Super data center dynamic energy efficiency optimization and carbon tracing system

Abstract

The invention relates to the technical field of energy efficiency management of a super data center, and discloses a dynamic energy efficiency optimization and carbon traceability system for a super data center, which comprises the following steps of: acquiring energy consumption, an energy supply chain and environmental data of the super data center through an acquisition module, and performing fusion and time sequence processing to provide basic data for an analysis module; and the analysis module establishes a carbon footprint prediction model based on the time sequence data, calculates layered carbon emission, and constructs a multi-objective optimization model to generate an energy efficiency optimization strategy. And the central control module adjusts the operation of the equipment according to the strategy, compares the adjusted data with the prediction result, judges whether to correct the strategy, and realizes closed-loop optimization. And the display module visually presents the time sequence data and the optimization strategy, so that a closed-loop intelligent management mechanism of data acquisition, analysis and prediction, strategy generation and execution correction is realized, and the energy efficiency management precision and carbon emission controllability of the data center are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy efficiency management of a super data center, and in particular to a dynamic energy efficiency optimization and carbon tracing system for a super data center. Background Art

[0002] As an important part of information infrastructure, existing data centers have been growing in scale and energy consumption, and have become a major source of energy consumption and carbon emissions.

[0003] However, traditional data centers rely primarily on static configuration and empirical adjustments for energy efficiency management, lacking systematic integration and dynamic analysis of operational energy consumption, energy supply chain, and external environmental data. This results in inaccurate energy efficiency optimization and difficulty effectively controlling carbon emissions. Furthermore, the types of energy used in data centers are becoming increasingly diverse, and the carbon emission characteristics of different energy sources within the supply chain vary significantly. For example, thermal power has a higher carbon emission factor, while clean energy sources such as wind power and photovoltaics have relatively lower carbon emission factors. Existing technologies generally lack the ability to jointly model and trace energy supply chain data and operational energy consumption data in real time, making it difficult to perform hierarchical calculations and dynamic tracking of carbon emission sources, resulting in inaccurate carbon emission accounting. Furthermore, while some research has attempted to monitor and analyze data center energy consumption through energy efficiency management platforms, deficiencies remain in multi-objective optimization decision-making, dynamic central control adjustments, and transparent display of carbon emissions. This makes it impossible to achieve closed-loop management of the entire process, from energy consumption collection and carbon footprint prediction to optimization strategy generation and policy revision.

[0004] Therefore, it is urgent to propose a hyper data center management system that can integrate operational data, supply chain data, and environmental data, and has the ability to trace carbon emissions in layers and dynamically optimize energy efficiency, in order to address the shortcomings of existing technologies. Summary of the Invention

[0005] In view of this, the present invention proposes a dynamic energy efficiency optimization and carbon traceability system for super data centers, aiming to solve the problem that existing data centers lack the full-process closed-loop capabilities of multi-source data fusion, layered carbon emission accounting and dynamic optimization in energy efficiency management and carbon emission traceability, resulting in inaccurate energy efficiency optimization and insufficient carbon emission control.

[0006] The present invention proposes a hyper data center dynamic energy efficiency optimization and carbon traceability system, comprising: The acquisition module is configured to obtain operating energy consumption data, energy supply chain data, and environmental data of the operating equipment of the hyper data center, perform fusion processing based on the operating energy consumption data, energy supply chain data, and environmental data, and construct time-series operating data of the hyper data center; an analysis module electrically connected to the acquisition module, the analysis module being configured to establish a carbon footprint prediction model based on the time series operation data and determine stratified carbon emission results. The analysis module is further configured to construct a multi-objective optimization model and generate an energy efficiency optimization strategy based on the stratified carbon emission results and the time series operation data; a central control module electrically connected to the analysis module, the central control module being configured to adjust the operating state of the super-data center's operating equipment according to the energy efficiency optimization strategy, and further configured to compare the super-data center's operating energy consumption data, energy supply chain data, and environmental data after the adjusted operating state with the stratified carbon emission results, and determine whether to modify the energy efficiency optimization strategy based on the comparison results; The display module is electrically connected to the acquisition module and the central control module respectively. The display module is configured to display the timing operation data and energy efficiency optimization strategy of the super data center.

[0007] Furthermore, the acquisition module integrates and processes the operating energy consumption data, energy supply chain data, and environmental data, and constructs the time series operating data of the hyper data center, including: The acquisition module is further configured to obtain timestamps of the operating energy consumption data, the energy supply chain data, and the environmental data, and unify the timestamps. The acquisition module is further configured to construct a time-synchronized energy consumption data set based on the operating energy consumption data after the unified timestamps; the acquisition module is further configured to construct a time-synchronized energy supply chain data set based on the energy supply chain data after the unified timestamps; the acquisition module is further configured to construct a time-synchronized environmental data set based on the environmental data after the unified timestamps; The acquisition module is further configured to perform data cleaning on the time-synchronized energy consumption dataset, the time-synchronized energy supply chain dataset, and the time-synchronized environment dataset respectively; The acquisition module is further configured to fuse the cleaned synchronized energy consumption dataset, the time-synchronized energy supply chain dataset, and the time-synchronized environment dataset based on a unified time reference, and generate a fused operation dataset; The acquisition module is further configured to perform time series modeling based on the fusion operation data set, and determine time series operation data based on the time series modeling.

[0008] Furthermore, the acquisition module performs time series modeling based on the fusion operation data set, and determines the time series operation data based on the time series modeling, including: The acquisition module is further configured to sort each data in the fusion running data set according to the timestamp and obtain the initial time series data set; The acquisition module is further configured to fill in missing values ​​or discontinuous data in the initial time series data set to obtain a complete time series data set; The acquisition module is further configured to extract energy consumption characteristics, environmental characteristics, and supply chain characteristics based on the complete time series data set, and establish a feature data set based on the energy consumption characteristics, environmental characteristics, and supply chain characteristics; The acquisition module is further configured to slice the feature dataset according to a preset time window and data granularity, and generate a sliced ​​dataset; The acquisition module is further configured to perform fitting modeling based on the time series slice data, and determine the time series slice data after fitting modeling as time series running data.

[0009] Furthermore, the analysis module is configured to establish a carbon footprint prediction model based on the time series operation data, and to determine the stratified carbon emission results, including: The analysis module is further configured to determine the energy type of the energy supply chain based on the supply chain characteristics in the time series operation data and based on the supply chain characteristics; The analysis module is further configured to determine a carbon factor based on the energy consumption characteristics and energy type in the time series operation data, and obtain carbon factor mapping data based on the carbon factor; The analysis module is further configured to determine the carbon emissions of each time segment and each device based on the carbon factor mapping data, and construct a carbon emissions dataset based on the carbon emissions of each time segment and each device; The analysis module is further configured to perform hierarchical aggregation on the carbon emission dataset and generate a hierarchical carbon emission result.

[0010] Furthermore, the analysis module is further configured to determine the carbon factor based on the energy consumption characteristics and energy types in the time series operation data, including: The analysis module is further configured to obtain energy type data from the time series operation data based on the energy consumption characteristics and the energy type; The analysis module is further configured to connect to an authoritative carbon factor database in real time based on the energy type data, retrieve and obtain the corresponding benchmark carbon factor, and form a carbon factor data set; The analysis module is further configured to match the energy consumption characteristic data with the carbon factor dataset in a time dimension based on time series data alignment and fusion to obtain preliminary mapping data; The analysis module is further configured to perform dynamic rule adjustment on the preliminary mapping data based on the carbon factor mapping rule engine to generate carbon factor mapping data.

[0011] Furthermore, the analysis module is further configured to dynamically adjust the rules of the preliminary mapping data based on the carbon factor mapping rule engine, including: The analysis module is further configured to compare the preliminary mapping data with the preset mapping data in the carbon factor mapping rule engine and normalize the comparison results; the analysis module is further configured to determine the accurate value of the preliminary mapping data based on the normalized comparison results; The analysis module is further configured to determine whether to perform dynamic rule adjustment on the preliminary mapping data based on a relationship between the accuracy value of the preliminary mapping data and the accuracy value of the preset mapping data pre-configured by the analysis module: When the accuracy value of the preliminary mapping data is lower than or equal to the preset accuracy value of the mapping data, the analysis module determines not to perform dynamic rule adjustment on the preliminary mapping data; When the accuracy value of the preliminary mapping data is higher than the preset accuracy value of the mapping data, the analysis module determines to perform dynamic rule adjustment on the preliminary mapping data.

[0012] Furthermore, when the accuracy value of the preliminary mapping data is higher than the preset accuracy value of the mapping data, the analysis module determines to adjust the dynamic rules of the preliminary mapping data, including: The analysis module is further configured to determine energy type, equipment type, and time period mapping items to be adjusted based on the preliminary mapping data and the carbon factor mapping rule engine rules; The analysis module is further configured to secondary determine the carbon factor value and corresponding weight based on the real-time energy type proportion, energy consumption characteristics and carbon factor rules; The analysis module is further configured to establish an adjustment parameter set based on the secondarily determined carbon factor value and the corresponding weight; The analysis module is further configured to adjust the preliminary mapping data based on the adjustment parameter set.

[0013] Furthermore, the analysis module constructs a multi-objective optimization model and generates an energy efficiency optimization strategy based on the layered carbon emission results and the time series operation data, including: The analysis module is further configured to construct a multi-objective optimization function based on the stratified carbon emission results and the time series operation data; The analysis module is further configured to establish device operation state constraints and system stability constraints based on the time series operation data; The analysis module is further configured to solve the optimization function based on the optimization algorithm and determine a preliminary energy efficiency optimization strategy combination; The analysis module is further configured to map the preliminary energy efficiency optimization strategy into an operating instruction of the super data center operating equipment, and generate a final energy efficiency optimization strategy based on the operating instruction of the super data center operating equipment.

[0014] Furthermore, the central control module determines whether to modify the energy efficiency optimization strategy based on the comparison results, including: The central control module is further configured to obtain deviation indicators of carbon intensity, PUE, and power / electricity based on the comparison results, and extract supply structure change indicators such as energy share change, carbon factor difference, and energy supply fluctuation based on the deviation indicators of carbon intensity, PUE, and power / electricity. The central control module is further configured to construct a deviation feature parameter set based on the deviation indicators of carbon intensity, PUE, and power / electricity and the supply structure change indicators such as energy share change, carbon factor difference, and energy supply fluctuation. The central control module is also configured to perform hard threshold verification on the deviation characteristic parameter set based on the preset SLA, and obtain the hard threshold determination result and the corresponding trigger item list; The central control module is also configured to perform statistical significance testing on the deviation characteristic parameter set and determine the statistical score of the abnormal sign; The central control module is further configured to determine an adjustment score based on the deviation characteristic parameter set, the hard threshold determination result, and the abnormal flag statistical score, and determine whether to modify the energy efficiency optimization strategy based on the relationship between the adjustment score and the preset adjustment score configured in the central control module: When the adjustment score is lower than the preset adjustment score, the central control module determines to revise the energy efficiency optimization strategy; When the adjustment score is higher than or equal to the preset adjustment score, the central control module determines not to modify the energy efficiency optimization strategy.

[0015] Furthermore, when the adjustment score is lower than the preset adjustment score, the central control module determines to revise the energy efficiency optimization strategy, including: The central control module is also configured to extract the deviation ratio of each indicator based on the deviation characteristic parameter set; The central control module is further configured to determine a correction coefficient based on the deviation ratio of each indicator and a preset weight factor corresponding to each indicator; The central control module is also configured to generate a correction parameter set based on the correction coefficient and to correct the energy efficiency optimization strategy based on the correction parameter set, wherein the correction parameter set includes cooling strategy parameters, energy call priority parameters and load migration parameters.

[0016] Compared with the existing technology, the beneficial effect of the present invention is that: through the acquisition module, the operating energy consumption data, energy supply chain data and environmental data are integrated and processed, and time-series operating data is constructed, which realizes the unified modeling and dynamic association of multi-source heterogeneous data, can avoid the information island problem brought by traditional single energy consumption monitoring, and thus provide high-precision, comprehensive basic data support for subsequent carbon footprint prediction and optimization analysis. Secondly, through the analysis module, a carbon footprint prediction model is established and the layered carbon emission results are determined, which can finely decompose and dynamically track the carbon emissions of different energy types and different operating links, and realize the transparency and accuracy of carbon emission accounting. At the same time, the analysis module constructs a multi-objective optimization model based on the layered carbon emission results and time-series operating data, and generates a highly targeted energy efficiency optimization strategy, thereby taking into account the dual goals of energy consumption reduction and carbon emission control. Thirdly, the central control module can implement the optimization strategy to the state adjustment of the operating equipment, and after execution, compare the new operating data with the layered carbon emission results to determine whether the optimization effect meets expectations. When optimization results are insufficient, the central control module promptly adjusts the energy efficiency optimization strategy, achieving dynamic closed-loop adjustments and significantly improving the real-time and adaptability of energy efficiency optimization. Finally, the display module visualizes operational data and optimization strategies, making system operating status, carbon emissions distribution, and optimization strategies transparent and visible. This helps operations personnel keep abreast of energy efficiency and carbon emissions, supporting scientific decision-making and continuous improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings: Figure 1 This is a functional block diagram of a hyper data center dynamic energy efficiency optimization and carbon traceability system provided by an embodiment of the present invention; Figure 2 A flow chart of a dynamic energy efficiency optimization and carbon traceability system for a hyper data center provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0019] like Figure 1-Figure 2 As shown, in some embodiments of the present application, this embodiment provides a super data center dynamic energy efficiency optimization and carbon traceability system, including: a collection module, an analysis module, a central control module and a display module.

[0020] Specifically, the acquisition module is configured to obtain the operating energy consumption data, energy supply chain data and environmental data of the super data center operating equipment, perform fusion processing based on the operating energy consumption data, energy supply chain data and environmental data, and construct the time series operating data of the super data center; the analysis module is electrically connected to the acquisition module, and the analysis module is configured to establish a carbon footprint prediction model based on the time series operating data and determine the stratified carbon emission results. The analysis module is also configured to construct a multi-objective optimization model and generate an energy efficiency optimization strategy based on the stratified carbon emission results and the time series operating data; the central control module is electrically connected to the analysis module, and the central control module is configured to adjust the operating status of the super data center operating equipment according to the energy efficiency optimization strategy. The central control module is also configured to compare the operating energy consumption data, energy supply chain data and environmental data of the super data center after the operating status is adjusted with the stratified carbon emission results, and determine whether to revise the energy efficiency optimization strategy based on the comparison results; the display module is electrically connected to the acquisition module and the central control module respectively, and the display module is configured to display the time series operating data and energy efficiency optimization strategy of the super data center.

[0021] As you can understand, the acquisition module, deployed on key equipment in the hyperscale data center, collects operational energy consumption data, energy supply chain data, and environmental data. Leveraging IoT technology and edge computing, the acquisition module integrates and processes multi-source heterogeneous data, eliminating data noise, standardizing data from different sources, and constructing unified time-series operational data. This process enables comprehensive and dynamic awareness of equipment operating status, energy supply conditions, and environmental factors, providing an accurate and continuous data foundation for subsequent analysis. Secondly, the analysis module establishes a carbon footprint prediction model based on the collected time-series operational data and performs stratified calculations of carbon emissions at different levels (device, region, and system), enabling refined carbon emissions tracking. Furthermore, the analysis module combines the stratified carbon emission results with the time-series operational data to construct a multi-objective optimization model. Through algorithmic analysis, it generates energy efficiency optimization strategies that reduce energy consumption while minimizing carbon emissions, forming a dual optimization decision-making mechanism that balances energy consumption and carbon emissions. Furthermore, the central control module adjusts operating equipment based on the optimization strategy while simultaneously monitoring the adjusted energy consumption, energy supply, and environmental data in real time. By comparing the results with the stratified carbon emissions, the central control module determines whether the current optimization strategy is achieving the desired effect and dynamically adjusts it when necessary, achieving closed-loop control for energy efficiency optimization. This ensures the real-time adaptability of the strategy and the reliability of its execution. Finally, the display module intuitively presents the collected time-series operation data and the generated optimization strategy to the user, visualizing system operating status, energy efficiency indicators, and carbon emissions information. This not only supports real-time monitoring and analysis by operations and maintenance personnel, but also provides data support for scientific decision-making, making the energy efficiency optimization and carbon traceability processes transparent and manageable.

[0022] Specifically, the collection module is deployed on various key devices in a hyperdata center, including servers, storage devices, network equipment, air conditioning systems, chillers, and lighting systems. Using high-precision sensors such as smart meters, temperature and humidity sensors, and flow sensors, the collection module collects real-time data on device operating energy consumption (active power, reactive power, and electrical energy), operating status parameters (temperature, humidity, wind speed, and flow), energy supply chain data (grid type and supply ratio, real-time energy prices), and environmental data (outdoor temperature, humidity, and wind speed). The collection module utilizes edge computing nodes to preprocess, fuse, and model the raw data in a time-series fashion, generating unified hyperdata center time-series operating data that provides accurate and continuous data input for the analysis module. The analysis module then receives the time-series operating data provided by the collection module and, based on this data, develops a carbon footprint prediction model, performing hierarchical calculations of carbon emissions at the device, regional, and system levels. For example, by analyzing server load, power source, and device energy consumption characteristics, it predicts the carbon emissions of each server over different time periods. The analysis module further combines the stratified carbon emission results with time-series operational data to construct a multi-objective optimization model. Using particle swarm optimization (PSO) or genetic algorithms (GA), it generates energy efficiency optimization strategies, including dynamic server frequency scaling, cooling system capacity adjustment, and optimized power load distribution. Furthermore, the central control module, based on the optimization strategies generated by the analysis module, adjusts the operating status of various hyperscale data center equipment in real time, such as switching underloaded servers to sleep mode, adjusting air conditioning fan speeds, or optimizing UPS charging and discharging strategies. After these adjustments, the central control module monitors operational energy consumption data, energy supply chain data, and environmental data in real time and compares them with the stratified carbon emission results. If actual operating energy consumption or carbon emissions deviate from predicted values ​​by more than a preset threshold, the central control module automatically adjusts the optimization strategy, ensuring a closed-loop implementation of energy efficiency optimization and carbon emission reduction goals. Finally, the display module visually presents the data results from the acquisition and central control modules to the user, including device-level and regional-level energy consumption curves, carbon emission trends, and optimization strategy execution. Users can use the interactive interface to select specific time periods, devices, or areas for detailed viewing, enabling transparent management of the hyperscale data center's operating status and carbon footprint, and optimizing operation and maintenance decisions based on system prompts.

[0023] Specifically, when the acquisition module performs fusion processing based on the operating energy consumption data, energy supply chain data and environmental data, and constructs the time-series operating data of the super data center, it includes: the acquisition module is also configured to obtain the timestamps of the operating energy consumption data, energy supply chain data and environmental data, and unify the timestamps, and the acquisition module is also configured to construct a time-synchronized energy consumption data set based on the operating energy consumption data after the unified timestamp; the acquisition module is also configured to construct a time-synchronized energy supply chain data set based on the energy supply chain data after the unified timestamp; the acquisition module is also configured to construct a time-synchronized environment data set based on the environmental data after the unified timestamp; the acquisition module is also configured to perform data cleaning on the time-synchronized energy consumption data set, the time-synchronized energy supply chain data set and the time-synchronized environment data set respectively; the acquisition module is also configured to perform data fusion on the synchronized energy consumption data set, the time-synchronized energy supply chain data set and the time-synchronized environment data set after data cleaning based on a unified time base, and generate a fused operating data set; the acquisition module is also configured to perform time-series modeling based on the fused operating data set, and determine the time-series operating data based on the time-series modeling.

[0024] Specifically, the acquisition module performs time series modeling based on the fusion operation data set, and determines the time series operation data based on the time series modeling, including: the acquisition module is also configured to sort the data in the fusion operation data set according to the timestamp, and obtain the initial time series data set; the acquisition module is also configured to complete the missing values ​​or discontinuous data in the initial time series data set to obtain a complete time series data set; the acquisition module is also configured to extract energy consumption characteristics, environmental characteristics and supply chain characteristics based on the complete time series data set, and establish a feature data set based on the energy consumption characteristics, environmental characteristics and supply chain characteristics; the acquisition module is also configured to slice the feature data set according to the preset time window and data granularity, and generate a slice data set; the acquisition module is also configured to perform fitting modeling based on the time series slice data, and determine the time series operation data based on the time series slice data after fitting modeling.

[0025] As can be understood, the acquisition module first obtains timestamps for operational energy consumption data, energy supply chain data, and environmental data, and then unifies the timestamps of these data types to ensure precise alignment of data from different sources using the same time base. Based on the unified timestamps, the acquisition module constructs time-synchronized energy consumption, energy supply chain, and environmental datasets. This time synchronization ensures accurate correspondence between data sources during subsequent analysis, improving the accuracy and comparability of the fused data. Secondly, based on this time synchronization, the acquisition module cleans each dataset, including addressing missing values, outliers, and noisy data to ensure data quality. Subsequently, the acquisition module fuses the cleaned energy consumption, energy supply chain, and environmental data based on the unified time base to generate a fused operational dataset. This fusion step integrates information from multiple sources to form a complete and continuous description of the operational status, providing a reliable data foundation for time series analysis. Furthermore, the acquisition module performs time series modeling on the fused operational dataset. First, the fused data is sorted according to timestamps to form an initial time series dataset, and missing or discontinuous data is supplemented to obtain a complete time series dataset. The acquisition module then extracts energy consumption characteristics, environmental characteristics, and energy supply chain characteristics from the complete time series data to construct a multidimensional feature dataset. By setting the time window and data granularity, the feature data is sliced ​​to generate a slice dataset, providing structured time series samples for subsequent modeling. Finally, based on the slice dataset, the acquisition module performs fitting modeling on each time series slice, generating predictable time series patterns by modeling the trends of energy consumption, environmental, and supply chain characteristics over time. Ultimately, the time series slice data after fitting modeling is integrated into complete time series operation data, providing accurate data input for the analysis module to establish carbon footprint prediction models and optimization models, laying the foundation for energy efficiency optimization and carbon traceability in hyperscale data centers.

[0026] Specifically, the acquisition module first acquires operational energy consumption data, energy supply chain data, and environmental data from the hyperdata center's operating equipment, simultaneously recording the timestamps for each data type. The acquisition module then uniformly processes the acquired timestamps, aligning data from different sources to a unified time base. It then constructs time-synchronized energy consumption, energy supply chain, and environmental datasets, respectively, to ensure accurate correspondence between the various data types at the same time point, providing a foundation for subsequent data fusion. Secondly, the acquisition module cleans the time-synchronized datasets, addressing missing values, outliers, and noise to ensure data quality. Subsequently, based on the unified time base, the cleaned energy consumption, energy supply chain, and environmental datasets are fused to generate a fused operational dataset. This fused dataset integrates information from multiple sources to form a complete description of the hyperdata center's operational status, providing accurate input data for time series analysis. Furthermore, based on data fusion, the acquisition module sorts the fused data according to timestamps to form an initial time series dataset, and then completes any missing or discontinuous data to obtain a complete time series dataset. The acquisition module then extracts energy consumption, environmental, and energy supply chain characteristics from the complete time series dataset to create a multidimensional feature dataset. The acquisition module then slices the feature data according to the preset time window and data granularity to generate a sliced ​​dataset, providing structured time series samples for subsequent fitting and modeling. Finally, the acquisition module performs fitting and modeling on the sliced ​​dataset, analyzing the changing trends of energy consumption, environmental, and supply chain characteristics over time to generate modeling results for each time slice. Finally, the fitted and modeled sliced ​​data is integrated to form complete time series operational data. This time series operational data is used by the analysis module to establish a carbon footprint prediction model and generate multi-objective optimization strategies, providing accurate data support for dynamic energy efficiency optimization and carbon traceability of hyperscale data centers.

[0027] Specifically, the analysis module is configured to establish a carbon footprint prediction model based on the time series operation data, and to determine the stratified carbon emission results, including: the analysis module is also configured to determine the energy type of the energy supply chain based on the supply chain characteristics in the time series operation data and the supply chain characteristics; the analysis module is also configured to determine the carbon factor based on the energy consumption characteristics and energy type in the time series operation data, and obtain carbon factor mapping data based on the carbon factor; the analysis module is also configured to determine the carbon emissions of each time segment and each device based on the carbon factor mapping data, and construct a carbon emission data set based on the carbon emissions of each time segment and each device; the analysis module is also configured to perform stratified aggregation on the carbon emission data set and generate stratified carbon emission results.

[0028] Specifically, the analysis module is also configured to determine the carbon factor based on the energy consumption characteristics and energy type in the time series operation data, including: the analysis module is also configured to obtain the energy type data in the time series operation data based on the energy consumption characteristics and energy type; the analysis module is also configured to connect to the authoritative carbon factor database in real time based on the energy type data, retrieve and obtain the corresponding benchmark carbon factor, and form a carbon factor data set; the analysis module is also configured to match the energy consumption characteristic data with the carbon factor data set in the time dimension based on time series data alignment and fusion to obtain preliminary mapping data; the analysis module is also configured to dynamically adjust the rules of the preliminary mapping data based on the carbon factor mapping rule engine to generate carbon factor mapping data.

[0029] As can be understood, the analysis module extracts supply chain characteristics from time-series operational data to identify various energy types within the energy supply chain (e.g., thermal power, wind power, photovoltaic power, etc.). Different energy types have distinct carbon emission characteristics within the supply chain, and identifying energy types is fundamental to carbon emission calculation and tiered analysis. This process automatically identifies and classifies energy types by modeling characteristic parameters of equipment and the energy supply chain, providing key input for subsequent carbon factor acquisition. Secondly, the analysis module integrates energy consumption characteristics with the identified energy types and connects to an authoritative carbon factor database in real time to retrieve corresponding benchmark carbon factors and generate a carbon factor dataset. Subsequently, using time-series data alignment and fusion technology, the energy consumption characteristic data and carbon factor data are matched across time to form preliminary mapping data. Using the carbon factor mapping rule engine, the analysis module dynamically adjusts the rules within this preliminary mapping data to achieve precise mapping of carbon factors across different time periods, equipment, and energy types, generating the final carbon factor mapping data. Furthermore, the analysis module uses this generated carbon factor mapping data to calculate carbon emissions for each time segment and equipment, organizing it into a complete carbon emission dataset. Through multi-dimensional data processing techniques, we model the relationships between energy consumption characteristics, energy types, and carbon factors, providing technical support for the accurate calculation of carbon emissions. Finally, the analysis module aggregates the carbon emissions dataset into layers, summarizing and stratifying carbon emissions according to different dimensions (such as equipment type, time period, and energy type), generating stratified carbon emissions results. This stratified result can be used for carbon footprint prediction, optimization model construction, and multi-objective energy efficiency optimization, providing a scientific basis for energy efficiency management and carbon emissions tracking in hyperscale data centers.

[0030] Specifically, the analysis module first extracts energy supply chain characteristics from time-series operational data, including energy sources, power line types, and supply chain node information. Based on these supply chain characteristics, the analysis module determines the types of energy sources within the energy supply chain, such as thermal power, wind power, and photovoltaic power, providing baseline data for subsequent carbon emissions calculations. The analysis module then combines the energy consumption characteristics and identified energy types in the time-series operational data to determine the carbon factor for each device in each time segment. This process involves obtaining energy type data and, through real-time access to authoritative carbon factor databases, such as the IPCC database or regional carbon factor data published by the National Energy Administration, retrieving and obtaining corresponding benchmark carbon factors to form a carbon factor dataset. Using time-series data alignment and fusion technology, the analysis module matches the energy consumption characteristic data with the carbon factor dataset along the time dimension, generating preliminary mapping data. The analysis module then uses a carbon factor mapping rule engine to dynamically adjust the rules within the preliminary mapping data, incorporating factors such as time period, device type, and energy type to generate final carbon factor mapping data. Based on this generated carbon factor mapping data, the analysis module calculates carbon emissions for each time segment and device, organizing the results into a carbon emissions dataset. Finally, the analysis module aggregates the carbon emissions dataset into layers, for example by device type, time period, or energy type, to generate stratified carbon emissions results. These stratified carbon emissions results can be used to predict carbon footprints, build multi-objective optimization models, and generate energy efficiency optimization strategies, enabling refined management and dynamic tracking of carbon emissions for hyperscale data centers.

[0031] Specifically, when the analysis module is also configured to perform dynamic rule adjustment on the preliminary mapping data based on the carbon factor mapping rule engine, it includes: the analysis module is also configured to compare the preliminary mapping data with the preset mapping data in the carbon factor mapping rule engine, and normalize the comparison result; the analysis module is also configured to determine the accurate value of the preliminary mapping data based on the comparison result after normalization; the analysis module is also configured to determine whether to perform dynamic rule adjustment on the preliminary mapping data based on the relationship between the accurate value of the preliminary mapping data and the accurate value of the preset mapping data pre-configured by the analysis module: when the accurate value of the preliminary mapping data is lower than or equal to the accurate value of the preset mapping data, the analysis module determines not to perform dynamic rule adjustment on the preliminary mapping data; when the accurate value of the preliminary mapping data is higher than the accurate value of the preset mapping data, the analysis module determines to perform dynamic rule adjustment on the preliminary mapping data.

[0032] Specifically, when the accuracy value of the preliminary mapping data is higher than the accuracy value of the preset mapping data, the analysis module determines to perform dynamic rule adjustments on the preliminary mapping data, including: the analysis module is also configured to determine the energy type, equipment type and time period mapping items to be adjusted based on the preliminary mapping data and the carbon factor mapping rule engine rules; the analysis module is also configured to secondary determine the carbon factor value and the corresponding weight based on the real-time energy type proportion, energy consumption characteristics and carbon factor rules; the analysis module is also configured to establish an adjustment parameter set based on the secondary determined carbon factor value and the corresponding weight; the analysis module is also configured to adjust the preliminary mapping data based on the adjustment parameter set.

[0033] Specifically, when the analysis module constructs a multi-objective optimization model and generates an energy efficiency optimization strategy based on the stratified carbon emission results and the time series operation data, it includes: the analysis module is also configured to construct a multi-objective optimization function based on the stratified carbon emission results and the time series operation data; the analysis module is also configured to establish equipment operation status constraints and system stability constraints based on the time series operation data; the analysis module is also configured to solve the optimization function based on the optimization algorithm to determine a preliminary energy efficiency optimization strategy combination; the analysis module is also configured to map the preliminary energy efficiency optimization strategy into operating instructions of the super data center operating equipment, and generate a final energy efficiency optimization strategy based on the operating instructions of the super data center operating equipment.

[0034] It is understood that the analysis module dynamically adjusts the preliminary mapping data using the carbon factor mapping rule engine. Technically, the module compares the preliminary mapping data with the preset mapping data in the rule engine and normalizes the comparison results to obtain the accurate value of the preliminary mapping data. The analysis module then determines whether dynamic rule adjustment is necessary based on the relationship between the accurate value and the accurate value of the preset mapping data. If the accurate value of the preliminary mapping data exceeds the accurate value of the preset mapping data, adjustment is triggered; otherwise, no adjustment is required. This method ensures that dynamic adjustment of the carbon factor mapping data is based on scientific and quantitative judgment criteria, improving the accuracy and real-time performance of carbon emissions calculations. Secondly, when the preliminary mapping data requires adjustment, the analysis module further determines the energy type, equipment type, and time period mapping items to be adjusted. Based on the real-time energy type share, energy consumption characteristics, and carbon factor rules, it recalculates the carbon factor value and corresponding weight to form an adjustment parameter set. Based on the adjustment parameter set, the preliminary mapping data is modified to achieve dynamic optimization of the carbon factor mapping, thereby ensuring the accuracy and traceability of carbon emissions calculations. Finally, the analysis module utilizes the layered carbon emissions results and time series operation data to construct a multi-objective optimization model. The technical approach involves establishing an optimization function based on layered carbon emissions and operational data. This optimization algorithm, combined with constraints on equipment operating status and system stability, determines a preliminary energy efficiency optimization strategy. This strategy is then mapped into operational instructions for each device in the hyperscale data center, generating the final energy efficiency optimization strategy. This allows for dynamic regulation of device energy consumption and overall system energy efficiency optimization.

[0035] In a specific example, the analysis module first compares the preliminary mapping data with the preset mapping data in the carbon factor mapping rule engine to obtain a comparison result, and normalizes the comparison result to calculate the accurate value of the preliminary mapping data. Subsequently, the analysis module compares the accurate value of the preliminary mapping data with the accurate value of the preset mapping data to determine whether dynamic rule adjustment is required. When the accurate value of the preliminary mapping data is higher than the accurate value of the preset mapping data, the analysis module determines the energy type, equipment type and time period mapping items to be adjusted based on the preliminary mapping data and the carbon factor mapping rule engine rules. Subsequently, the analysis module combines the real-time energy type proportion, energy consumption characteristics and carbon factor rules to calculate the carbon factor value and the corresponding weight for a second time, and establishes an adjustment parameter set accordingly. Finally, the analysis module dynamically adjusts the preliminary mapping data based on the adjustment parameter set to generate carbon factor mapping data. Secondly, The analysis module builds a multi-objective optimization model based on the layered carbon emission results and time series operation data, and generates an energy efficiency optimization strategy. The specific steps include: Constructing an optimization function: Constructing a multi-objective optimization function based on layered carbon emission data and time series operation data.

[0036] Establish constraints: Establish equipment operating status constraints and system stability constraints based on time series operation data.

[0037] Solve the optimization function: Use the optimization algorithm to solve the optimization function and determine the preliminary energy efficiency optimization strategy combination.

[0038] Mapping to operation instructions: Map the preliminary energy efficiency optimization strategy to the operation instructions of each operating device in the hyper data center, and generate the final energy efficiency optimization strategy.

[0039] Table 1 shows an example of key parameters in energy efficiency optimization strategy generation:

[0040] Through the above embodiments, the analysis module realizes the dynamic adjustment of carbon factor mapping data and the generation of multi-objective energy efficiency optimization strategies based on layered carbon emission results, thereby ensuring the accuracy of carbon emission prediction and optimized control of operating energy efficiency of the super data center.

[0041] It can be seen that by comparing the preliminary mapping data with the preset mapping data in the carbon factor mapping rule engine, combined with normalization and accuracy determination, deviations can be dynamically identified and rules adjusted, thereby ensuring the accuracy of the carbon factor mapping data and improving the precision of carbon emissions accounting for hyperscale data centers. Secondly, when the accuracy of the preliminary mapping data exceeds the accuracy of the preset mapping data, the system performs a secondary calculation based on the real-time energy type ratio, energy consumption characteristics, and carbon factor rules, and generates an adjustment parameter set to achieve dynamic adjustment of the carbon factor mapping. This mechanism can adaptively correct carbon emission data based on actual operating conditions, improving the system's real-time responsiveness and intelligence. Furthermore, by constructing a multi-objective optimization model, combining stratified carbon emission results with time-series operating data, and setting equipment operating status constraints and system stability constraints, the analysis module can generate a preliminary energy efficiency optimization strategy combination and map it into executable operational instructions, achieving efficient operation and energy consumption control of hyperscale data center equipment, thereby reducing energy consumption while maintaining system stability. Finally, through the combination of dynamic rule adjustment and multi-objective optimization strategies, a closed-loop management of carbon emission calculation and energy efficiency optimization of the hyper data center is formed, which not only ensures data accuracy, but also continuously optimizes equipment operation, improves energy utilization efficiency, reduces carbon emission levels, and achieves coordinated optimization of energy efficiency and carbon emissions.

[0042] Specifically, when the central control module determines whether to revise the energy efficiency optimization strategy based on the comparison results, it includes: the central control module is also configured to obtain the deviation indicators of carbon intensity, PUE, and power / electricity according to the comparison results, and extract the supply structure change indicators of energy proportion change, carbon factor difference, and energy supply fluctuation according to the deviation indicators of carbon intensity, PUE, and power / electricity; the central control module is also configured to construct a deviation feature parameter set based on the deviation indicators of carbon intensity, PUE, power / electricity and the supply structure change indicators of energy proportion change, carbon factor difference, and energy supply fluctuation; the central control module is also configured to perform deviation feature parameter set optimization based on the preset SLA Hard threshold verification, obtaining the hard threshold judgment result and the corresponding trigger item list; the central control module is also configured to perform statistical significance detection on the deviation feature parameter set to determine the abnormal sign statistical score; the central control module is also configured to determine the adjustment score based on the deviation feature parameter set, the hard threshold judgment result and the abnormal sign statistical score, and determine whether to revise the energy efficiency optimization strategy based on the relationship between the adjustment score and the preset adjustment score configured in the central control module: when the adjustment score is lower than the preset adjustment score, the central control module determines to revise the energy efficiency optimization strategy; when the adjustment score is higher than or equal to the preset adjustment score, the central control module determines not to revise the energy efficiency optimization strategy.

[0043] Specifically, when the adjustment score is lower than the preset adjustment score, the central control module determines to correct the energy efficiency optimization strategy, including: the central control module is also configured to extract the deviation ratio of each indicator based on the deviation characteristic parameter set; the central control module is also configured to determine the correction coefficient based on the deviation ratio of each indicator and the preset weight factor corresponding to each indicator; the central control module is also configured to generate a correction parameter set based on the correction coefficient, and correct the energy efficiency optimization strategy based on the correction parameter set, wherein the correction parameter set includes cooling strategy parameters, energy call priority parameters and load migration parameters.

[0044] As can be understood, the central control module compares the adjusted operating energy consumption data, energy supply chain data, and environmental data with the stratified carbon emissions results to obtain deviation indicators for carbon intensity, PUE (power efficiency ratio), and power / electricity. Furthermore, it combines supply structure fluctuations, such as changes in energy share, carbon factor differences, and energy supply volatility, to construct a deviation characteristic parameter set. This process quantitatively characterizes operational status deviations, providing foundational data for subsequent decision-making. Secondly, the central control module performs a hard threshold check on the deviation characteristic parameter set based on the preset service level agreement (SLA), obtaining the hard threshold determination results and a list of trigger items. Furthermore, the deviation characteristic parameter set is analyzed through statistical significance testing to generate anomaly indicator statistical scores. This step identifies operational anomalies or deviations from expectations, providing a quantitative basis for policy adjustments. Furthermore, the central control module combines the deviation characteristic parameter set, hard threshold determination results, and anomaly indicator statistical scores to calculate an overall adjustment score and compare it with the preset adjustment score. When the adjustment score is lower than the preset adjustment score, it indicates that there is a significant deviation in the system and the energy efficiency optimization strategy needs to be revised. When the adjustment score is higher than or equal to the preset value, no correction is required, achieving intelligent decision-making. Finally, when it is determined that correction is necessary, the central control module calculates the deviation ratio of each indicator based on the deviation characteristic parameter set, and combines the preset weight factors of each indicator to determine the correction coefficient. The correction coefficient is then used to generate a correction parameter set, including cooling strategy parameters, energy call priority parameters, and load migration parameters. The energy efficiency optimization strategy is then revised based on the correction parameter set, achieving closed-loop optimization of the energy efficiency and carbon emissions of the hyperscale data center.

[0045] In a specific example, the central control module is configured to determine whether to modify the energy efficiency optimization strategy of the hyperscale data center based on the comparison results. Specifically, the central control module first obtains deviation indicators for carbon intensity, power usage effectiveness (PUE), and power / electricity based on the comparison results. Combining changes in energy share, carbon factor differences, and energy supply fluctuations, it extracts supply structure change indicators to construct a deviation characteristic parameter set. The central control module then performs a hard threshold check on the deviation characteristic parameter set based on a preset service level agreement (SLA), obtaining the hard threshold determination results and a corresponding list of trigger items. Simultaneously, the deviation characteristic parameter set is statistically significant, generating an anomaly indicator statistical score. Next, the central control module calculates an overall adjustment score based on the deviation characteristic parameter set, the hard threshold determination results, and the anomaly indicator statistical score, and compares this adjustment score with a preset adjustment score. If the adjustment score is lower than the preset adjustment score, it indicates a significant deviation in system operation and requires adjustment to the energy efficiency optimization strategy. If the adjustment score is higher than or equal to the preset adjustment score, no adjustment is made, achieving dynamic closed-loop control. Furthermore, if the adjusted score falls below the preset adjustment score, the central control module extracts the deviation ratio of each indicator based on the deviation characteristic parameter set and calculates a correction coefficient based on the preset weighting factors corresponding to each indicator. Subsequently, a correction parameter set is generated based on the correction coefficient and used to adjust the energy efficiency optimization strategy. This correction parameter set primarily includes cooling strategy parameters, energy utilization priority parameters, and load migration parameters, thereby achieving dynamic optimization of the hyperscale data center's operating status.

[0046] Table 2 is an example table of the correction parameter set, showing the deviation index, corresponding weight factor and correction parameter

[0047] Through the above steps, the central control module can dynamically modify the energy efficiency optimization strategy, realize real-time optimization and closed-loop management of the energy consumption and carbon emissions of the hyper data center, improve system stability and energy utilization efficiency, and ensure the satisfaction of the service level agreement (SLA).

[0048] As can be seen, by comparing key indicators such as carbon intensity, PUE, and power / electricity before and after the implementation of energy efficiency optimization strategies, and combining this with information on supply structure changes for comprehensive analysis, the central control module can determine the effectiveness of the current energy efficiency strategy in real time. When the adjustment score falls below a preset threshold, the system proactively adjusts the optimization strategy, implementing dynamic closed-loop control of the hyperscale data center's operating status and ensuring that energy efficiency optimization measures consistently align with actual operating conditions. Secondly, by constructing a deviation characteristic parameter set, performing hard threshold verification, and statistical significance testing, the central control module accurately quantifies various deviation indicators and their impact on the energy efficiency strategy. It then generates a correction parameter set based on the indicator deviation ratio and preset weighting factors. This mechanism not only improves the accuracy of energy efficiency strategy adjustments but also enables refined management of key aspects of carbon emissions and energy usage, effectively reducing the carbon footprint. Furthermore, by dynamically adjusting cooling strategies, energy allocation priorities, and load migration parameters, the central control module optimizes energy allocation and load management within the data center while ensuring that service level agreements (SLAs) are met. This not only enhances overall system stability but also improves the efficiency of energy and equipment resource utilization, achieving dual optimization of energy conservation, emission reduction, and operational performance. Finally, through multi-indicator comprehensive evaluation and rule-driven correction strategies, the central control module can independently determine whether the optimization strategy needs to be adjusted and automatically generate correction parameters for execution, reducing manual intervention, improving the intelligence and automation level of super data center operations, and supporting efficient operation and maintenance management of large-scale, highly complex data centers.

[0049] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0050] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0051] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0052] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A dynamic energy efficiency optimization and carbon traceability system for a hyper data center, characterized in that: include: The acquisition module is configured to obtain operating energy consumption data, energy supply chain data, and environmental data of the operating equipment of the hyper data center, perform fusion processing based on the operating energy consumption data, energy supply chain data, and environmental data, and construct time-series operating data of the hyper data center; an analysis module electrically connected to the acquisition module, the analysis module being configured to establish a carbon footprint prediction model based on the time series operation data and determine stratified carbon emission results. The analysis module is further configured to construct a multi-objective optimization model and generate an energy efficiency optimization strategy based on the stratified carbon emission results and the time series operation data; a central control module electrically connected to the analysis module, the central control module being configured to adjust the operating state of the super-data center's operating equipment according to the energy efficiency optimization strategy, and further configured to compare the super-data center's operating energy consumption data, energy supply chain data, and environmental data after the adjusted operating state with the stratified carbon emission results, and determine whether to modify the energy efficiency optimization strategy based on the comparison results; The display module is electrically connected to the acquisition module and the central control module respectively. The display module is configured to display the timing operation data and energy efficiency optimization strategy of the super data center.

2. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 1, characterized in that: The acquisition module integrates and processes the operating energy consumption data, energy supply chain data, and environmental data, and constructs the time series operating data of the hyper data center, including: The acquisition module is further configured to obtain timestamps of the operating energy consumption data, the energy supply chain data, and the environmental data, and unify the timestamps. The acquisition module is further configured to construct a time-synchronized energy consumption data set based on the operating energy consumption data after the unified timestamps; the acquisition module is further configured to construct a time-synchronized energy supply chain data set based on the energy supply chain data after the unified timestamps; the acquisition module is further configured to construct a time-synchronized environmental data set based on the environmental data after the unified timestamps; The acquisition module is further configured to perform data cleaning on the time-synchronized energy consumption dataset, the time-synchronized energy supply chain dataset, and the time-synchronized environment dataset respectively; The acquisition module is further configured to fuse the cleaned synchronized energy consumption dataset, the time-synchronized energy supply chain dataset, and the time-synchronized environment dataset based on a unified time reference, and generate a fused operation dataset; The acquisition module is further configured to perform time series modeling based on the fusion operation data set, and determine time series operation data based on the time series modeling.

3. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 2, characterized in that: The acquisition module performs time series modeling based on the fusion operation data set and determines the time series operation data based on the time series modeling, including: The acquisition module is further configured to sort each data in the fusion running data set according to the timestamp and obtain the initial time series data set; The acquisition module is further configured to fill in missing values ​​or discontinuous data in the initial time series data set to obtain a complete time series data set; The acquisition module is further configured to extract energy consumption characteristics, environmental characteristics, and supply chain characteristics based on the complete time series data set, and establish a feature data set based on the energy consumption characteristics, environmental characteristics, and supply chain characteristics; The acquisition module is further configured to slice the feature dataset according to a preset time window and data granularity, and generate a sliced ​​dataset; The acquisition module is further configured to perform fitting modeling based on the time series slice data, and determine the time series slice data after fitting modeling as time series running data.

4. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 3, characterized in that: The analysis module is configured to establish a carbon footprint prediction model based on the time series operation data, and to determine the stratified carbon emission results, including: The analysis module is further configured to determine the energy type of the energy supply chain based on the supply chain characteristics in the time series operation data and based on the supply chain characteristics; The analysis module is further configured to determine a carbon factor based on the energy consumption characteristics and energy type in the time series operation data, and obtain carbon factor mapping data based on the carbon factor; The analysis module is further configured to determine the carbon emissions of each time segment and each device based on the carbon factor mapping data, and construct a carbon emissions dataset based on the carbon emissions of each time segment and each device; The analysis module is further configured to perform hierarchical aggregation on the carbon emission dataset and generate a hierarchical carbon emission result.

5. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 4, characterized in that: The analysis module is further configured to determine the carbon factor based on the energy consumption characteristics and energy types in the time series operation data, including: The analysis module is further configured to obtain energy type data from the time series operation data based on the energy consumption characteristics and the energy type; The analysis module is further configured to connect to an authoritative carbon factor database in real time based on the energy type data, retrieve and obtain the corresponding benchmark carbon factor, and form a carbon factor data set; The analysis module is further configured to match the energy consumption characteristic data with the carbon factor dataset in a time dimension based on time series data alignment and fusion to obtain preliminary mapping data; The analysis module is further configured to perform dynamic rule adjustment on the preliminary mapping data based on the carbon factor mapping rule engine to generate carbon factor mapping data.

6. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 5, characterized in that: The analysis module is further configured to dynamically adjust rules on the preliminary mapping data based on the carbon factor mapping rule engine, including: The analysis module is further configured to compare the preliminary mapping data with the preset mapping data in the carbon factor mapping rule engine and normalize the comparison results; the analysis module is further configured to determine the accurate value of the preliminary mapping data based on the normalized comparison results; The analysis module is further configured to determine whether to perform dynamic rule adjustment on the preliminary mapping data based on a relationship between the accuracy value of the preliminary mapping data and the accuracy value of the preset mapping data pre-configured by the analysis module: When the accuracy value of the preliminary mapping data is lower than or equal to the preset accuracy value of the mapping data, the analysis module determines not to perform dynamic rule adjustment on the preliminary mapping data; When the accuracy value of the preliminary mapping data is higher than the preset accuracy value of the mapping data, the analysis module determines to perform dynamic rule adjustment on the preliminary mapping data.

7. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 6, characterized in that: When the accuracy value of the preliminary mapping data is higher than the preset accuracy value of the mapping data, the analysis module determines to adjust the dynamic rules of the preliminary mapping data, including: The analysis module is further configured to determine energy type, equipment type, and time period mapping items to be adjusted based on the preliminary mapping data and the carbon factor mapping rule engine rules; The analysis module is further configured to secondary determine the carbon factor value and corresponding weight based on the real-time energy type proportion, energy consumption characteristics and carbon factor rules; The analysis module is further configured to establish an adjustment parameter set based on the secondarily determined carbon factor value and the corresponding weight; The analysis module is further configured to adjust the preliminary mapping data based on the adjustment parameter set.

8. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 7, characterized in that: The analysis module constructs a multi-objective optimization model and generates an energy efficiency optimization strategy based on the layered carbon emission results and the time series operation data, including: The analysis module is further configured to construct a multi-objective optimization function based on the stratified carbon emission results and the time series operation data; The analysis module is further configured to establish device operation state constraints and system stability constraints based on the time series operation data; The analysis module is further configured to solve the optimization function based on the optimization algorithm and determine a preliminary energy efficiency optimization strategy combination; The analysis module is further configured to map the preliminary energy efficiency optimization strategy into an operating instruction of the super data center operating equipment, and generate a final energy efficiency optimization strategy based on the operating instruction of the super data center operating equipment.

9. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 1, characterized in that: The central control module determines whether to modify the energy efficiency optimization strategy based on the comparison results, including: The central control module is further configured to obtain deviation indicators of carbon intensity, PUE, and power / electricity based on the comparison results, and extract supply structure change indicators such as energy share change, carbon factor difference, and energy supply fluctuation based on the deviation indicators of carbon intensity, PUE, and power / electricity. The central control module is further configured to construct a deviation feature parameter set based on the deviation indicators of carbon intensity, PUE, and power / electricity and the supply structure change indicators such as energy share change, carbon factor difference, and energy supply fluctuation. The central control module is also configured to perform hard threshold verification on the deviation characteristic parameter set based on the preset SLA, and obtain the hard threshold determination result and the corresponding trigger item list; The central control module is also configured to perform statistical significance testing on the deviation characteristic parameter set and determine the statistical score of the abnormal sign; The central control module is further configured to determine an adjustment score based on the deviation characteristic parameter set, the hard threshold determination result, and the abnormal flag statistical score, and determine whether to modify the energy efficiency optimization strategy based on the relationship between the adjustment score and the preset adjustment score configured in the central control module: When the adjustment score is lower than the preset adjustment score, the central control module determines to revise the energy efficiency optimization strategy; When the adjustment score is higher than or equal to the preset adjustment score, the central control module determines not to modify the energy efficiency optimization strategy.

10. The hyper data center dynamic energy efficiency optimization and carbon traceability system according to claim 9, characterized in that: When the adjustment score is lower than the preset adjustment score, the central control module determines to revise the energy efficiency optimization strategy, including: The central control module is also configured to extract the deviation ratio of each indicator based on the deviation characteristic parameter set; The central control module is further configured to determine a correction coefficient based on the deviation ratio of each indicator and a preset weight factor corresponding to each indicator; The central control module is also configured to generate a correction parameter set based on the correction coefficient and to correct the energy efficiency optimization strategy based on the correction parameter set, wherein the correction parameter set includes cooling strategy parameters, energy call priority parameters and load migration parameters.

Images