for determining the liquid cooling power consumption of each equipment surface based on the temperature of each equipment surface of the equipment temperature model and determining the total liquid cooling power consumption; the single-module power consumption calculation module is used for calculating a single-module power consumption value; and the power consumption abnormity judgment module is used for calculating a total module power consumption value based on the single module power consumption value, calculating a predicted PUE value based on the total module power consumption value and judging whether power consumption abnormity occurs or not.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a green data center energy consumption monitoring and management system and method. Background Art

[0002] Energy consumption monitoring is particularly important in research on power supply for modular data centers. The "Green Data Center Evaluation Specification" not only provides clear guidance for the construction and operation of green data centers, but also profoundly reveals the critical role of energy consumption monitoring in operations. In the current context of global energy tensions and increasing pressure on environmental protection, power plants, as the core link in energy supply, are becoming increasingly important for monitoring energy consumption in modular data centers.

[0003] The Green Data Center Evaluation Standards clearly outline requirements for efficient energy resource utilization, green design, green procurement, and green energy resource management within data centers. These requirements aim to guide data centers toward a green development path characterized by high efficiency, low carbon, intensive use, and recycling. These requirements not only address the operation and management of data centers themselves, but also pose significant challenges to the power plants that provide their energy.

[0004] In summary, energy consumption monitoring in green data centers plays a vital role in achieving efficient energy utilization, reducing operating costs, addressing environmental protection policies, and promoting the green development of data centers. With the deepening implementation of the "Green Data Center Evaluation Standards" and the increasing global awareness of environmental protection, power plant green data centers should attach greater importance to energy consumption monitoring, continuously improving its accuracy and effectiveness, and providing strong support for the green development of data centers. However, energy consumption monitoring in existing power plant green data centers often only allows for simple data statistics during the data collection phase, without enabling effective regulation and making it difficult to effectively control PUE values.

[0005] In view of this, the present invention is proposed. Summary of the Invention

[0006] The purpose of the present invention is to provide a green data center energy consumption monitoring and management system and method. This solution uses real-time predicted PUE values ​​to perform real-time control, so that the PUE value of the data center is always kept below the PUE threshold, and determines whether power consumption anomalies occur.

[0007] The present invention provides a green data center energy consumption monitoring and management system. The green data center includes multiple micro-module computer rooms, each of which is equipped with IT equipment. The system includes:

[0008] The data prediction module obtains the power consumption value and temperature of each temperature measurement point at the predicted time point based on the power consumption value of IT equipment in the micro-module room at multiple historical time points and the temperature of each temperature measurement point at multiple historical time points;

[0009] A model building module builds an ambient temperature model based on the temperature of the ambient temperature measuring points among the temperature measuring points, and builds an equipment temperature model based on the equipment temperature measuring points among the temperature measuring points;

[0010] an air cooling power consumption determination module, which determines an ambient temperature based on the ambient temperature model, determines an air cooling requirement intensity based on the ambient temperature, and determines an external air cooling power consumption and an internal air cooling power consumption based on the air cooling requirement intensity;

[0011] a liquid cooling power consumption determination module, which determines the temperature of each device surface based on the device temperature model, determines the liquid cooling power consumption of each device surface based on the temperature of each device surface, and determines the total liquid cooling power consumption based on the liquid cooling power consumption of each device surface;

[0012] Single-module power consumption calculation module, which calculates the power consumption of a single module based on the external air cooling power consumption, internal air cooling power consumption, and total liquid cooling power consumption of the micro-module room;

[0013] The power consumption abnormality determination module calculates the total module power consumption value based on the single module power consumption value, calculates the predicted PUE value based on the total module power consumption value, and determines whether power consumption abnormality occurs based on comparison between the predicted PUE value and a preset PUE threshold.

[0014] The above scheme is adopted. This scheme is first based on the structure of a data center composed of multiple micro-module computer rooms. First, data statistics are performed through the input end of each micro-module computer room. Based on the principle that the temperature of the micro-module computer room is mainly caused by IT equipment, the temperature of each temperature measuring point is predicted through the power consumption value of IT equipment and the temperature of the temperature measuring point statistically calculated at multiple historical time points. The predicted temperature of the temperature measuring point is used to determine the predicted external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption based on the principle that the temperature control system self-regulates based on temperature. Finally, the predicted PUE value of the data center is calculated through the predicted external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption of each micro-module computer room, and real-time management and control are performed through the real-time predicted PUE value to ensure that the PUE value of the data center is always kept below the PUE threshold to determine whether power consumption abnormalities occur.

[0015] In some embodiments of the present invention, the system also includes a feedback adjustment module. If power consumption abnormality occurs, the single-module power consumption of each micro-module computer room at a preset number of time points is counted, and the maximum and minimum values ​​of the single-module power consumption of each micro-module computer room at a preset number of time points are used as the upper limit and lower limit respectively. The upper limit of fluctuation is determined based on the upper limit value, the lower limit of fluctuation is determined based on the lower limit value, the fluctuation range is determined based on the upper limit of fluctuation and the lower limit of fluctuation, and whether the corresponding micro-module computer room needs to be checked is determined based on the fluctuation range.

[0016] With the above solution, since the overall PUE anomaly is often related to the consumption of temperature regulation in the micro-module computer room, after determining the overall PUE anomaly, this solution identifies the abnormal micro-module computer room by statistically analyzing the historical situation of each micro-module computer room, which can provide certain guidance for staff management and improve the accuracy of subsequent staff investigations.

[0017] In some embodiments of the present invention, in the steps of determining the upper limit of fluctuation based on the upper limit value and determining the lower limit of fluctuation based on the lower limit value, the difference between the upper limit value and the power consumption of each single module in the power consumption of the single module at a previous preset number of time points is calculated, and the number of single module power consumptions in the difference that is less than the fluctuation threshold is counted as a first number, and the number of differences between the power consumption of the single modules at the previous preset number of time points and the lower limit value that is less than the fluctuation threshold is calculated as a second number. The upper limit of fluctuation and the lower limit of fluctuation are calculated based on the initial fluctuation value, the first number, and the second number. The upper limit of fluctuation and the lower limit of fluctuation are then calculated based on the following formula:

[0018]

[0019] Among them, R T Indicates the upper limit of fluctuation, R L represents the lower limit of fluctuation, r represents the preset initial fluctuation value, Indicates the upper limit value, Indicates the lower limit, N1 indicates the first number, N2 indicates the second number, N t Indicates the number of time points before the preset number, N ρ Indicates the preset basic fluctuation quantity value.

[0020] Using the above solution, in the process of determining whether each micro-module computer room is normal, the solution calculates the number of values ​​near the upper limit and the lower limit by counting the first number and the second number. When the first number is large, it means that the number of values ​​close to the upper limit is large. When calculating the corresponding fluctuation upper limit, If the value is larger, the final calculated If the second number is larger, it can be guaranteed that the number of values ​​close to the upper limit value is larger, and the corresponding upper limit value of the fluctuation increases; when the second number is larger, it means that the number of values ​​close to the lower limit value is larger, then when calculating the corresponding lower limit value of the fluctuation, If the value is larger, the final calculated If it is larger, it can ensure that the number of values ​​close to the lower limit is larger, and the corresponding lower limit of fluctuation is lowered, thus ensuring statistical accuracy.

[0021] In some embodiments of the present invention, in the step of obtaining the power consumption value at a predicted time point and the temperature of each temperature measuring point based on the power consumption values ​​of IT equipment in a micromodule computer room at multiple historical time points and the temperatures of each temperature measuring point at multiple historical time points, the power consumption value at each historical time point and the temperature of the temperature measuring point are used as the value of a dimension in a statistical vector to construct a statistical vector, and the statistical vectors of multiple historical time points are constructed into a statistical matrix, which is input into a trained first prediction model, and the first prediction model outputs a first prediction vector, which includes the power consumption value at the predicted time point and the temperature of each temperature measuring point.

[0022] In some embodiments of the present invention, the model structure of the first prediction model adopts the model structure of a recurrent neural network model.

[0023] In some embodiments of the present invention, in the step of constructing a statistical vector by taking the power consumption value and the temperature of each temperature measurement point at each historical time point as a value of a dimension in the statistical vector:

[0024] The temperature of each temperature measuring point at multiple historical time points is used to construct a temperature change curve for each temperature measuring point, and a total fitting curve is obtained based on the temperature change curve of each temperature measuring point;

[0025] The power consumption change curve of the power consumption value of the micro-module room at multiple historical time points;

[0026] The power consumption variation curve and the total fitting curve are input into a hysteresis calculation model to obtain a hysteresis time step.

[0027] In some embodiments of the present invention, in the step of constructing a temperature change curve for each temperature measuring point by using the temperatures of each temperature measuring point in multiple historical time points, and obtaining a total fitting curve based on the temperature change curve of each temperature measuring point, Gaussian process regression, weighted least squares method or cubic spline interpolation are used for fitting to obtain the total fitting curve.

[0028] In some embodiments of the present invention, in the step of constructing a statistical vector by taking the power consumption value and the temperature of the temperature measurement point at each historical time point as the value of a dimension in the statistical vector, the temperature change time point corresponding to the time point of the power consumption value is determined based on the numerical value of the lag time step, and the power consumption value at the time point and the temperature of each temperature measurement point in the corresponding temperature change time point are constructed as a statistical vector.

[0029] The above scheme is adopted. First, based on the fact that power consumption value is often correlated with temperature, this scheme constructs the power consumption value and the temperature of each temperature measurement point into a statistical vector. In the process of constructing the statistical vector, based on the fact that the change in temperature often lags behind the change in power consumption value, this scheme fits the temperature change curve of each temperature measurement point. The fitted temperature change curve and power consumption change curve of the temperature measurement point are input into the lag calculation model to obtain the lag time step. The time point of the corresponding power consumption value and the time point of the temperature measurement value are determined through the lag time step, and constructed as a statistical vector to ensure the calculation accuracy.

[0030] In some embodiments of the present invention, in the steps of constructing an environmental temperature model based on the temperature of the environmental temperature measuring points among the temperature measuring points, and constructing a device temperature model based on the device temperature measuring points among the temperature measuring points;

[0031] Determine a temperature distribution map of a plane based on the predicted values ​​of the temperature measurement points of each plane set on the inner wall of the micro-module computer room among the environmental temperature measurement points, and construct an environmental temperature model based on the temperature distribution maps of all the inner walls;

[0032] The temperature of each device is determined based on the temperature measuring points set on the outer surface of the IT device among the device temperature measuring points, and a three-dimensional framework pre-built for the device is rendered based on the temperature of the device to obtain a device temperature model.

[0033] In some embodiments of the present invention, in the step of determining a temperature distribution map of a plane based on the predicted values ​​of the temperature measurement points of each plane set on the inner wall of the micromodule room among the environmental temperature measurement points,

[0034] For the model blocks covered by the temperature measurement points on the plane, the corresponding pixel values ​​are determined based on the temperature values ​​and rendered;

[0035] For a model block on the plane that is not covered by a temperature measurement point, calculate the distance value between the model block and the temperature measurement point within the threshold distance range of the model block, and determine the temperature influence weight based on the distance value;

[0036] Based on the temperature influence weight, a weighted calculation is performed on the temperature values ​​of each temperature measurement point within a threshold distance range from the model block to determine the temperature value of the corresponding model block, and a pixel value corresponding to the temperature value is determined for rendering.

[0037] During the specific implementation process, in the step of determining the temperature of each device based on the temperature measuring points set on the outer surface of the IT equipment among the equipment temperature measuring points, and rendering the three-dimensional framework pre-built for the equipment based on the temperature of the equipment to obtain the equipment temperature model, the rendering method of the temperature distribution map of the outer surface of the IT equipment is the same as the method of determining the temperature distribution map of a plane based on the predicted values ​​of the temperature measuring points on each plane set on the inner wall of the micro-module computer room among the environmental temperature measuring points.

[0038] Adopting the above scheme, this scheme ensures the calculation accuracy of the overall temperature distribution by constructing the temperature distribution map of each plane in the temperature rendering model of the micro-module computer room and each plane of the IT equipment, and thus determines the calculation accuracy of the overall temperature.

[0039] In some embodiments of the present invention, in the steps of determining the ambient temperature based on the ambient temperature model, determining the air cooling demand intensity based on the ambient temperature, and determining the external air cooling power consumption and the internal air cooling power consumption based on the air cooling demand intensity, the temperature distribution map of each plane in the ambient temperature model is input into a preset temperature calculation model, the ambient temperature is determined by the preset temperature calculation model, the temperature difference between the ambient temperature and the preset ambient temperature threshold is calculated, and the external air cooling power consumption and the internal air cooling power consumption are determined based on the temperature difference.

[0040] In the specific implementation process, in the step of inputting the temperature distribution map of each plane in the ambient temperature model into a preset temperature calculation model and determining the ambient temperature through the preset temperature calculation model, the pixel value of each position in the temperature distribution map of each plane is used as the value of a position in the corresponding matrix to obtain a temperature distribution matrix for the temperature distribution map of each plane, and all the temperature distribution matrices are input into the preset temperature calculation model, and the temperature calculation model outputs the ambient temperature.

[0041] In some embodiments of the present invention, in the step of determining the temperature of each device surface based on the device temperature model, the liquid cooling power consumption is calculated based on the temperature of each device and a preset device temperature threshold.

[0042] By using the above solution, the present invention accurately predicts the temperature of the device and the temperature of the environment in which the device is located, and determines the corresponding power consumption through a preset comparison table, thereby ensuring the accuracy of the calculation of the power consumption.

[0043] In some embodiments of the present invention, in the step of calculating the single module power consumption value based on the external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption of the micro-module computer room, the external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption are added to obtain the single module power consumption value.

[0044] In some embodiments of the present invention, in the steps of calculating the total module power consumption value based on the single module power consumption value and calculating the predicted PUE value based on the total module power consumption value, the predicted PUE value is calculated using the following formula:

[0045]

[0046] Among them, PUE pe represents the predicted PUE value, N represents the total number of micromodule rooms, and W n represents the power consumption of a single module in the micro-module room n, w IT Indicates the total power consumption of IT equipment, W PD Indicates the power consumption of the power distribution equipment, W + Indicates the power consumption of auxiliary equipment.

[0047] In some embodiments of the present invention, the following formula is used to calculate the power consumption value of the power distribution equipment and the power consumption value of the auxiliary equipment:

[0048]

[0049] Among them, δ1 represents the total proportion of the power consumption values ​​of each micro-module room, δ2 represents the proportion of the total power consumption of IT equipment, δ3 represents the proportion of the power consumption value of the power distribution equipment, and δ4 represents the proportion of the power consumption value of the auxiliary equipment.

[0050] In a specific implementation process, δ1+δ2=0.85, δ3=0.1, δ4=0.05.

[0051] By adopting the above scheme, this scheme can accurately predict the power consumption value of the micro-module computer room to achieve accurate prediction of the PUE value. It can perform real-time management and control through the real-time predicted PUE value, so that the PUE value of the data center is always kept below the PUE threshold. This scheme completes data prediction through data statistics, determines the predicted PUE value through the predicted data, realizes effective adjustment, and realizes effective control of the PUE value.

[0052] Another aspect of the present invention also relates to a green data center energy consumption monitoring and management method, the method comprising the following steps:

[0053] Based on the power consumption values ​​of IT equipment in the micro-module room at multiple historical time points and the temperatures of various temperature measurement points at multiple historical time points, the power consumption values ​​and temperatures of various temperature measurement points at the predicted time point are obtained;

[0054] An ambient temperature model is constructed based on the temperature of the ambient temperature measurement points among the temperature measurement points, and an equipment temperature model is constructed based on the temperature of the equipment temperature measurement points among the temperature measurement points;

[0055] determining an ambient temperature based on the ambient temperature model, determining an air cooling requirement intensity based on the ambient temperature, and determining an external air cooling power consumption and an internal air cooling power consumption based on the air cooling requirement intensity;

[0056] determining the temperature of each device surface based on the device temperature model, determining the liquid cooling power consumption of each device surface based on the temperature of each device surface, and determining the total liquid cooling power consumption based on the liquid cooling power consumption of each device surface;

[0057] Calculate the power consumption of a single module based on the external air cooling power consumption, internal air cooling power consumption, and total liquid cooling power consumption of the micro-module room;

[0058] The total module power consumption value is calculated based on the single module power consumption value, the predicted PUE value is calculated based on the total module power consumption value, and whether power consumption abnormality occurs is determined based on the comparison between the predicted PUE value and a preset PUE threshold.

[0059] In summary, the present invention has the following beneficial effects:

[0060] 1. This solution is based on the structure of a data center consisting of multiple micro-module computer rooms. First, data statistics are collected through the input end of each micro-module computer room. Based on the principle that the temperature of the micro-module computer room is mainly caused by IT equipment, the temperature of each temperature measuring point is predicted through the power consumption values ​​of IT equipment and the temperature of the temperature measuring point statistically collected at multiple historical time points. Based on the predicted temperature of the temperature measuring point and the principle that the temperature control system self-regulates based on temperature, the predicted external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption are determined. Finally, the predicted PUE value of the data center is calculated based on the predicted external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption of each micro-module computer room. Real-time management and control are carried out through the real-time predicted PUE value to ensure that the PUE value of the data center is always kept below the PUE threshold and to determine whether abnormal power consumption temperature occurs.

[0061] 2. In the process of determining whether each micro-module computer room is normal, this solution counts the number of values ​​near the upper limit and the lower limit by counting the first number and the second number. When the first number is large, it means that there are more values ​​close to the upper limit. In calculating the corresponding fluctuation upper limit, If the value is larger, the final calculated If the second number is larger, it can be guaranteed that the number of values ​​close to the upper limit value is larger, and the corresponding upper limit value of the fluctuation increases; when the second number is larger, it means that the number of values ​​close to the lower limit value is larger, then when calculating the corresponding lower limit value of the fluctuation, If the value is larger, the final calculated If it is larger, it can ensure that the number of values ​​close to the lower limit is larger, and the corresponding fluctuation lower limit is lowered, thus ensuring statistical accuracy;

[0062] 3. This solution is based on the fact that power consumption values ​​are often correlated with temperature. This solution constructs the power consumption value and the temperature of each temperature measurement point into a statistical vector. In the process of constructing the statistical vector, since the change in temperature often lags behind the change in power consumption value, this solution fits the temperature change curves of each temperature measurement point. The fitted temperature change curves and power consumption change curves of the temperature measurement points are input into the hysteresis calculation model to obtain the hysteresis time step. The corresponding time points of power consumption values ​​and temperature measurement values ​​are determined through the hysteresis time step, and then constructed into a statistical vector to ensure calculation accuracy.

[0063] 4. This solution accurately predicts the power consumption of the micro-module computer room to achieve accurate prediction of the PUE value. It can perform real-time management and control through the real-time predicted PUE value, so that the PUE value of the data center is always kept below the PUE threshold. This solution completes data prediction through data statistics, determines the predicted PUE value based on the predicted data, realizes effective adjustment, and realizes effective control of the PUE value. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0065] Figure 1 A schematic diagram of an embodiment of the green data center energy consumption monitoring and management system of the present invention;

[0066] Figure 2 A schematic diagram of another embodiment of the green data center energy consumption monitoring and management system of the present invention;

[0067] Figure 3 Schematic diagram of the processing process of the feedback adjustment module in the green data center energy consumption monitoring and management system of the present invention;

[0068] Figure 4 Schematic diagram of the processing process of the data prediction module in the green data center energy consumption monitoring and management system of the present invention;

[0069] Figure 5 This is a schematic diagram of an implementation of the green data center energy consumption monitoring and management method of the present invention. DETAILED DESCRIPTION

[0070] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present invention. Rather, they are merely examples of systems and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0071] The terms used in this invention are for the purpose of describing specific embodiments only and are not intended to limit the invention. The singular forms "a," "the," and "the" used in this invention and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0072] like Figure 1 As shown, the present invention provides a green data center energy consumption monitoring and management system, wherein the green data center comprises a plurality of micro-module computer rooms, each of which is equipped with IT equipment. The system comprises:

[0073] The data prediction module 100 obtains the power consumption value and the temperature of each temperature measurement point at a predicted time point based on the power consumption values ​​of IT equipment in the micro-module computer room at multiple historical time points and the temperatures of each temperature measurement point at multiple historical time points;

[0074] In a specific implementation process, the power consumption values ​​of the multiple historical time points and the temperatures of the temperature measurement points of the multiple historical time points are all time points adjacent to the current time point.

[0075] A model building module 200 builds an environmental temperature model based on the temperature of the environmental temperature measuring points among the temperature measuring points, and builds a device temperature model based on the device temperature measuring points among the temperature measuring points;

[0076] an air cooling power consumption determination module 300 for determining an ambient temperature based on the ambient temperature model, determining an air cooling requirement intensity based on the ambient temperature, and determining external air cooling power consumption and internal air cooling power consumption based on the air cooling requirement intensity;

[0077] a liquid cooling power consumption determination module 400 that determines the temperature of each device surface based on the device temperature model, determines the liquid cooling power consumption of each device surface based on the temperature of each device surface, and determines the total liquid cooling power consumption based on the liquid cooling power consumption of each device surface;

[0078] During the specific implementation process, a comparison table is preset in this solution, in which an air cooling demand intensity value is set corresponding to the ambient temperature, and a liquid cooling power consumption value is set corresponding to the temperature of each device surface of the equipment.

[0079] The single module power consumption calculation module 500 calculates the power consumption value of a single module based on the external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption of the micro-module computer room;

[0080] In the specific implementation process, the sum of the three is taken as the power consumption value of a single module.

[0081] The power consumption abnormality determination module 600 calculates the total module power consumption value based on the single module power consumption value, calculates the predicted PUE value based on the total module power consumption value, and determines whether power consumption abnormality occurs based on the comparison between the predicted PUE value and a preset PUE threshold.

[0082] In a specific implementation process, in the step of determining whether power consumption anomaly occurs based on comparing the predicted PUE value with the preset PUE threshold, if the predicted PUE value is greater than the preset PUE threshold, it is determined that power consumption anomaly occurs.

[0083] The above scheme is adopted. This scheme is first based on the structure of a data center composed of multiple micro-module computer rooms. First, data statistics are performed through the input end of each micro-module computer room. Based on the principle that the temperature of the micro-module computer room is mainly caused by IT equipment, the temperature of each temperature measuring point is predicted through the power consumption value of IT equipment and the temperature of the temperature measuring point statistically calculated at multiple historical time points. The predicted temperature of the temperature measuring point is used to determine the predicted external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption based on the principle that the temperature control system self-regulates based on temperature. Finally, the predicted PUE value of the data center is calculated through the predicted external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption of each micro-module computer room, and real-time management and control are performed through the real-time predicted PUE value to ensure that the PUE value of the data center is always kept below the PUE threshold to determine whether power consumption abnormalities occur.

[0084] like Figure 2 and 3 As shown, in some embodiments of the present invention, the system further includes a feedback adjustment module 700. If abnormal power consumption occurs,

[0085] Step 710, counting the power consumption of a single module in each micro-module room at a preset number of time points;

[0086] Step 720: The maximum and minimum power consumption values ​​of the single modules in each micro-module computer room at the previous preset number of time points are used as the upper limit and lower limit, respectively.

[0087] The upper limit of fluctuation is determined based on the upper limit, the lower limit of fluctuation is determined based on the lower limit, the fluctuation range is determined based on the upper limit and the lower limit, and whether the corresponding micromodule computer room needs to be checked is determined based on the fluctuation range.

[0088] During the specific implementation process, in the step of determining whether the corresponding micro-module computer room needs to be verified based on the fluctuation range, it is determined whether the predicted single-module power consumption value of the micro-module computer room is within the fluctuation range. If not, verification is required; if so, verification is not required.

[0089] With the above solution, since the overall PUE anomaly is often related to the consumption of temperature regulation in the micro-module computer room, after determining the overall PUE anomaly, this solution identifies the abnormal micro-module computer room by statistically analyzing the historical situation of each micro-module computer room, which can provide certain guidance for staff management and improve the accuracy of subsequent staff investigations.

[0090] In some embodiments of the present invention, the steps of determining the upper limit of fluctuation based on the upper limit and determining the lower limit of fluctuation based on the lower limit include step 730 of calculating the difference between the upper limit and the power consumption of each single module in the power consumption of the single module at a previous preset number of time points, and counting the number of single modules whose power consumption is less than the fluctuation threshold in the difference as a first number; step 740 of calculating the number of single modules whose power consumption is less than the fluctuation threshold in the difference between the power consumption of the single module at a previous preset number of time points and the lower limit as a second number; and step 750 of calculating the upper limit of fluctuation and the lower limit of fluctuation based on the initial fluctuation value, the first number, and the second number, and then calculating the upper limit of fluctuation and the lower limit of fluctuation based on the following formula:

[0091]

[0092] Among them, R T Indicates the upper limit of fluctuation, R L represents the lower limit of fluctuation, r represents the preset initial fluctuation value, Indicates the upper limit value, Indicates the lower limit, N1 indicates the first number, N2 indicates the second number, N t Indicates the number of time points before the preset number, N ρ Indicates the preset basic fluctuation quantity value.

[0093] In some embodiments of the present invention, step 760 determines a fluctuation range based on the fluctuation upper limit and the fluctuation lower limit, and determines whether the corresponding micro-module computer room needs to be checked based on the fluctuation range.

[0094] Using the above solution, in the process of determining whether each micro-module computer room is normal, the solution calculates the number of values ​​near the upper limit and the lower limit by counting the first number and the second number. When the first number is large, it means that the number of values ​​close to the upper limit is large. When calculating the corresponding fluctuation upper limit, If the value is larger, the final calculated If the second number is larger, it can be guaranteed that the number of values ​​close to the upper limit value is larger, and the corresponding upper limit value of the fluctuation increases; when the second number is larger, it means that the number of values ​​close to the lower limit value is larger, then when calculating the corresponding lower limit value of the fluctuation, If the value is larger, the final calculated If it is larger, it can ensure that the number of values ​​close to the lower limit is larger, and the corresponding lower limit of fluctuation is lowered, thus ensuring statistical accuracy.

[0095] like Figure 4 As shown, in some embodiments of the present invention, the step of obtaining the power consumption value at a predicted time point and the temperature of each temperature measuring point based on the power consumption values ​​of IT equipment in a micromodule computer room at multiple historical time points and the temperatures of each temperature measuring point at multiple historical time points includes: step S110, taking the power consumption value at each historical time point and the temperature of the temperature measuring point as the value of a dimension in a statistical vector to construct a statistical vector; step S120, constructing the statistical vectors of multiple historical time points into a statistical matrix, inputting the statistical matrix into a trained first prediction model, and the first prediction model outputting a first prediction vector, which includes the power consumption value at the predicted time point and the temperature of each temperature measuring point.

[0096] In some embodiments of the present invention, the model structure of the first prediction model adopts the model structure of a recurrent neural network model.

[0097] Specifically, a bidirectional LSTM model structure (Bi-LSTM) can be used. The bidirectional LSTM model structure uses a bidirectional processing sequence, combines forward and backward information, and improves context understanding capabilities.

[0098] In some embodiments of the present invention, in the step of constructing a statistical vector by taking the power consumption value and the temperature of each temperature measurement point at each historical time point as a value of a dimension in the statistical vector:

[0099] The temperature of each temperature measuring point at multiple historical time points is used to construct a temperature change curve for each temperature measuring point, and a total fitting curve is obtained based on the temperature change curve of each temperature measuring point;

[0100] The power consumption change curve of the power consumption value of the micro-module room at multiple historical time points;

[0101] The power consumption variation curve and the total fitting curve are input into a hysteresis calculation model to obtain a hysteresis time step.

[0102] In some embodiments of the present invention, in the step of constructing a temperature change curve for each temperature measuring point by using the temperatures of each temperature measuring point in multiple historical time points, and obtaining a total fitting curve based on the temperature change curve of each temperature measuring point, Gaussian process regression, weighted least squares method or cubic spline interpolation are used for fitting to obtain the total fitting curve.

[0103] In some embodiments of the present invention, the hysteresis calculation model adopts a recurrent neural network (RNN) structure, specifically an LSTM model structure, in which the impact of early power consumption changes on temperature is long-term memorized through cell state.

[0104] In some embodiments of the present invention, in the step of constructing a statistical vector by taking the power consumption value and the temperature of the temperature measurement point at each historical time point as the value of a dimension in the statistical vector, the temperature change time point corresponding to the time point of the power consumption value is determined based on the numerical value of the lag time step, and the power consumption value at the time point and the temperature of each temperature measurement point in the corresponding temperature change time point are constructed as a statistical vector.

[0105] The above scheme is adopted. First, based on the fact that power consumption value is often correlated with temperature, this scheme constructs the power consumption value and the temperature of each temperature measurement point into a statistical vector. In the process of constructing the statistical vector, based on the fact that the change in temperature often lags behind the change in power consumption value, this scheme fits the temperature change curve of each temperature measurement point. The fitted temperature change curve and power consumption change curve of the temperature measurement point are input into the lag calculation model to obtain the lag time step. The time point of the corresponding power consumption value and the time point of the temperature measurement value are determined through the lag time step, and constructed as a statistical vector to ensure the calculation accuracy.

[0106] In some embodiments of the present invention, in the steps of constructing an environmental temperature model based on the temperature of the environmental temperature measuring points among the temperature measuring points, and constructing a device temperature model based on the device temperature measuring points among the temperature measuring points;

[0107] Determine a temperature distribution map of a plane based on the predicted values ​​of the temperature measurement points of each plane set on the inner wall of the micro-module computer room among the environmental temperature measurement points, and construct an environmental temperature model based on the temperature distribution maps of all the inner walls;

[0108] The temperature of each device is determined based on the temperature measuring points set on the outer surface of the IT device among the device temperature measuring points, and a three-dimensional framework pre-built for the device is rendered based on the temperature of the device to obtain a device temperature model.

[0109] In some embodiments of the present invention, in the step of determining a temperature distribution map of a plane based on the predicted values ​​of the temperature measurement points of each plane set on the inner wall of the micromodule room among the environmental temperature measurement points,

[0110] For the model blocks covered by the temperature measurement points on the plane, the corresponding pixel values ​​are determined based on the temperature values ​​and rendered;

[0111] For a model block on the plane that is not covered by a temperature measurement point, calculate the distance value between the model block and the temperature measurement point within the threshold distance range of the model block, and determine the temperature influence weight based on the distance value;

[0112] Based on the temperature influence weight, a weighted calculation is performed on the temperature values ​​of each temperature measurement point within a threshold distance range from the model block to determine the temperature value of the corresponding model block, and a pixel value corresponding to the temperature value is determined for rendering.

[0113] In the specific implementation process, for each plane surface, this solution divides it into a preset number of model blocks.

[0114] Specifically, for the correspondence between temperature values ​​and pixel values, this solution provides a comparison table.

[0115] During the specific implementation process, in the step of determining the temperature of each device based on the temperature measuring points set on the outer surface of the IT equipment among the equipment temperature measuring points, and rendering the three-dimensional framework pre-built for the equipment based on the temperature of the equipment to obtain the equipment temperature model, the rendering method of the temperature distribution map of the outer surface of the IT equipment is the same as the method of determining the temperature distribution map of a plane based on the predicted values ​​of the temperature measuring points on each plane set on the inner wall of the micro-module computer room among the environmental temperature measuring points.

[0116] Adopting the above scheme, this scheme ensures the calculation accuracy of the overall temperature distribution by constructing the temperature distribution map of each plane in the temperature rendering model of the micro-module computer room and each plane of the IT equipment, and thus determines the calculation accuracy of the overall temperature.

[0117] In some embodiments of the present invention, in the steps of determining the ambient temperature based on the ambient temperature model, determining the air cooling demand intensity based on the ambient temperature, and determining the external air cooling power consumption and the internal air cooling power consumption based on the air cooling demand intensity, the temperature distribution map of each plane in the ambient temperature model is input into a preset temperature calculation model, the ambient temperature is determined by the preset temperature calculation model, the temperature difference between the ambient temperature and the preset ambient temperature threshold is calculated, and the external air cooling power consumption and the internal air cooling power consumption are determined based on the temperature difference.

[0118] During the specific implementation process, in the step of determining the external air cooling power consumption and the internal air cooling power consumption based on the temperature difference, the power consumption level of the external air cooling is determined based on the temperature difference. Specifically, the external air cooling power consumption and the internal air cooling power consumption levels corresponding to the temperature difference are determined based on a preset comparison table, and the external air cooling power consumption and the internal air cooling power consumption are determined.

[0119] In the specific implementation process, in the step of inputting the temperature distribution map of each plane in the ambient temperature model into a preset temperature calculation model and determining the ambient temperature through the preset temperature calculation model, the pixel value of each position in the temperature distribution map of each plane is used as the value of a position in the corresponding matrix to obtain a temperature distribution matrix for the temperature distribution map of each plane, and all the temperature distribution matrices are input into the preset temperature calculation model, and the temperature calculation model outputs the ambient temperature.

[0120] In the specific implementation process, the temperature calculation model adopts the LSTM model structure.

[0121] In the specific implementation process, the external air cooling power consumption is the power consumption of achieving air cooling by connecting with the outside world and introducing external air; the internal air cooling power consumption is the power consumption of achieving air cooling through air-conditioning equipment.

[0122] Specifically, when the outside temperature is higher than a preset outside threshold temperature, the operation of the external air cooling device is stopped.

[0123] In some embodiments of the present invention, in the step of determining the temperature of each device surface based on the device temperature model, the liquid cooling power consumption is calculated based on the temperature of each device and a preset device temperature threshold.

[0124] Specifically, the liquid cooling power consumption corresponding to the temperature of each device surface is determined by using a comparison table. The comparison table is provided with a temperature threshold, and each liquid cooling power consumption level is divided according to the temperature threshold.

[0125] By using the above solution, the present invention accurately predicts the temperature of the device and the temperature of the environment in which the device is located, and determines the corresponding power consumption through a preset comparison table, thereby ensuring the accuracy of the calculation of the power consumption.

[0126] In some embodiments of the present invention, in the step of calculating the single module power consumption value based on the external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption of the micro-module computer room, the external air cooling power consumption, internal air cooling power consumption and total liquid cooling power consumption are added to obtain the single module power consumption value.

[0127] In some embodiments of the present invention, in the steps of calculating the total module power consumption value based on the single module power consumption value and calculating the predicted PUE value based on the total module power consumption value, the predicted PUE value is calculated using the following formula:

[0128]

[0129] Among them, PUE pe represents the predicted PUE value, N represents the total number of micromodule rooms, and W n represents the power consumption of a single module in the micro-module room n, w ITIndicates the total power consumption of IT equipment, W PD Indicates the power consumption of the power distribution equipment, W + Indicates the power consumption of auxiliary equipment.

[0130] In some embodiments of the present invention, the following formula is used to calculate the power consumption value of the power distribution equipment and the power consumption value of the auxiliary equipment:

[0131]

[0132] Among them, δ1 represents the total proportion of the power consumption values ​​of each micro-module room, δ2 represents the proportion of the total power consumption of IT equipment, δ3 represents the proportion of the power consumption value of the power distribution equipment, and δ4 represents the proportion of the power consumption value of the auxiliary equipment.

[0133] In a specific implementation process, δ1+δ2=0.85, δ3=0.1, δ4=0.05.

[0134] By adopting the above scheme, this scheme can accurately predict the power consumption value of the micro-module computer room to achieve accurate prediction of the PUE value. It can perform real-time management and control through the real-time predicted PUE value, so that the PUE value of the data center is always kept below the PUE threshold. This scheme completes data prediction through data statistics, determines the predicted PUE value through the predicted data, realizes effective adjustment, and realizes effective control of the PUE value.

[0135] like Figure 5 As shown, another aspect of the present invention also relates to a green data center energy consumption monitoring and management method, the method comprising the following steps:

[0136] Step S100, based on the power consumption values ​​of IT equipment in the micro-module computer room at multiple historical time points and the temperatures of each temperature measurement point at multiple historical time points, obtain the power consumption value and the temperature of each temperature measurement point at a predicted time point;

[0137] Step S200, constructing an environmental temperature model based on the temperature of the environmental temperature measurement points among the temperature measurement points, and constructing an equipment temperature model based on the equipment temperature measurement points among the temperature measurement points;

[0138] Step S300, determining the ambient temperature based on the ambient temperature model, determining the air cooling requirement intensity based on the ambient temperature, and determining the external air cooling power consumption and the internal air cooling power consumption based on the air cooling requirement intensity;

[0139] Step S400: determining the temperature of each device surface based on the device temperature model, determining the liquid cooling power consumption of each device surface based on the temperature of each device surface, and determining the total liquid cooling power consumption based on the liquid cooling power consumption of each device surface;

[0140] Step S500, calculating the power consumption value of a single module based on the external air cooling power consumption, internal air cooling power consumption, and total liquid cooling power consumption of the micro-module computer room;

[0141] Step S600 , calculating the total module power consumption value based on the single module power consumption value, calculating the predicted PUE value based on the total module power consumption value, and determining whether power consumption abnormality occurs based on comparing the predicted PUE value with a preset PUE threshold.

[0142] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the aforementioned green data center energy consumption monitoring and management system. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, floppy disk, hard disk, removable storage disk, CD-ROM, or any other form of storage medium known in the art.

[0143] It should be understood by those skilled in the art that the various exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is specifically performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0144] It should be understood that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted. In the above embodiments, several specific steps are described and illustrated as examples. However, the method of the present invention is not limited to the specific steps described and illustrated. Those skilled in the art may make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present invention.

[0145] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or replace features of other embodiments.

[0146] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations to the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A green data center energy consumption monitoring and management system, characterized in that: The green data center is composed of multiple micro-module computer rooms, each of which is equipped with IT equipment. The system includes: The data prediction module obtains the power consumption value and temperature of each temperature measurement point at the predicted time point based on the power consumption value of IT equipment in the micro-module room at multiple historical time points and the temperature of each temperature measurement point at multiple historical time points; A model building module builds an ambient temperature model based on the temperature of the ambient temperature measuring points among the temperature measuring points, and builds an equipment temperature model based on the equipment temperature measuring points among the temperature measuring points; an air cooling power consumption determination module, which determines an ambient temperature based on the ambient temperature model, determines an air cooling requirement intensity based on the ambient temperature, and determines an external air cooling power consumption and an internal air cooling power consumption based on the air cooling requirement intensity; a liquid cooling power consumption determination module, which determines the temperature of each device surface based on the device temperature model, determines the liquid cooling power consumption of each device surface based on the temperature of each device surface, and determines the total liquid cooling power consumption based on the liquid cooling power consumption of each device surface; Single-module power consumption calculation module, which calculates the power consumption of a single module based on the external air cooling power consumption, internal air cooling power consumption, and total liquid cooling power consumption of the micro-module computer room; The power consumption abnormality determination module calculates the total module power consumption value based on the single module power consumption value, calculates the predicted PUE value based on the total module power consumption value, and determines whether power consumption abnormality occurs based on comparison between the predicted PUE value and a preset PUE threshold.

2. The green data center energy consumption monitoring and management system according to claim 1, characterized in that: The system also includes a feedback adjustment module. If power consumption abnormality occurs, the power consumption of a single module of each micro-module computer room at a preset number of time points is counted, and the maximum and minimum values ​​of the single module power consumption of each micro-module computer room at the preset number of time points are used as the upper limit and lower limit respectively. The upper limit of fluctuation is determined based on the upper limit, the lower limit of fluctuation is determined based on the lower limit, the fluctuation range is determined based on the upper limit and the lower limit, and whether the corresponding micro-module computer room needs to be checked is determined based on the fluctuation range.

3. The green data center energy consumption monitoring and management system according to claim 2, characterized in that: In the steps of determining the upper limit of fluctuation based on the upper limit value and determining the lower limit of fluctuation based on the lower limit value, the difference between the upper limit value and the power consumption of each single module in the power consumption of the single module at a previous preset number of time points is calculated, and the number of single modules whose power consumption is less than the fluctuation threshold value in the difference is counted as a first number, and the number of single modules whose power consumption is less than the fluctuation threshold value in the difference between the power consumption of the single module at a previous preset number of time points and the lower limit value is counted as a second number. The upper limit of fluctuation and the lower limit of fluctuation are calculated based on the initial fluctuation value, the first number, and the second number. The upper limit of fluctuation and the lower limit of fluctuation are then calculated based on the following formula: Among them, R T Indicates the upper limit of fluctuation, R L represents the lower limit of fluctuation, r represents the preset initial fluctuation value, Indicates the upper limit value, Indicates the lower limit, N1 indicates the first number, N2 indicates the second number, N t Indicates the number of time points before the preset number, N ρ Indicates the preset basic fluctuation quantity value.

4. The green data center energy consumption monitoring and management system according to claim 1, characterized in that: In the step of obtaining the power consumption value at a predicted time point and the temperature of each temperature measuring point based on the power consumption values ​​of IT equipment in a micro-module computer room at multiple historical time points and the temperatures of each temperature measuring point at multiple historical time points, the power consumption value at each historical time point and the temperature of the temperature measuring point are used as the value of a dimension in a statistical vector to construct a statistical vector, and the statistical vectors of multiple historical time points are constructed into a statistical matrix. The statistical matrix is ​​input into a trained first prediction model, and the first prediction model outputs a first prediction vector, which includes the power consumption value at the predicted time point and the temperature of each temperature measuring point.

5. The green data center energy consumption monitoring and management system according to claim 4, characterized in that: The model structure of the first prediction model adopts the model structure of a recurrent neural network model.

6. The green data center energy consumption monitoring and management system according to claim 4, characterized in that: In the step of constructing a statistical vector by taking the power consumption value and the temperature of each temperature measurement point at each historical time point as the value of a dimension in the statistical vector: The temperature of each temperature measuring point at multiple historical time points is used to construct a temperature change curve for each temperature measuring point, and a total fitting curve is obtained based on the temperature change curve of each temperature measuring point; The power consumption change curve of the power consumption value of the micro-module room at multiple historical time points; The power consumption variation curve and the total fitting curve are input into a hysteresis calculation model to obtain a hysteresis time step.

7. The green data center energy consumption monitoring and management system according to claim 6, characterized in that: In the step of constructing a temperature change curve for each temperature measuring point by using the temperatures of each temperature measuring point at multiple historical time points, and obtaining a total fitting curve based on the temperature change curve of each temperature measuring point, Gaussian process regression, weighted least squares method or cubic spline interpolation are used for fitting to obtain the total fitting curve.

8. The green data center energy consumption monitoring and management system according to claim 6, characterized in that: In the step of constructing a statistical vector by taking the power consumption value and the temperature of the temperature measurement point at each historical time point as the value of a dimension in the statistical vector, the temperature change time point corresponding to the time point of the power consumption value is determined based on the numerical value of the lag time step, and the power consumption value at the time point and the temperature of each temperature measurement point at the corresponding temperature change time point are constructed into a statistical vector.

9. The green data center energy consumption monitoring and management system according to claim 1, characterized in that: In the steps of building an environmental temperature model based on the temperature of the environmental temperature measuring points among the temperature measuring points, and building a device temperature model based on the device temperature measuring points among the temperature measuring points; Determine a temperature distribution map of a plane based on the predicted values ​​of the temperature measurement points of each plane set on the inner wall of the micro-module computer room among the environmental temperature measurement points, and construct an environmental temperature model based on the temperature distribution maps of all the inner walls; The temperature of each device is determined based on the temperature measuring points set on the outer surface of the IT device among the device temperature measuring points, and a three-dimensional framework pre-built for the device is rendered based on the temperature of the device to obtain a device temperature model.

10. The green data center energy consumption monitoring and management system according to claim 9, characterized in that: In the steps of determining the ambient temperature based on the ambient temperature model, determining the air cooling demand intensity based on the ambient temperature, and determining the external air cooling power consumption and the internal air cooling power consumption based on the air cooling demand intensity, the temperature distribution map of each plane in the ambient temperature model is input into a preset temperature calculation model, the ambient temperature is determined by the preset temperature calculation model, the temperature difference between the ambient temperature and a preset ambient temperature threshold is calculated, and the external air cooling power consumption and the internal air cooling power consumption are determined based on the temperature difference.

11. The green data center energy consumption monitoring and management system according to claim 10, characterized in that: In the step of inputting the temperature distribution map of each plane in the ambient temperature model into a preset temperature calculation model and determining the ambient temperature through the preset temperature calculation model, the pixel value of each position in the temperature distribution map of each plane is used as the value of a position in the corresponding matrix to obtain a temperature distribution matrix for the temperature distribution map of each plane, and all the temperature distribution matrices are input into the preset temperature calculation model, and the temperature calculation model outputs the ambient temperature.

12. The green data center energy consumption monitoring and management system according to claim 1, characterized in that: In the step of determining the temperature of each device surface based on the device temperature model, the liquid cooling power consumption is calculated based on the temperature of each device and a preset device temperature threshold.

13. The green data center energy consumption monitoring and management system according to claim 1, characterized in that: In the step of calculating the power consumption value of a single module based on the external air cooling power consumption, the internal air cooling power consumption, and the total liquid cooling power consumption of the micro-module computer room, the external air cooling power consumption, the internal air cooling power consumption, and the total liquid cooling power consumption are added together to obtain the power consumption value of the single module.

14. The green data center energy consumption monitoring and management system according to any one of claims 1 to 13, characterized in that: In the steps of calculating the total module power consumption value based on the single module power consumption value and calculating the predicted PUE value based on the total module power consumption value, the predicted PUE value is calculated using the following formula: Among them, PUE pe represents the predicted PUE value, N represents the total number of micromodule rooms, and W n represents the power consumption of a single module in the micro-module room n, w IT Indicates the total power consumption of IT equipment, W PD Indicates the power consumption of the power distribution equipment, W + Indicates the power consumption of auxiliary equipment.

15. The green data center energy consumption monitoring and management system according to claim 14, characterized in that: Use the following formula to calculate the power consumption of the power distribution equipment and auxiliary equipment; Among them, δ1 represents the total proportion of the power consumption values ​​of each micro-module room, δ2 represents the proportion of the total power consumption of IT equipment, δ3 represents the proportion of the power consumption value of the power distribution equipment, and δ4 represents the proportion of the power consumption value of the auxiliary equipment.

16. A green data center energy consumption monitoring and management method, characterized by: The steps of the method include: Based on the power consumption values ​​of IT equipment in the micro-module room at multiple historical time points and the temperatures of various temperature measurement points at multiple historical time points, the power consumption values ​​and temperatures of various temperature measurement points at the predicted time point are obtained; An ambient temperature model is constructed based on the temperature of the ambient temperature measurement points among the temperature measurement points, and an equipment temperature model is constructed based on the temperature of the equipment temperature measurement points among the temperature measurement points; determining an ambient temperature based on the ambient temperature model, determining an air cooling requirement intensity based on the ambient temperature, and determining an external air cooling power consumption and an internal air cooling power consumption based on the air cooling requirement intensity; determining the temperature of each device surface based on the device temperature model, determining the liquid cooling power consumption of each device surface based on the temperature of each device surface, and determining the total liquid cooling power consumption based on the liquid cooling power consumption of each device surface; Calculate the power consumption of a single module based on the external air cooling power consumption, internal air cooling power consumption, and total liquid cooling power consumption of the micro-module room; The total module power consumption value is calculated based on the single module power consumption value, the predicted PUE value is calculated based on the total module power consumption value, and whether power consumption abnormality occurs is determined based on the comparison between the predicted PUE value and a preset PUE threshold.

Images