A data center energy consumption prediction method and system based on DCIM

Abstract

The embodiment of the present application provides a data center energy consumption prediction method and system based on DCIM. Among them, the real-time operating parameters of the power monitoring, refrigeration equipment control and environmental sensing modules are synchronously obtained through the heterogeneous data bus, and the dynamic topology calibration mechanism is used to perform time-space alignment processing to generate a dynamic topology matrix, and a propagation path weight matrix is ​​constructed. According to the timestamp sequence of historical equipment start and stop events and the phase offset of the energy consumption fluctuation curve, the event trigger signal is injected into the propagation path weight matrix to form a multi-dimensional prediction map. Based on the transient peak distribution and real-time environmental parameter phase deviation in the prediction map, the path tracing mechanism is used to locate the key propagation path identifier, and the impedance coefficient and pressure loss threshold in the propagation path weight matrix are adjusted to generate the refrigeration equipment power gradient adjustment instruction and the power supply link impedance matching instruction. The technical solution provided by the embodiment of the present application effectively improves energy efficiency management and operational stability.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of DCIM technology, and in particular to a method and system for predicting energy consumption of a data center based on DCIM. Background Art

[0002] As data centers continue to expand, how to efficiently manage energy consumption has become a key issue. Data centers need to monitor and adjust their power usage, cooling equipment operation, and environmental parameters (such as temperature and humidity) in real time to ensure stable operation and reduce energy consumption. In addition, accurate prediction of equipment start-stop events and energy consumption fluctuations is crucial to optimizing operating costs.

[0003] Current data centers usually rely on traditional DCIM systems to collect data from power monitoring, cooling equipment control, and environmental sensor modules. These systems can provide basic data monitoring functions, but lack the ability to conduct in-depth analysis and dynamic adjustment. For example, they cannot accurately simulate energy propagation paths or predict future energy consumption fluctuations based on historical data.

[0004] Existing DCIM solutions are inefficient when processing heterogeneous data sources, and it is difficult to achieve spatiotemporal alignment, resulting in inaccurate models. In addition, traditional methods lack support for virtual twin models and cannot effectively superimpose dynamic weight coefficients of energy propagation paths, limiting the accurate calculation of thermodynamic propagation delays of equipment state change events. This leads to inadequate optimization of energy consumption management strategies, affecting the overall energy efficiency and stability of data centers. Summary of the invention

[0005] The embodiments of the present application provide a method and system for predicting energy consumption of a data center based on DCIM, so as to solve the problems of inaccurate data processing and insufficient optimization of energy consumption management strategies in the prior art.

[0006] In a first aspect, an embodiment of the present application provides a method for predicting energy consumption of a data center based on DCIM, comprising:

[0007] The real-time operating parameters of the power monitoring module, refrigeration equipment control module and environmental sensor module in the DCIM system are synchronously obtained through the heterogeneous data bus. The dynamic topology calibration mechanism is used to perform spatiotemporal alignment processing on the equipment operation status data and the room temperature and humidity distribution data to generate a dynamic topology matrix.

[0008] Based on the topological matching degree between the dynamic topological matrix and the virtual twin model, the dynamic weight coefficient of the energy propagation path is superimposed on the physical connection relationship of the device to construct a propagation path weight matrix;

[0009] According to the phase offset between the timestamp sequence of the equipment start and stop events in the historical operation cycle and the energy consumption fluctuation curve, an event trigger signal is injected into the propagation path weight matrix, and the thermodynamic propagation delay of the equipment state change event is calculated through the path superposition effect to generate a multi-dimensional prediction map;

[0010] Based on the phase deviation between the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, the key propagation path identifier is located through the path tracing mechanism, the impedance coefficient and the pressure loss threshold in the propagation path weight matrix are dynamically adjusted, and the power gradient adjustment instructions of the refrigeration equipment and the power supply link impedance matching instructions are generated.

[0011] Optionally, according to the phase offset between the timestamp sequence of the equipment start and stop events in the historical operation cycle and the energy consumption fluctuation curve, an event trigger signal is injected into the propagation path weight matrix, and the thermodynamic propagation delay of the equipment state change event is calculated through the path superposition effect to generate a multi-dimensional prediction map, including:

[0012] The dynamic topology calibration engine performs nonlinear coupling analysis on the equipment power fluctuation coefficient and the refrigerant flow gradient, and generates a dynamic topology matrix by combining the cabinet inlet air temperature gradient distribution and the air conditioner return air temperature phase difference.

[0013] Based on the impedance coefficient and pressure loss threshold of the propagation path weight matrix, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage parameters are integrated to build a thermodynamic disturbance propagation simulation model for equipment state change events.

[0014] According to the phase deviation of the transient peak distribution in the prediction map, the key propagation path identifier is located and the impedance correction coefficient is calculated through the path tracing engine, and the chilled water flow fluctuation parameters and the cooling water pump output percentage are synchronously integrated for dynamic compensation;

[0015] Combined with the dynamic matching relationship between real-time environmental parameters and impedance correction coefficients, a closed-loop feedback mechanism is used to collaboratively optimize the chilled water valve opening threshold and the power supply link impedance parameters to generate the power gradient adjustment instructions for the refrigeration equipment and the power supply link reconstruction strategy.

[0016] Optionally, based on the impedance coefficient and pressure loss threshold of the propagation path weight matrix, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage parameter are integrated to construct a thermodynamic disturbance propagation simulation model of the equipment state change event, including:

[0017] The dynamic fluid dynamics engine is used to synchronize the pressure loss gradient of the chilled water circulation path with the cooling tower fan output percentage, and the chilled water return temperature gradient distribution and the power supply link harmonic distortion rate parameters are integrated to generate a fluid dynamics topology matrix.

[0018] Based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient are superimposed to generate an impedance matching map with time series markings.

[0019] According to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid mechanics topology matrix and the impedance matching map are synchronously integrated for cross-validation;

[0020] The fluid mechanics topology matrix and the power supply link impedance parameters are integrated, and the dynamic adjustment curve of the chilled water valve opening threshold and the power supply link harmonic suppression parameters are combined to generate a collaborative control strategy.

[0021] Optionally, according to the phase deviation of the transient peak distribution in the prediction spectrum, the key propagation path identifier is located and the impedance correction coefficient is calculated through the path tracing engine, and the chilled water flow fluctuation parameter and the cooling water pump output percentage are synchronously integrated for dynamic compensation, including:

[0022] The dynamic fluid dynamics engine is used to synchronize the pressure loss gradient of the chilled water circulation path with the cooling tower fan output percentage, and the power supply link harmonic distortion rate parameters are integrated to generate a fluid dynamics topology matrix.

[0023] Based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, the dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and an impedance matching map with time series marking is generated;

[0024] According to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction factor is calculated through the path tracing engine and the impedance matching map is updated synchronously;

[0025] The fluid mechanics topology matrix and the updated impedance matching map are integrated, and the dynamic adjustment curve of the chilled water valve opening threshold is combined to generate the power gradient adjustment instructions and power supply link reconstruction strategy of the refrigeration equipment.

[0026] Optionally, based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient are superimposed to generate an impedance matching map with a time series mark, including:

[0027] The chilled water circulation system is modeled through a dynamic fluid dynamics engine, and the chilled water flow fluctuation parameters, cooling water pump output percentage and its correction coefficient are integrated to generate a fluid dynamics topology matrix;

[0028] Based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, the dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is analyzed, and the influence of the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient is considered.

[0029] Combined with the temperature gradient distribution and pressure loss gradient parameters in the environmental parameters, the cooling water pump output percentage correction factor is calculated, and the impedance matching map is updated synchronously;

[0030] The fluid mechanics topology matrix and the updated impedance matching map are integrated, and the dynamic adjustment curve of the chilled water valve opening threshold is combined to generate the refrigeration equipment power gradient adjustment instruction and the power supply link reconstruction strategy.

[0031] Optionally, according to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid mechanics topology matrix and the impedance matching map are synchronously integrated for cross-validation, including:

[0032] Collect temperature gradient distribution and pressure loss gradient parameters from each monitoring point in the data center in real time, process these data using specific algorithms, and identify key features of system state changes;

[0033] Based on the above data analysis results, the correction factor of the cooling water pump output percentage is calculated through a closed-loop feedback mechanism, taking into account the temperature and pressure loss variation patterns in different time periods and their impact on cooling efficiency;

[0034] Combine the calculated cooling water pump output percentage correction factor with the dynamic compensation factor in the fluid mechanics topology matrix to cross-validate the impedance matching map to ensure data consistency and continuity;

[0035] Develop a collaborative control strategy, including a dynamic adjustment curve for the chilled water valve opening threshold and optimal setting of the power supply link harmonic suppression parameters, to ensure the optimal operating state of the refrigeration equipment and power supply system.

[0036] Optionally, according to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through the path tracing engine and the impedance matching map is updated synchronously, including:

[0037] Collect temperature gradient distribution and pressure loss gradient parameters from each monitoring point in the data center in real time, process these data using advanced data analysis algorithms, extract key features reflecting the system status, and convert them into intermediate parameters that can be used in subsequent steps;

[0038] Based on the above data analysis results, the key propagation path identifiers are identified through the path tracing engine to predict future environmental change trends, and the working status of the cooling water pump is adjusted accordingly to calculate the correction coefficient of the cooling water pump output percentage;

[0039] Combine the calculated cooling water pump output percentage correction factor with other relevant parameters to synchronously update the impedance matching map to ensure coordinated operation among all components and optimize overall system performance;

[0040] Combining the results of all steps, formulate or optimize the collaborative control strategy, including the dynamic adjustment curve of the chilled water valve opening threshold, the setting of the power supply link harmonic suppression parameters and other necessary operating instructions, to ensure that the system can respond to changes in real-time environmental parameters and is forward-looking and flexible.

[0041] In a second aspect, an embodiment of the present application provides a data center energy consumption prediction system based on DCIM, including:

[0042] The acquisition module is used to synchronously acquire the real-time operating parameters of the power monitoring module, refrigeration equipment control module and environmental sensor module in the DCIM system through the heterogeneous data bus, and use the dynamic topology calibration mechanism to perform spatiotemporal alignment processing on the equipment operation status data and the room temperature and humidity distribution data to generate a dynamic topology matrix;

[0043] A construction module, used to superimpose the dynamic weight coefficient of the energy propagation path on the physical connection relationship of the device based on the topological matching degree between the dynamic topology matrix and the virtual twin model, and construct a propagation path weight matrix;

[0044] A calculation module is used to inject an event trigger signal into the propagation path weight matrix according to the timestamp sequence of the equipment start and stop events in the historical operation cycle and the phase offset of the energy consumption fluctuation curve, calculate the thermodynamic propagation delay of the equipment state change event through the path superposition effect, and generate a multi-dimensional prediction map;

[0045] The adjustment module is used to locate the key propagation path identifier through the path tracing mechanism based on the phase deviation between the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, dynamically adjust the impedance coefficient and the pressure loss threshold in the propagation path weight matrix, and generate the power gradient adjustment instruction of the refrigeration equipment and the impedance matching instruction of the power supply link.

[0046] In a third aspect, an embodiment of the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a DCIM-based data center energy consumption prediction method as described in the first aspect above.

[0047] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, a DCIM-based data center energy consumption prediction method as described in the first aspect is implemented.

[0048] In an embodiment of the present application, real-time operating parameters of a power monitoring module, a refrigeration equipment control module, and an environmental sensing module in a DCIM system are synchronously acquired through a heterogeneous data bus, and a dynamic topology calibration mechanism is used to perform spatiotemporal alignment processing on equipment operating status data and room temperature and humidity distribution data to generate a dynamic topology matrix; based on the topological matching degree between the dynamic topology matrix and the virtual twin model, the dynamic weight coefficient of the energy propagation path is superimposed on the physical connection relationship of the equipment to construct a propagation path weight matrix; according to the phase offset of the timestamp sequence of equipment start and stop events and the energy consumption fluctuation curve in the historical operating cycle, an event trigger signal is injected into the propagation path weight matrix, and the thermodynamic propagation delay of the equipment state change event is calculated through the path superposition effect to generate a multi-dimensional prediction map; based on the phase deviation of the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, the key propagation path identifier is located through the path tracing mechanism, the impedance coefficient and the pressure loss threshold in the propagation path weight matrix are dynamically adjusted, and the power gradient adjustment instruction of the refrigeration equipment and the impedance matching instruction of the power supply link are generated.

[0049] The technical solution of this application has the following beneficial effects:

[0050] The real-time operating parameters of power monitoring, refrigeration equipment control and environmental sensing modules are synchronously obtained through heterogeneous data buses, and the dynamic topology calibration mechanism is used for spatiotemporal alignment. This method ensures that data from different sources can be integrated accurately and improves the accuracy of subsequent analysis; based on the generated dynamic topology matrix and the topological matching degree of the virtual twin model, the dynamic weight coefficient of the energy propagation path is superimposed on the physical connection relationship of the equipment to construct the propagation path weight matrix. This helps to more accurately simulate the energy flow inside the data center, thereby providing support for the formulation of more effective energy management strategies; using the timestamp sequence of equipment start and stop events in the historical operation cycle and the phase offset of the energy consumption fluctuation curve, the event trigger signal is injected into the propagation path weight matrix, the thermodynamic propagation delay of the equipment state change event is calculated, and a multi-dimensional prediction map is generated. This process enables the system to foresee potential problems and take measures in advance to deal with them, thereby improving the stability and efficiency of the system; based on the phase deviation between the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, the key propagation path identifier is located through the path tracing mechanism, the impedance coefficient and the pressure loss threshold in the propagation path weight matrix are dynamically adjusted, and the power gradient adjustment instruction of the refrigeration equipment and the impedance matching instruction of the power supply link are generated. This method realizes intelligent perception and automatic adjustment of the data center environment, further improving energy efficiency and reducing operating costs.

[0051] Furthermore, the dynamic topology calibration engine is used to perform nonlinear coupling analysis on the equipment power fluctuation coefficient and the refrigerant flow rate gradient, and the dynamic topology matrix is ​​generated by combining the cabinet inlet air temperature gradient distribution and the air conditioner return air temperature phase difference. On this basis, based on the impedance coefficient and pressure loss threshold of the propagation path weight matrix, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage parameters are integrated to construct a thermodynamic perturbation propagation simulation model of the equipment state change event. According to the phase deviation of the transient peak distribution in the prediction spectrum, the path tracing engine is used to locate the key propagation path identifier and calculate the impedance correction coefficient, and the chilled water flow fluctuation parameters and the cooling water pump output percentage are synchronously integrated for dynamic compensation. Finally, combined with the dynamic matching relationship between real-time environmental parameters and the impedance correction coefficient, a closed-loop feedback mechanism is used to coordinately optimize the chilled water valve opening threshold and the power supply link impedance parameters, and generate the refrigeration equipment power gradient adjustment instructions and power supply link reconstruction strategy. This method, through in-depth analysis of the timestamp sequence of equipment start and stop events and the phase offset of the energy consumption fluctuation curve, combined with advanced dynamic topology calibration engine and nonlinear coupling analysis technology, accurately simulates the thermodynamic propagation delay of equipment state change events, and generates a multi-dimensional prediction map, achieving accurate prediction and intelligent regulation of energy flow and cooling demand in the data center. Using the path tracing engine and closed-loop feedback mechanism, the system can dynamically adjust the chilled water valve opening threshold and power supply link impedance parameters to ensure efficient energy use and optimized operating costs, significantly improving the overall energy efficiency and stability of the data center.

[0052] Furthermore, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage are processed in phase synchronization through the dynamic fluid dynamics engine, and the chilled water return temperature gradient distribution and the power supply link harmonic distortion rate parameters are integrated to generate a fluid dynamics topology matrix. On this basis, based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, the dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient are superimposed to generate an impedance matching map with a time series mark. According to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated using a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid dynamics topology matrix is ​​synchronously integrated with the impedance matching map for cross-validation. Finally, the fluid dynamics topology matrix and the power supply link impedance parameters are combined, and the dynamic adjustment curve of the chilled water valve opening threshold and the power supply link harmonic suppression parameters are used to generate a collaborative control strategy. This method accurately simulates the influence of the chilled water circulation path and cooling tower fan operation on the system's thermodynamic performance by introducing a dynamic fluid dynamics engine and a thermodynamic disturbance propagation simulation model. It also optimizes the configuration of the power supply link impedance parameters and the chilled water valve opening threshold by generating a fluid dynamics topology matrix and an impedance matching map with timing markers. Cross-validation using a closed-loop feedback mechanism and a dynamic compensation factor ensures accurate regulation of the cooling water pump output and chilled water flow fluctuation parameters. In addition, through a collaborative control strategy, the dynamic adjustment of the chilled water valve opening threshold and the power supply link harmonic suppression are effectively combined, significantly improving the stability and energy efficiency of the data center cooling system while reducing operating costs. This method not only improves energy efficiency, but also enhances the system's response speed and stability, providing a more intelligent and efficient operation mode for data centers.

[0053] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0055] Figure 1 A flow chart of a DCIM-based data center energy consumption prediction method provided by the present application is shown;

[0056] Figure 2 A schematic diagram of the structure of a data center energy consumption prediction system based on DCIM provided by the present application is shown;

[0057] Figure 3 A schematic diagram of the structure of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0058] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0059] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0060] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0061] Figure 1 A flow chart of a data center energy consumption prediction method based on DCIM is provided for an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0062] 101. The real-time operating parameters of the power monitoring module, refrigeration equipment control module and environmental sensor module in the DCIM system are synchronously obtained through the heterogeneous data bus, and the dynamic topology calibration mechanism is used to perform spatiotemporal alignment processing on the equipment operation status data and the temperature and humidity distribution data of the computer room to generate a dynamic topology matrix;

[0063] In this step, the real-time operating parameters of the power monitoring module, cooling equipment control module and environmental sensing module in the DCIM system are synchronously obtained through the heterogeneous data bus. These parameters include but are not limited to power consumption, cooling equipment status, and temperature and humidity distribution in the computer room, which are used to comprehensively monitor the internal environmental conditions of the data center and their changing trends.

[0064] In the embodiment of this application, the heterogeneous data bus technology is first used to collect real-time operating parameters from different sources (such as power monitoring, refrigeration equipment control and environmental sensing modules), and the dynamic topology calibration mechanism is used to align these data in time and space to ensure the consistency and accuracy of the data. This process ultimately generates a dynamic topology matrix that reflects the relationship between the equipment operating status and the temperature and humidity distribution, laying the foundation for subsequent analysis.

[0065] Suppose a data center deploys multiple sensors to monitor the temperature, humidity, and energy consumption of each area. Through the heterogeneous data bus, the system can collect this information in real time and use the dynamic topology calibration mechanism to adjust the timestamp and spatial position of the data to accurately reflect the actual working status and environmental conditions of each device, thereby generating a detailed dynamic topology matrix for subsequent analysis.

[0066] 102. Based on the topological matching degree between the dynamic topological matrix and the virtual twin model, the dynamic weight coefficient of the energy propagation path is superimposed on the physical connection relationship of the device to construct a propagation path weight matrix;

[0067] In this step, based on the topological matching degree of the previously generated dynamic topology matrix and the virtual twin model, the dynamic weight coefficient of the energy propagation path is superimposed on the physical connection relationship of the device. This includes the direction and intensity of energy flow determined according to the actual physical layout and simulation results, which is used to construct a propagation path weight matrix that accurately reflects how energy is transmitted within the system.

[0068] In the embodiment of the present application, the dynamic topology matrix is ​​first compared and analyzed with the virtual twin model to identify the physical connection relationship between the devices, and on this basis, the dynamic weight coefficient of the energy propagation path representing the energy propagation efficiency is added. This process helps to create a detailed propagation path weight matrix, which not only describes how energy moves within the system, but also takes into account the impact of various factors on energy propagation.

[0069] Suppose we have a data center model that includes all devices and how they are connected. By analyzing this model and comparing it with actual operational data, we can assign corresponding energy propagation weight coefficients to each connection, thus forming a detailed propagation path weight matrix. This allows us to more accurately predict how energy flows in the system and optimize resource allocation.

[0070] 103. According to the phase offset of the timestamp sequence of the equipment start and stop events in the historical operation cycle and the energy consumption fluctuation curve, an event trigger signal is injected into the propagation path weight matrix, and the thermodynamic propagation delay of the equipment state change event is calculated through the path superposition effect to generate a multi-dimensional prediction map;

[0071] In this step, event trigger signals are injected into the propagation path weight matrix according to the phase offset between the timestamp sequence of the equipment start and stop events in the historical operation cycle and the energy consumption fluctuation curve. These signals reflect the impact of equipment start or stop on the system energy distribution, and are used to calculate the thermodynamic propagation delay caused by equipment state change events and generate a multi-dimensional prediction map.

[0072] In the embodiment of the present application, by analyzing the equipment start and stop time and energy consumption fluctuations in the historical data, combined with the propagation path weight matrix, the specific impact of each state change on the system thermodynamic performance is evaluated. This method allows us to predict future energy consumption patterns through the path superposition effect, generate a multi-dimensional prediction map, and help formulate more effective management strategies.

[0073] Suppose we have recorded the start and stop times of all devices in the data center and the corresponding energy consumption changes over the past few months. By analyzing this data and applying it to the propagation path weight matrix, we can predict how the status change of a certain device will affect the overall energy consumption in the future, so that we can prepare in advance.

[0074] 104. Based on the phase deviation between the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, the key propagation path identifier is located through the path tracing mechanism, the impedance coefficient and the pressure loss threshold in the propagation path weight matrix are dynamically adjusted, and the power gradient adjustment instruction of the refrigeration equipment and the power supply link impedance matching instruction are generated.

[0075] In this step, based on the phase deviation between the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, the key propagation path identifier is located through the path tracing mechanism. This involves identifying those energy propagation paths that have a significant impact on system performance, and dynamically adjusting the impedance coefficient and pressure loss threshold in the propagation path weight matrix accordingly, generating the power gradient adjustment instructions for the refrigeration equipment and the impedance matching instructions for the power supply link.

[0076] In the embodiment of the present application, the key paths that may cause problems are first identified based on the prediction map, and then the relevant parameters are adjusted according to the characteristics of these paths to optimize the operating efficiency of the entire system. This method ensures that the optimal operating state can be maintained even when environmental conditions change.

[0077] Assume that our prediction map shows that certain specific paths may become bottlenecks in the future. Through the path tracing mechanism, we can adjust the impedance coefficient and voltage loss threshold on these paths in a targeted manner to ensure that the system is always in the optimal working state and avoid potential problems.

[0078] In summary, steps 101 to 104 cover the complete process from data collection, model building, predictive analysis to dynamic adjustment, aiming to provide a comprehensive data center energy management system to meet the operational needs of high efficiency and low energy consumption while ensuring the stability and reliability of the system.

[0079] In order to further improve the accuracy and efficiency of data center energy consumption prediction, in some embodiments, in step 103, according to the phase offset between the timestamp sequence of equipment start and stop events in the historical operation cycle and the energy consumption fluctuation curve, an event trigger signal is injected into the propagation path weight matrix, and the thermodynamic propagation delay of the equipment state change event is calculated through the path superposition effect to generate a multi-dimensional prediction map, including:

[0080] The dynamic topology calibration engine is used to perform nonlinear coupling analysis on the equipment power fluctuation coefficient and the refrigerant flow rate gradient, and the dynamic topology matrix is ​​generated by combining the cabinet inlet air temperature gradient distribution and the air conditioner return air temperature phase difference. Based on the impedance coefficient and pressure loss threshold of the propagation path weight matrix, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage parameters are integrated to build a thermodynamic disturbance propagation simulation model of the equipment state change event. According to the phase deviation of the transient peak distribution in the prediction map, the path tracing engine is used to locate the key propagation path identifier and calculate the impedance correction coefficient, and the chilled water flow fluctuation parameters and the cooling water pump output percentage are synchronously integrated for dynamic compensation. Combined with the dynamic matching relationship between real-time environmental parameters and the impedance correction coefficient, a closed-loop feedback mechanism is used to coordinately optimize the chilled water valve opening threshold and the power supply link impedance parameters to generate the refrigeration equipment power gradient adjustment instructions and power supply link reconstruction strategy.

[0081] In this embodiment, this method uses a dynamic topology calibration engine to process equipment power fluctuations and refrigerant flow gradient data, and generates a dynamic topology matrix in combination with cabinet inlet air temperature and air conditioner return air temperature information. In addition, it also uses the impedance coefficient and pressure loss threshold in the propagation path weight matrix, combined with the chilled water circulation path and cooling tower fan parameters to construct a thermodynamic disturbance propagation simulation model to simulate the impact of equipment state changes on the system. Finally, the critical path is identified through the path tracing engine, and the chilled water valve and power supply link parameters are adjusted through a closed-loop feedback mechanism to ensure efficient operation of the system.

[0082] In the embodiment of the present application, a dynamic topology calibration engine is first used to perform nonlinear coupling analysis on equipment power fluctuations and refrigerant flow rates; secondly, based on the impedance coefficient and pressure loss threshold in the propagation path weight matrix, a simulation model is constructed in combination with the chilled water circulation path and cooling tower fan parameters; thirdly, according to the transient peak distribution in the prediction map, the impedance correction coefficient is calculated through the path tracing engine, and the chilled water flow fluctuation parameters are synchronously integrated for dynamic compensation; finally, in combination with real-time environmental parameters, a closed-loop feedback mechanism is used to collaboratively optimize the chilled water valve opening and power supply link parameters to generate adjustment instructions.

[0083] Here is a specific example:

[0084] Suppose a large data center wants to improve energy efficiency by optimizing its energy management system. First, the data center uses the dynamic topology calibration engine to analyze the equipment power fluctuations and refrigerant flow rate data in the past three months, and generates a dynamic topology matrix based on the cabinet inlet air temperature and air conditioning return air temperature information; secondly, based on the generated dynamic topology matrix, the data center evaluates the energy transmission efficiency between different devices, and builds a thermodynamic perturbation propagation simulation model based on the chilled water circulation path and cooling tower fan parameters; thirdly, based on the simulation results, the data center uses the path tracing engine to identify the critical paths that may cause performance bottlenecks, and calculates the corresponding impedance correction coefficients, while integrating the chilled water flow fluctuation parameters for dynamic compensation; finally, the data center combines the current environmental conditions and adjusts the chilled water valve opening and power supply link parameters through a closed-loop feedback mechanism to ensure that the system is always in the optimal operating state. Through the above steps, the data center not only improves energy efficiency, but also enhances the stability and response speed of the system.

[0085] In order to further improve the accuracy and response speed of data center energy consumption prediction, in some embodiments, based on the impedance coefficient and pressure loss threshold of the propagation path weight matrix, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage parameter are integrated to construct a thermodynamic disturbance propagation simulation model of the equipment state change event, including:

[0086] The pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage are phase-synchronized through the dynamic fluid mechanics engine, and the chilled water return temperature gradient distribution and the power supply link harmonic distortion rate parameters are integrated to generate a fluid mechanics topology matrix; based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, the dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient are superimposed to generate an impedance matching map with a time series mark; according to the temperature gradient distribution and the pressure loss gradient parameter in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid mechanics topology matrix and the impedance matching map are synchronously integrated for cross-validation; the fluid mechanics topology matrix and the power supply link impedance parameters are integrated, and the dynamic adjustment curve of the chilled water valve opening threshold and the power supply link harmonic suppression parameters are combined to generate a collaborative control strategy.

[0087] In this embodiment, the method uses a dynamic fluid dynamics engine to process the operating data of the chilled water circulation path and the cooling tower fan, and generates a fluid dynamics topology matrix in combination with the chilled water return temperature and the power supply link harmonic distortion rate. In addition, it also uses the propagation delay coefficient to construct a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold, and generates an impedance matching map with a timing mark. Finally, the cooling water pump output and the chilled water valve opening are optimized through a closed-loop feedback mechanism to ensure the efficient operation of the system.

[0088] In an embodiment of the present application, firstly, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage are analyzed through a dynamic fluid mechanics engine to generate a fluid mechanics topology matrix; secondly, based on the propagation delay coefficient, a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed; thirdly, according to the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid mechanics topology matrix and the impedance matching map are synchronously integrated for cross-validation; finally, the fluid mechanics topology matrix and the power supply link impedance parameters are combined to generate a collaborative control strategy to optimize the chilled water valve opening and power supply link settings.

[0089] Here is a specific example:

[0090] Suppose a data center wants to improve energy efficiency by optimizing its cooling system. First, the data center uses a dynamic fluid mechanics engine to analyze the pressure loss gradient of the chilled water circulation path and the output percentage of the cooling tower fan, and generates a fluid mechanics topology matrix in combination with the chilled water return temperature and the harmonic distortion rate of the power supply link; secondly, based on the propagation delay coefficient, the data center constructs a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold; thirdly, according to the temperature gradient and pressure loss gradient in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid mechanics topology matrix is ​​synchronously integrated with the impedance matching map for cross-validation; finally, the data center combines the fluid mechanics topology matrix with the power supply link impedance parameters to generate a collaborative control strategy to optimize the chilled water valve opening and power supply link settings. Through the above steps, the data center not only improves energy efficiency, but also enhances the stability and response speed of the system, ensuring the best operating state.

[0091] In order to further improve the accuracy of data center energy consumption prediction and the response speed of the system, in some embodiments, according to the phase deviation of the transient peak distribution in the prediction map, the key propagation path identifier is located and the impedance correction coefficient is calculated through the path tracing engine, and the chilled water flow fluctuation parameter and the cooling water pump output percentage are synchronously integrated for dynamic compensation, including:

[0092] The pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage are phase-synchronized through the dynamic fluid mechanics engine, and the power supply link harmonic distortion rate parameters are integrated to generate a fluid mechanics topology matrix. Based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, the dynamic matching relationship between the power supply link impedance parameters and the chilled water valve opening threshold is constructed, and an impedance matching map with timing marks is generated. According to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through the path tracing engine and the impedance matching map is updated synchronously. The fluid mechanics topology matrix and the updated impedance matching map are integrated, and the power gradient adjustment instructions of the refrigeration equipment and the power supply link reconstruction strategy are generated in combination with the dynamic adjustment curve of the chilled water valve opening threshold.

[0093] In this embodiment, the method uses a dynamic fluid dynamics engine to process the operating data of the chilled water circulation path and the cooling tower fan, and generates a fluid dynamics topology matrix in combination with the power supply link harmonic distortion rate. In addition, it also uses the propagation delay coefficient to construct a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold, and generates an impedance matching map with a timing mark. Finally, the cooling water pump output percentage correction coefficient is calculated through the path tracing engine, and the impedance matching map is synchronously updated to optimize the chilled water valve opening and power supply link settings.

[0094] In an embodiment of the present application, firstly, the pressure loss gradient of the chilled water circulation path and the output percentage of the cooling tower fan are analyzed through a dynamic fluid mechanics engine, and a fluid mechanics topology matrix is ​​generated in combination with the harmonic distortion rate of the power supply link; secondly, based on the propagation delay coefficient, a dynamic matching relationship between the impedance parameters of the power supply link and the chilled water valve opening threshold is constructed and an impedance matching map with timing marks is generated; thirdly, according to the temperature gradient and pressure loss gradient in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through the path tracing engine, and the impedance matching map is updated synchronously; finally, the fluid mechanics topology matrix and the updated impedance matching map are integrated, and the power gradient adjustment instruction of the refrigeration equipment and the power supply link reconstruction strategy are generated in combination with the dynamic adjustment curve of the chilled water valve opening threshold.

[0095] Here is a specific example:

[0096] Suppose a data center wants to improve energy efficiency by optimizing its cooling system. First, the data center uses a dynamic fluid dynamics engine to analyze the pressure loss gradient of the chilled water circulation path and the output percentage of the cooling tower fan, and generates a fluid dynamics topology matrix in combination with the harmonic distortion rate of the power supply link; secondly, based on the propagation delay coefficient, the data center constructs a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold, and generates an impedance matching map with a timing mark; thirdly, according to the temperature gradient and pressure loss gradient in the real-time environmental parameters, the path tracing engine calculates the cooling water pump output percentage correction coefficient, and synchronously updates the impedance matching map; finally, the data center integrates the fluid dynamics topology matrix with the updated impedance matching map, and generates the refrigeration equipment power gradient adjustment instructions and power supply link reconstruction strategy in combination with the dynamic adjustment curve of the chilled water valve opening threshold. Through the above steps, the data center not only improves the energy efficiency, but also enhances the stability and response speed of the system, ensuring the best operating state.

[0097] In order to further improve the accuracy of data center energy consumption prediction and optimize the response speed of the cooling system, in some embodiments, based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient are superimposed to generate an impedance matching map with a time series mark, including:

[0098] The chilled water circulation system is modeled through a dynamic fluid mechanics engine, and the chilled water flow fluctuation parameters, the cooling water pump output percentage and its correction coefficient are integrated to generate a fluid mechanics topology matrix; based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, the dynamic matching relationship between the power supply link impedance parameters and the chilled water valve opening threshold is analyzed, and the influence of the chilled water flow fluctuation parameters and the cooling water pump output percentage correction coefficient is considered; combined with the temperature gradient distribution and the pressure loss gradient parameters in the environmental parameters, the cooling water pump output percentage correction coefficient is calculated, and the impedance matching map is updated synchronously; the fluid mechanics topology matrix and the updated impedance matching map are integrated, and the power gradient adjustment instructions of the refrigeration equipment and the power supply link reconstruction strategy are generated in combination with the dynamic adjustment curve of the chilled water valve opening threshold.

[0099] In this embodiment, the method uses a dynamic fluid dynamics engine to model the chilled water circulation path in detail, including chilled water flow fluctuation parameters, cooling water pump output percentage and its correction coefficient, to generate a fluid dynamics topology matrix. In addition, it also uses the propagation delay coefficient to analyze the dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold, and considers various influencing factors to generate an impedance matching map with timing marks. Finally, the impedance matching map is updated through a closed-loop feedback mechanism to ensure the efficient operation of the system.

[0100] In an embodiment of the present application, the chilled water circulation system is firstly modeled by a dynamic fluid mechanics engine, and the chilled water flow fluctuation parameters, the cooling water pump output percentage and its correction coefficient are integrated to generate a fluid mechanics topology matrix; secondly, based on the propagation delay coefficient, the dynamic matching relationship between the power supply link impedance parameters and the chilled water valve opening threshold is analyzed, and the influence of the chilled water flow fluctuation parameters and the cooling water pump output percentage correction coefficient is considered; thirdly, according to the temperature gradient distribution and the pressure loss gradient parameters in the environmental parameters, the cooling water pump output percentage correction coefficient is calculated, and the impedance matching map is updated synchronously; finally, the fluid mechanics topology matrix and the updated impedance matching map are integrated, and the power gradient adjustment instructions and power supply link reconstruction strategies of the refrigeration equipment are generated in combination with the dynamic adjustment curve of the chilled water valve opening threshold.

[0101] Here is a specific example:

[0102] Suppose a large data center wants to improve energy efficiency by optimizing its cooling system. First, the data center uses a dynamic fluid dynamics engine to model the chilled water circulation system, integrates the chilled water flow fluctuation parameters, the cooling water pump output percentage and its correction coefficient, and generates a fluid dynamics topology matrix; secondly, based on the propagation delay coefficient, the data center analyzes the dynamic matching relationship between the power supply link impedance parameters and the chilled water valve opening threshold, and considers the influence of the chilled water flow fluctuation parameters and the cooling water pump output percentage correction coefficient; thirdly, according to the temperature gradient distribution and pressure loss gradient parameters in the environmental parameters, the data center calculates the cooling water pump output percentage correction coefficient and updates the impedance matching map synchronously; finally, the data center integrates the fluid dynamics topology matrix and the updated impedance matching map, and combines the dynamic adjustment curve of the chilled water valve opening threshold to generate the refrigeration equipment power gradient adjustment instructions and power supply link reconstruction strategy. Through the above steps, the data center not only improves the energy efficiency, but also enhances the stability and response speed of the system, ensuring the best operating state.

[0103] In order to further improve the accuracy of data center energy consumption prediction and optimize the response speed of the cooling system, in some embodiments, according to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid mechanics topology matrix and the impedance matching map are synchronously integrated for cross-validation, including:

[0104] The temperature gradient distribution and pressure loss gradient parameters are collected in real time from each monitoring point in the data center, and these data are processed using a specific algorithm to identify the key features of system state changes. Based on the above data analysis results, the correction factor of the cooling water pump output percentage is calculated through a closed-loop feedback mechanism, taking into account the temperature and pressure loss change patterns in different time periods and their impact on cooling efficiency. The calculated cooling water pump output percentage correction factor is combined with the dynamic compensation factor in the fluid mechanics topology matrix, and the impedance matching map is cross-validated to ensure data consistency and continuity. A collaborative control strategy is formulated, including a dynamic adjustment curve for the chilled water valve opening threshold and optimal setting of the power supply link harmonic suppression parameters to ensure the optimal operating state of the refrigeration equipment and power supply system.

[0105] In this embodiment, the method involves collecting temperature gradient distribution and pressure loss gradient parameters from various monitoring points in the data center in real time, and using a specific algorithm to process these data to identify key features of system state changes. In addition, it also uses a closed-loop feedback mechanism to calculate the correction factor of the cooling water pump output percentage, and combines the dynamic compensation factor in the fluid mechanics topology matrix to cross-validate the impedance matching map to ensure the consistency and continuity of the data, and finally generates a collaborative control strategy to optimize the chilled water valve opening and power supply link settings.

[0106] In an embodiment of the present application, first, the temperature gradient distribution and pressure loss gradient parameters are collected in real time from each monitoring point in the data center, and a specific algorithm is used to process these data to identify key features of system state changes; secondly, based on the above data analysis results, the correction coefficient of the cooling water pump output percentage is calculated through a closed-loop feedback mechanism, taking into account the temperature and pressure loss change patterns in different time periods and their impact on the cooling efficiency; thirdly, the calculated cooling water pump output percentage correction coefficient is combined with the dynamic compensation factor in the fluid mechanics topology matrix, and the impedance matching map is cross-validated to ensure data consistency and continuity; finally, a collaborative control strategy is formulated, including a dynamic adjustment curve of the chilled water valve opening threshold and an optimized setting of the harmonic suppression parameters of the power supply link, to ensure the optimal operating state of the refrigeration equipment and the power supply system.

[0107] Here is a specific example:

[0108] Suppose a large data center wants to improve energy efficiency by optimizing its cooling system. First, the data center collects temperature gradient distribution and pressure loss gradient parameters from each monitoring point in real time, and uses a specific algorithm to process these data to identify the key features of system state changes; secondly, based on the above data analysis results, the correction coefficient of the cooling water pump output percentage is calculated through a closed-loop feedback mechanism, taking into account the temperature and pressure loss change patterns in different time periods and their impact on cooling efficiency; thirdly, the calculated cooling water pump output percentage correction coefficient is combined with the dynamic compensation factor in the fluid mechanics topology matrix, and the impedance matching map is cross-validated to ensure the consistency and continuity of the data; finally, the data center formulates a collaborative control strategy, including the dynamic adjustment curve of the chilled water valve opening threshold and the optimization setting of the power supply link harmonic suppression parameters, to ensure the optimal operating state of the refrigeration equipment and power supply system. Through the above steps, the data center not only improves the energy efficiency, but also enhances the stability and response speed of the system, ensuring the optimal operating state.

[0109] In order to further improve the accuracy of data center energy consumption prediction and optimize the response speed of the cooling system, in some embodiments, according to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through the path tracing engine and the impedance matching map is updated synchronously, including:

[0110] The temperature gradient distribution and pressure loss gradient parameters are collected from each monitoring point in the data center in real time. The data are processed using advanced data analysis algorithms to extract key features reflecting the system status and convert them into intermediate parameters that can be used in subsequent steps. Based on the above data analysis results, the key propagation path identifiers are identified through the path tracing engine to predict future environmental change trends, and the working status of the cooling water pump is adjusted accordingly, and the correction factor of the cooling water pump output percentage is calculated. The calculated cooling water pump output percentage correction factor is combined with other relevant parameters, and the impedance matching map is updated synchronously to ensure the coordinated work of all components and optimize the overall system performance. Combined with the results of all steps, a collaborative control strategy is formulated or optimized, including the dynamic adjustment curve of the chilled water valve opening threshold, the setting of the power supply link harmonic suppression parameters and other necessary operating instructions to ensure that the system can respond to changes in real-time environmental parameters and is forward-looking and flexible.

[0111] In this embodiment, the method involves collecting temperature gradient distribution and pressure loss gradient parameters from various monitoring points in the data center in real time, and processing these data using advanced data analysis algorithms to extract key features reflecting the system status. In addition, it also uses a path tracing engine to identify key propagation path identifiers and calculates correction factors for the cooling water pump output percentage. Finally, by synchronously updating the impedance matching map, the coordination between all components is ensured, and a collaborative control strategy is formulated to optimize the chilled water valve opening and power supply link settings.

[0112] In an embodiment of the present application, first, the temperature gradient distribution and pressure loss gradient parameters are collected in real time from each monitoring point in the data center, and the data are processed using an advanced data analysis algorithm to extract key features reflecting the system status and convert them into intermediate parameters that can be used in subsequent steps; secondly, based on the above data analysis results, the key propagation path identifier is identified through the path tracing engine, the future environmental change trend is predicted, and the working state of the cooling water pump is adjusted accordingly, and the correction coefficient of the cooling water pump output percentage is calculated; thirdly, the calculated cooling water pump output percentage correction coefficient is combined with other relevant parameters, and the impedance matching map is updated synchronously to ensure the coordinated work between all components and optimize the overall system performance; finally, based on the results of all steps, a collaborative control strategy is formulated or optimized, including a dynamic adjustment curve of the chilled water valve opening threshold, the setting of the power supply link harmonic suppression parameters and other necessary operating instructions to ensure that the system can respond to changes in real-time environmental parameters and is forward-looking and flexible.

[0113] Here is a specific example:

[0114] Suppose a large data center wants to improve energy efficiency by optimizing its cooling system. First, the data center collects temperature gradient distribution and pressure loss gradient parameters from each monitoring point in real time, and uses advanced data analysis algorithms to process these data, extract key features reflecting the system status, and convert them into intermediate parameters that can be used in subsequent steps; secondly, based on the above data analysis results, the key propagation path identifier is identified through the path tracing engine, the future environmental change trend is predicted, and the working state of the cooling water pump is adjusted accordingly, and the correction coefficient of the cooling water pump output percentage is calculated; thirdly, the calculated cooling water pump output percentage correction coefficient is combined with other related parameters, and the impedance matching map is updated synchronously to ensure the coordinated work between all components and optimize the overall system performance; finally, combined with the results of all steps, the data center has formulated a collaborative control strategy, including the dynamic adjustment curve of the chilled water valve opening threshold, the setting of the power supply link harmonic suppression parameters and other necessary operation instructions, to ensure that the system can respond to changes in real-time environmental parameters and has foresight and flexibility. Through the above steps, the data center not only improves energy efficiency, but also enhances the stability and response speed of the system, ensuring the best operating state.

[0115] Figure 2 A schematic diagram of a DCIM-based data center energy consumption prediction system is provided for an embodiment of the present application. Figure 2 As shown, the device comprises:

[0116] The acquisition module 21 is used to synchronously acquire the real-time operating parameters of the power monitoring module, the refrigeration equipment control module and the environmental sensor module in the DCIM system through the heterogeneous data bus, and use the dynamic topology calibration mechanism to perform spatiotemporal alignment processing on the equipment operation status data and the temperature and humidity distribution data of the computer room to generate a dynamic topology matrix;

[0117] A construction module 22 is used to superimpose the dynamic weight coefficient of the energy propagation path on the physical connection relationship of the device based on the topological matching degree between the dynamic topology matrix and the virtual twin model to construct a propagation path weight matrix;

[0118] The calculation module 23 is used to inject an event trigger signal into the propagation path weight matrix according to the timestamp sequence of the equipment start and stop events in the historical operation cycle and the phase offset of the energy consumption fluctuation curve, calculate the thermodynamic propagation delay of the equipment state change event through the path superposition effect, and generate a multi-dimensional prediction map;

[0119] The adjustment module 24 is used to locate the key propagation path identifier through the path tracing mechanism based on the phase deviation between the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, dynamically adjust the impedance coefficient and the pressure loss threshold in the propagation path weight matrix, and generate the power gradient adjustment instruction of the refrigeration equipment and the power supply link impedance matching instruction.

[0120] Figure 2 The DCIM-based data center energy consumption prediction device can be executed Figure 1 The implementation principle and technical effect of the DCIM-based data center energy consumption prediction method described in the illustrated embodiment will not be repeated. The specific manner in which each module and unit performs operations in the DCIM-based data center energy consumption prediction device in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0121] In one possible design, Figure 2 The DCIM-based data center energy consumption prediction device of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0122] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0123] The processing component 32 is used for the above Figure 1 The embodiment provides a method for predicting energy consumption of a data center based on DCIM.

[0124] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0125] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0126] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0127] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0128] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0129] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0130] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is a method for predicting energy consumption of a data center based on DCIM.

[0131] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0132] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0133] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A data center energy consumption prediction method based on DCIM, characterized in that: include: The real-time operating parameters of the power monitoring module, refrigeration equipment control module and environmental sensor module in the DCIM system are synchronously obtained through the heterogeneous data bus. The dynamic topology calibration mechanism is used to perform spatiotemporal alignment processing on the equipment operation status data and the room temperature and humidity distribution data to generate a dynamic topology matrix. Based on the topological matching degree between the dynamic topological matrix and the virtual twin model, the dynamic weight coefficient of the energy propagation path is superimposed on the physical connection relationship of the device to construct a propagation path weight matrix; According to the phase offset between the timestamp sequence of the equipment start and stop events in the historical operation cycle and the energy consumption fluctuation curve, an event trigger signal is injected into the propagation path weight matrix, and the thermodynamic propagation delay of the equipment state change event is calculated through the path superposition effect to generate a multi-dimensional prediction map; Based on the phase deviation between the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, the key propagation path identifier is located through the path tracing mechanism, the impedance coefficient and the pressure loss threshold in the propagation path weight matrix are dynamically adjusted, and the power gradient adjustment instructions of the refrigeration equipment and the power supply link impedance matching instructions are generated.

2. The method according to claim 1, characterized in that According to the phase offset between the timestamp sequence of equipment start and stop events in the historical operation cycle and the energy consumption fluctuation curve, an event trigger signal is injected into the propagation path weight matrix, and the thermodynamic propagation delay of the equipment state change event is calculated through the path superposition effect to generate a multi-dimensional prediction map, including: The dynamic topology calibration engine performs nonlinear coupling analysis on the equipment power fluctuation coefficient and the refrigerant flow gradient, and generates a dynamic topology matrix by combining the cabinet inlet air temperature gradient distribution and the air conditioner return air temperature phase difference. Based on the impedance coefficient and pressure loss threshold of the propagation path weight matrix, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage parameters are integrated to build a thermodynamic disturbance propagation simulation model for equipment state change events. According to the phase deviation of the transient peak distribution in the prediction map, the key propagation path identifier is located and the impedance correction coefficient is calculated through the path tracing engine, and the chilled water flow fluctuation parameters and the cooling water pump output percentage are synchronously integrated for dynamic compensation; Combined with the dynamic matching relationship between real-time environmental parameters and impedance correction coefficients, a closed-loop feedback mechanism is used to collaboratively optimize the chilled water valve opening threshold and the power supply link impedance parameters to generate the power gradient adjustment instructions for the refrigeration equipment and the power supply link reconstruction strategy.

3. The method according to claim 2, characterized in that Based on the impedance coefficient and pressure loss threshold of the propagation path weight matrix, the pressure loss gradient of the chilled water circulation path and the cooling tower fan output percentage parameters are integrated to build a thermodynamic disturbance propagation simulation model for equipment state change events, including: The dynamic fluid dynamics engine is used to synchronize the pressure loss gradient of the chilled water circulation path with the cooling tower fan output percentage, and the chilled water return temperature gradient distribution and the power supply link harmonic distortion rate parameters are integrated to generate a fluid dynamics topology matrix. Based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient are superimposed to generate an impedance matching map with time series markings. According to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid mechanics topology matrix and the impedance matching map are synchronously integrated for cross-validation; The fluid mechanics topology matrix and the power supply link impedance parameters are integrated, and the dynamic adjustment curve of the chilled water valve opening threshold and the power supply link harmonic suppression parameters are combined to generate a collaborative control strategy.

4. The method according to claim 2, characterized in that: According to the phase deviation of the transient peak distribution in the prediction map, the key propagation path identifier is located and the impedance correction coefficient is calculated through the path tracing engine, and the chilled water flow fluctuation parameters and the cooling water pump output percentage are synchronously integrated for dynamic compensation, including: The dynamic fluid dynamics engine is used to synchronize the pressure loss gradient of the chilled water circulation path with the cooling tower fan output percentage, and the power supply link harmonic distortion rate parameters are integrated to generate a fluid dynamics topology matrix. Based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, the dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and an impedance matching map with time series marking is generated; According to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction factor is calculated through the path tracing engine and the impedance matching map is updated synchronously; The fluid mechanics topology matrix and the updated impedance matching map are integrated, and the dynamic adjustment curve of the chilled water valve opening threshold is combined to generate the power gradient adjustment instructions and power supply link reconstruction strategy of the refrigeration equipment.

5. The method according to claim 3, characterized in that: Based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, a dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is constructed, and the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient are superimposed to generate an impedance matching map with time series markings, including: The chilled water circulation system is modeled through a dynamic fluid dynamics engine, and the chilled water flow fluctuation parameters, cooling water pump output percentage and its correction coefficient are integrated to generate a fluid dynamics topology matrix; Based on the propagation delay coefficient in the thermodynamic disturbance propagation simulation model, the dynamic matching relationship between the power supply link impedance parameter and the chilled water valve opening threshold is analyzed, and the influence of the chilled water flow fluctuation parameter and the cooling water pump output percentage correction coefficient is considered; Combined with the temperature gradient distribution and pressure loss gradient parameters in the environmental parameters, the cooling water pump output percentage correction factor is calculated, and the impedance matching map is updated synchronously; The fluid mechanics topology matrix and the updated impedance matching map are integrated, and the dynamic adjustment curve of the chilled water valve opening threshold is combined to generate the refrigeration equipment power gradient adjustment instruction and the power supply link reconstruction strategy.

6. The method according to claim 3, characterized in that According to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction coefficient is calculated through a closed-loop feedback mechanism, and the dynamic compensation factor in the fluid mechanics topology matrix and the impedance matching map are synchronously integrated for cross-validation, including: Collect temperature gradient distribution and pressure loss gradient parameters from each monitoring point in the data center in real time, process the temperature gradient distribution and pressure loss gradient parameters using a specific algorithm, identify key features reflecting changes in system status, and obtain data analysis results; Based on the data analysis results, a correction factor for the cooling water pump output percentage is calculated through a closed-loop feedback mechanism, taking into account the temperature and pressure loss variation patterns in different time periods and their impact on cooling efficiency; Combine the calculated cooling water pump output percentage correction factor with the dynamic compensation factor in the fluid mechanics topology matrix to cross-validate the impedance matching map to ensure data consistency and continuity; Develop a collaborative control strategy, including a dynamic adjustment curve for the chilled water valve opening threshold and optimal setting of the power supply link harmonic suppression parameters, to ensure the optimal operating state of the refrigeration equipment and power supply system.

7. The method according to claim 4, characterized in that According to the temperature gradient distribution and pressure loss gradient parameters in the real-time environmental parameters, the cooling water pump output percentage correction factor is calculated through the path tracing engine and the impedance matching map is updated synchronously, including: Collect temperature gradient distribution and pressure loss gradient parameters from each monitoring point in the data center in real time, process the temperature gradient distribution and pressure loss gradient parameters using advanced data analysis algorithms, extract key features reflecting the system status, and convert them into intermediate parameters that can be used in subsequent steps; Based on the data analysis results, the path tracing engine is used to identify key propagation path identifiers, predict future environmental change trends, adjust the working status of the cooling water pump accordingly, and calculate the correction coefficient of the cooling water pump output percentage; Combine the calculated cooling water pump output percentage correction factor with other relevant parameters to synchronously update the impedance matching map to ensure coordinated operation among all components and optimize overall system performance; Combining the results of all steps, formulate or optimize the collaborative control strategy, including the dynamic adjustment curve of the chilled water valve opening threshold, the setting of the power supply link harmonic suppression parameters and other necessary operating instructions, to ensure that the system can respond to changes in real-time environmental parameters and is forward-looking and flexible.

8. A data center energy consumption prediction system based on DCIM, characterized in that: include: The acquisition module is used to synchronously acquire the real-time operating parameters of the power monitoring module, refrigeration equipment control module and environmental sensor module in the DCIM system through the heterogeneous data bus, and use the dynamic topology calibration mechanism to perform spatiotemporal alignment processing on the equipment operation status data and the room temperature and humidity distribution data to generate a dynamic topology matrix; A construction module, used to superimpose the dynamic weight coefficient of the energy propagation path on the physical connection relationship of the device based on the topological matching degree between the dynamic topology matrix and the virtual twin model, and construct a propagation path weight matrix; A calculation module is used to inject an event trigger signal into the propagation path weight matrix according to the timestamp sequence of the equipment start and stop events in the historical operation cycle and the phase offset of the energy consumption fluctuation curve, calculate the thermodynamic propagation delay of the equipment state change event through the path superposition effect, and generate a multi-dimensional prediction map; The adjustment module is used to locate the key propagation path identifier through the path tracing mechanism based on the phase deviation between the transient peak distribution and the real-time environmental parameters in the multi-dimensional prediction map, dynamically adjust the impedance coefficient and the pressure loss threshold in the propagation path weight matrix, and generate the power gradient adjustment instruction of the refrigeration equipment and the impedance matching instruction of the power supply link.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a DCIM-based data center energy consumption prediction method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a data center energy consumption prediction method based on DCIM is implemented as described in any one of claims 1 to 7.

Images