A full-life-cycle-based data center carbon emission optimization method and system

Abstract

The application discloses a kind of based on full life cycle data center carbon emission optimization method and system, belong to data center energy efficiency management technical field.The method includes: the full life cycle of data center is divided into building material acquisition, construction, equipment procurement, operation and maintenance and demolition five stages, and establishes hierarchical carbon footprint accounting mathematical model;Regional emission factor data system and data collection specification are constructed;Based on time-sharing power data and real-time power grid carbon emission factor, dynamic carbon footprint characterization is realized, and carbon emission responsibility apportioning mechanism based on cabinet power quota, IT energy consumption and computing power output is established;According to the accounting result, optimization strategy is generated and feedback adjustment is executed, complete closed-loop management is formed, and standardized carbon footprint accounting report is generated.The application realizes the complete carbon footprint coverage of the full life cycle of data center and dynamic fine carbon emission management, and overcomes the problems of incomplete boundary, lack of dynamic characterization and missing apportioning mechanism in traditional accounting method.

Description

Technical Field

[0001] This invention relates to the field of data center energy efficiency management technology, and more particularly to a method and system for optimizing data center carbon emissions based on the entire life cycle. Background Technology

[0002] As a core infrastructure of the digital economy era, data centers bear the responsibility of storing, computing, and interacting with massive amounts of data. With the rapid development of technologies such as artificial intelligence, cloud computing, and the Internet of Things, the scale and energy consumption of data centers are showing a continuous and rapid growth trend. With the explosive growth of high-performance computing applications such as AI, the power consumption of data centers has surged, and their carbon emissions will also rise accordingly. The green and low-carbon transformation of the data center industry has become a major issue that urgently needs to be addressed, and accurately and comprehensively quantifying the carbon emission levels of data centers and constructing a scientific and reasonable carbon footprint accounting methodology are the fundamental prerequisites for achieving carbon management in the industry.

[0003] Currently, there are some related technological research and patent achievements regarding carbon emissions from data centers. For example, Chinese invention patent document with publication number CN112183907A proposes a method and device for determining the carbon utilization rate of data centers. This scheme uses the ratio of the total carbon dioxide emissions from data centers to the IT load as the carbon utilization rate, which can accurately calculate the carbon utilization efficiency index during the operation phase of data centers, providing a quantitative tool for measuring the carbon efficiency of data center operation. The calculation boundary of this scheme is the carbon emissions from electricity consumption during the data center operation and maintenance phase.

[0004] Chinese invention patent document CN115358859A proposes a method for calculating the energy efficiency and carbon emissions of data centers. This method comprehensively considers three dimensions: resource utilization efficiency, computing power utilization rate, and computing power energy efficiency. It determines the overall energy efficiency of the data center through weighted fusion and determines a carbon emission benchmark value based on the energy efficiency benchmark value, providing a scientific basis for carbon emission quota setting. This method falls under the category of energy efficiency evaluation in the data center operation and maintenance phase.

[0005] Chinese invention patent document CN122092323A proposes a method for optimizing the configuration of hybrid energy storage in data centers, considering carbon emissions throughout the entire lifecycle. It constructs a dynamic PUE model based on environmental thermal balance and a unified matrix model for batch processing loads. A lifecycle carbon emission evaluation model is established, incorporating both implicit carbon emissions from equipment and carbon emissions from electricity generated during operation. A two-layer, multi-objective optimization framework is employed to optimize the capacity configuration of a hybrid energy storage system combining lithium batteries and supercapacitors. This scheme quantifies the implicit carbon emissions from the manufacturing and replacement of energy storage equipment, and its application focuses on capacity configuration optimization for data center energy storage systems.

[0006] In summary, while existing technologies have made progress in certain aspects of data center carbon emissions (such as carbon utilization efficiency measurement during operation, comprehensive energy efficiency evaluation, or energy storage optimization), the following systemic shortcomings still exist: lack of a unified system boundary definition and emission source classification system for the entire life cycle; lack of a regional carbon emission factor data system and data collection standards for the data center industry; insufficient characterization of carbon emissions under the dynamic operating characteristics and shared infrastructure of data centers; and lack of an effective integration path between carbon accounting methodology and practical business applications. Summary of the Invention

[0007] To address the technical problems in existing data center carbon emission accounting methods, such as inconsistent accounting boundaries, incomplete coverage of life cycle stages, lack of emission factor systems, and insufficient dynamic carbon management capabilities, this invention adopts the following technical solution: On the one hand, this application provides a method for optimizing data center carbon emissions based on the entire life cycle, including: Construct a classification system for the entire lifecycle of data centers, including system boundaries and carbon emission sources.

[0008] Based on the physical entity and operational nature of a data center, its complete lifecycle is deconstructed into multiple consecutive stages with clearly defined start and end points. Specifically, these lifecycle stages include: material acquisition, construction, equipment procurement, operation and maintenance, and demolition. The building materials acquisition phase covers carbon emissions from the production process of the main building materials required for data center construction, as well as carbon emissions from the transportation process from the production site to the construction site.

[0009] The construction phase encompasses carbon emissions from fossil fuel consumption by construction machinery and vehicles, carbon emissions from the activities of construction workers, and carbon emissions from electricity consumption during construction.

[0010] The equipment procurement phase covers carbon emissions from the production process, transportation process, on-site installation and commissioning process, and recycling of equipment packaging materials for all equipment required for the data center.

[0011] The operation and maintenance phase covers carbon emissions from fossil fuel combustion, non-fossil fuel carbon emissions, water use carbon emissions, non-carbon dioxide greenhouse gas emissions from refrigerant escaping, carbon emission reductions from waste heat utilization, and carbon emissions from replacing or adding equipment during operation.

[0012] The demolition phase covers carbon emissions from the demolition process when the data center is decommissioned, as well as carbon emissions from the recycling and disposal of solid waste such as construction waste.

[0013] Establish a hierarchical mathematical model and calculation rules for carbon footprint accounting.

[0014] Based on the phased division of the entire lifecycle, a unified calculation path is established for each carbon emission source. The total carbon emissions of a data center throughout its entire lifecycle are calculated using the following formula:

[0015] in, Carbon emissions during the building materials acquisition phase, Carbon emissions during the building construction phase. Carbon emissions during the equipment procurement phase. Carbon emissions during the operation and maintenance phase Carbon emissions during the demolition phase.

[0016] The specific calculation methods for carbon emissions at each stage are as follows: Carbon emissions during the building materials production stage are calculated as the product of building materials consumption and its corresponding carbon emission factor; carbon emissions during the building materials transportation stage are calculated as the product of building materials consumption, transportation distance, and the carbon emission factor of the transportation mode.

[0017] Carbon emissions during the construction phase are calculated by multiplying the actual consumption of various energy sources during construction by the corresponding carbon emission factor.

[0018] Carbon emissions during the equipment production phase should be calculated using data from product carbon footprint reports or environmental product statements verified by a third party; alternatively, the emissions can be calculated by multiplying the amount of raw materials and energy consumed during equipment production by the corresponding emission factor.

[0019] Carbon emissions during the operation and maintenance phase are calculated by subtracting the carbon emission reduction from waste heat utilization from the sum of carbon emissions from fossil fuel combustion, non-fossil fuel combustion, water consumption, and refrigerant emissions. Specifically, refrigerant emissions are calculated by multiplying the number of devices, refrigerant charge, annual leakage rate, and global warming potential; waste heat utilization carbon emission reductions are calculated by multiplying the heat value of steam or hot water sold from the data center by the corresponding emission factor.

[0020] Carbon emissions during the demolition phase are calculated as the sum of carbon emissions from energy consumption during the demolition construction process and carbon emissions from solid waste disposal and recycling.

[0021] Establish a regionalized emission factor data system and data collection standards.

[0022] The system assesses and screens the sources of emission factors for data centers. For carbon emissions from electricity, the average carbon footprint factor of the regional power grid is used; for carbon emissions from fossil fuels of various energy sources, emission factors calculated based on GB / T2589 calorific value data and IPCC default emission coefficients are used; for embodied carbon emissions from various equipment and building materials, factor data from domestic life cycle assessment databases are used first, and cross-validation and supplementation are conducted using authoritative international databases.

[0023] For each carbon emission source, the statistical scope, measurement methods, and data source requirements for its activity level data are clearly defined. The activity level data includes, but is not limited to: equipment list and material usage data (from design drawings and procurement contracts), construction machinery fuel consumption and transportation mileage data (from construction logs and transportation documents), energy consumption data during operation (from energy bills and smart meter records), and refrigerant filling and leakage data (from equipment maintenance records).

[0024] To achieve dynamic carbon footprint characterization and carbon emission responsibility sharing.

[0025] To address the dynamic changes in data center load rate, energy efficiency, and external environmental conditions, a dynamic carbon footprint characterization method based on time-varying data is established. Specifically, based on time-of-use electricity consumption data and real-time / hourly grid carbon emission factors, the dynamic carbon usage efficiency of data centers under different load scenarios is calculated, and a method for measuring carbon usage efficiency based on IT load rate is defined and standardized.

[0026] For scenarios where multiple users share the same physical infrastructure under models such as hybrid cloud and managed services, a carbon emission responsibility sharing mechanism should be established. The sharing mechanism should be based on rack power quotas, actual energy consumption metering of IT equipment, or computing power output ratios to fairly and transparently distribute the total carbon emissions of the data center to each tenant or business unit.

[0027] Carbon footprint accounting results are used to generate carbon emission optimization strategies and implement feedback adjustments.

[0028] Based on carbon footprint accounting results, the main carbon emission sources and emission reduction potential points are identified at each stage of the data center's life cycle, and targeted carbon emission optimization strategies are generated. These optimization strategies include, but are not limited to: low-carbon selection schemes during the building materials and equipment procurement stage, low-carbon construction schemes during the construction stage, energy structure optimization schemes during the operation and maintenance stage, and resource recovery schemes during equipment decommissioning.

[0029] The optimization strategy is translated into executable operational instructions and sent to the data center's equipment control system or operations management system for execution. During execution, operational data is continuously collected, and the actual carbon emission results are compared with the predicted values ​​to determine the effectiveness of the optimization strategy. When the deviation between the actual and predicted values ​​exceeds a preset threshold, the strategy is automatically corrected, forming a closed-loop management mechanism.

[0030] Generate standardized carbon footprint accounting reports.

[0031] Based on the above accounting and optimization results, a data center full lifecycle carbon footprint accounting report is generated according to a unified report template. The report includes at least: an explanation of the accounting boundaries (including lifecycle stage division, physical boundaries and time boundaries), a carbon emission source inventory and activity data summary, carbon emissions and their proportions at each stage, key performance indicators such as carbon use efficiency, and carbon emission optimization strategies and implementation recommendations.

[0032] On the other hand, this application provides a data center carbon emission optimization system based on the entire life cycle, including a data acquisition layer, a data fusion and processing layer, an accounting engine layer, an optimization decision layer, an execution feedback layer, and a visualization reporting layer.

[0033] The data acquisition layer is configured to acquire raw data at each stage of the data center's entire lifecycle, including a building materials and equipment data acquisition unit, a construction data acquisition unit, an operation data acquisition unit, and a decommissioning data acquisition unit.

[0034] The building materials and equipment data acquisition unit is configured to obtain the types, specifications and quantities of building materials and equipment from design drawings, purchase contracts and equipment lists; The construction data acquisition unit is configured to obtain the fuel consumption of construction machinery, material transportation mileage and transportation method from construction logs, fuel purchase records and transportation documents; The operational data acquisition unit is configured to collect data on the data center's power consumption, water consumption, fuel consumption, refrigerant filling, and leakage in real time through sensors such as smart meters, water meters, and heat meters. The decommissioning data acquisition unit is configured to obtain energy consumption and solid waste disposal data during the demolition process from demolition construction records and solid waste disposal documents.

[0035] The data fusion and processing layer is communicatively connected to the data acquisition layer and is configured to perform timestamp unification, data cleaning, format standardization and fusion processing on the acquired multi-source heterogeneous data to construct a time-series basic dataset, including a time synchronization unit, a data cleaning unit, a feature extraction unit and a time-series modeling unit.

[0036] The time synchronization unit is configured to unify data from different sources and with different time granularities to the same time reference. The data cleaning unit is configured to identify and process missing values, outliers, and noisy data in the original data. The feature extraction unit is configured to extract energy consumption features, carbon emission features, and supply chain features from the cleaned data. The time-series modeling unit is configured to construct a time-series running dataset based on the extracted feature data.

[0037] The accounting engine layer is communicatively connected to the data fusion and processing layer and is configured to calculate the carbon emissions of each stage of the data center's entire life cycle based on the basic dataset and according to the hierarchical carbon footprint accounting mathematical model. This includes a life cycle stage accounting unit, an emission factor management unit, a dynamic carbon footprint calculation unit, a responsibility allocation calculation unit, and a carbon emission reduction accounting unit.

[0038] The life cycle stage accounting unit is configured to calculate the carbon emissions of the building material acquisition stage, the construction stage, the equipment procurement stage, the operation and maintenance stage, and the demolition stage, respectively. The emission factor management unit is configured to store, update, and retrieve various types of carbon emission factor data, including electricity carbon footprint factors, fossil fuel emission factors, and carbon factors implied in building materials and equipment. The dynamic carbon footprint calculation unit is configured to calculate dynamic carbon use efficiency based on time-of-use electricity consumption data and real-time grid carbon emission factors. The responsibility allocation calculation unit is configured to allocate the total carbon emissions to each tenant or business unit based on the cabinet power quota, IT equipment energy consumption or computing power output ratio. The carbon emission reduction calculation unit is configured to calculate the carbon emission reduction from green electricity consumption and the carbon emission reduction from waste heat utilization.

[0039] The optimization decision layer is communicatively connected to the accounting engine layer and is configured to generate carbon emission optimization strategies based on the accounting results, including an emission reduction potential analysis unit, a strategy generation unit, and a strategy evaluation unit.

[0040] The emission reduction potential analysis unit is configured to identify the main carbon emission contributions and emission reduction potential of each stage and each emission source based on the accounting results. The strategy generation unit is configured to generate low-carbon procurement plans, low-carbon construction plans, energy structure optimization plans, and resource utilization plans based on the emission reduction potential analysis results. The strategy evaluation unit is configured to quantitatively evaluate and compare the expected emission reduction effects and economic costs of different optimization strategies.

[0041] The execution feedback layer is communicatively connected to the optimization decision layer and is configured to convert optimization strategies into equipment operation instructions and issue them for execution. At the same time, it collects the running data after execution and feeds it back to the accounting engine layer for effect verification and strategy correction. It includes an instruction conversion unit, an execution monitoring unit, a deviation detection unit, and a strategy correction unit.

[0042] The instruction conversion unit is configured to convert the optimization strategy into operation instructions that can be recognized by the equipment control system or the operation management system. The execution monitoring unit is configured to monitor the execution status and effect of the optimization strategy in real time; The deviation detection unit is configured to compare the actual carbon emission result after execution with the expected value to determine whether the deviation exceeds a preset threshold. The strategy correction unit is configured to automatically trigger the correction and adjustment of the optimization strategy when the deviation exceeds a preset threshold.

[0043] The visualization report layer is communicatively connected to the accounting engine layer and the optimization decision layer, and is configured to display carbon emission accounting results and optimization strategies in a graphical interface, and generate standardized carbon footprint accounting reports, including a dashboard display unit, a report generation unit and an access control unit.

[0044] The dashboard display unit is configured to display key indicators such as carbon emissions, carbon emission structure, and carbon use efficiency at each stage of the life cycle in real time in the form of charts. The report generation unit is configured to automatically generate a data center full lifecycle carbon footprint accounting report according to a standardized template. The permission management unit is configured to set differentiated data access and operation permissions for different user roles.

[0045] The beneficial effects of this invention are as follows: This invention incorporates the entire process of data centers, from material acquisition, construction, equipment procurement, operation and maintenance to demolition and disposal, into the accounting system, achieving full carbon footprint coverage of the entire life cycle of data centers; This invention clearly defines the statistical scope, measurement methods and data source requirements for activity level data for each emission source at each stage of the data center's life cycle, and systematically integrates the carbon footprint factor of electricity, the emission factor of fossil fuels and the carbon factor of building materials and equipment, establishing a unified and operable hierarchical accounting framework and regional emission factor system. This invention introduces time-of-use electricity consumption data and real-time grid carbon emission factors to realize dynamic carbon utilization efficiency calculation based on actual load scenarios, transforming carbon management from post-event statistics to process insight, and providing real-time quantitative feedback and decision support for dynamic emission reduction measures such as intelligent operation and maintenance and demand-side response. This invention addresses the common industry practice of multiple users sharing the same physical infrastructure in hybrid cloud and managed services models. It proposes a multi-dimensional responsibility sharing model based on rack power quotas and the actual energy consumption or computing power output ratio of IT equipment. This effectively solves the problem of unclear carbon emission responsibility in the upstream and downstream of the supply chain and provides a methodological basis for carbon labeling and green procurement of digital products. This invention transforms the complex carbon accounting methodology into a standardized technical tool that can be embedded into actual business processes such as data center engineering consulting, design optimization, procurement management, operation and maintenance auditing, and certification rating through standardized data collection templates, step-by-step accounting workflows, and unified report formats, thus achieving a deep integration of methodology and business processes. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of a data center carbon emission optimization method based on the entire life cycle according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the S1 process of a data center carbon emission optimization method based on the entire life cycle according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the S3 process of a data center carbon emission optimization method based on the entire life cycle according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the strategy correction process of a data center carbon emission optimization method based on the entire life cycle according to an embodiment of the present invention; Figure 5 This is an overall architecture diagram of a data center carbon emission optimization system based on the entire life cycle, according to an embodiment of the present invention. Detailed Implementation

[0047] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0048] As a core infrastructure of the digital economy era, data centers are facing increasingly prominent issues regarding energy consumption and carbon emissions. Carbon emissions from data centers persist throughout their entire lifecycle, from material acquisition and construction to equipment procurement, operation and maintenance, and final dismantling. However, existing carbon emission accounting methods generally only focus on electricity emissions during the operation phase, neglecting carbon emissions from upstream equipment manufacturing, construction processes, refrigerant emissions of non-CO2 greenhouse gases, and decommissioning. This incompleteness in accounting boundaries leads to a significant underestimation of the carbon footprint of data centers, making it difficult to provide comprehensive and accurate data support for carbon reduction decisions.

[0049] Example 1 To address the aforementioned issues, this embodiment provides a data center carbon emission optimization method based on the entire lifecycle. This method incorporates the complete lifecycle of the data center into the accounting system, establishing a unified accounting boundary definition, a hierarchical accounting mathematical model, a regionalized emission factor system, and data collection specifications.

[0050] Figure 1 A schematic diagram of a data center carbon emission optimization method based on the entire lifecycle of the present invention is shown below. Please refer to [link / reference]. Figure 1 As shown, the specific steps include the following: S1 constructs a system boundary and carbon emission source classification system for the entire life cycle of data centers, establishing a clear and complete accounting boundary for the carbon footprint accounting of data centers. Figure 2 This diagram illustrates a process flow S1 of a data center carbon emission optimization method based on the entire lifecycle of the present invention. Please refer to [link / reference]. Figure 2 Specifically, it includes the following steps: S11, the entire lifecycle of a data center is divided into five stages: material acquisition, construction, equipment procurement, operation and maintenance, and decommissioning. This division is based on the complete journey of a data center's physical entity from its inception to its operation and eventual decommissioning.

[0051] The building materials acquisition phase encompasses the entire process of obtaining raw materials (including but not limited to steel, concrete, aluminum, and glass) from nature for the main structure of the data center building, processing them into building materials, and transporting these materials from the production site to the construction site. This phase marks the beginning of the data center's life cycle, and its carbon emissions primarily originate from the combustion of fossil fuels and electricity consumption during the building materials production process, as well as fuel consumption during transportation.

[0052] The construction phase encompasses the entire process of data center building construction, including main structure construction, interior decoration, and electromechanical system installation. This includes carbon emissions from the operation of construction machinery and vehicles, personnel activities, and temporary power supply at the construction site. This phase begins when the construction team arrives on site and ends when the data center passes final acceptance and is put into use.

[0053] The equipment procurement phase encompasses the acquisition of all equipment required for the normal operation of the data center. This equipment includes, but is not limited to: IT equipment such as servers, storage devices, and network equipment; power distribution equipment such as uninterruptible power supplies (UPS), transformers, and distribution cabinets; refrigeration equipment such as chillers, precision air conditioners, and cooling towers; and backup power equipment such as diesel generators. Carbon emissions during this phase include the implicit carbon emissions from the equipment's manufacturing process, transportation carbon emissions from the manufacturer to the data center, energy consumption carbon emissions during on-site installation and commissioning, and carbon emissions from the recycling and disposal of equipment packaging materials. The equipment procurement phase begins with equipment selection and the signing of the procurement contract and continues until the equipment is installed, commissioned, and ready for operation.

[0054] The operation and maintenance (O&M) phase covers the entire lifecycle of a data center after its delivery and commissioning, including but not limited to: daily power consumption, fossil fuel consumption (such as periodic testing of diesel generators), water consumption, non-CO2 greenhouse gas emissions from refrigerant leaks, and carbon emissions from the replacement or addition of equipment during operation. The O&M phase begins when the data center officially begins operation and ends when it is decommissioned and ceases operation. The accounting period for carbon emissions during the O&M phase is typically a full calendar year. For data centers operating continuously across years, annual carbon emission statistics and reporting are conducted on a yearly basis.

[0055] The demolition phase encompasses the dismantling and disposal process of data centers after they have reached their designed service life or are decommissioned for other reasons. This includes carbon emissions from various energy consumption during demolition construction, as well as carbon emissions from the transportation and disposal (including landfill, incineration, and recycling) of solid waste such as construction waste and obsolete equipment. The timeframe for this phase begins when the demolition team enters the site and ends when the site is cleared and solid waste is disposed of.

[0056] S12, the physical boundary is calculated based on the data center with clearly defined property rights. The data center includes the main server room, auxiliary area, support area, and administrative area. The main server room is the building space primarily used for the installation and operation of data processing equipment, including server rooms, network rooms, storage rooms, and other functional areas. The auxiliary area is the location for the installation, debugging, maintenance, operation monitoring, and management of electronic information equipment and software, including incoming line rooms, test rooms, central control centers, fire and security control rooms, unpacking areas, spare parts warehouses, printing rooms, and repair rooms. The support area provides power support and security to the main server room and auxiliary area, including power distribution rooms, diesel generator rooms, battery rooms, air conditioning rooms, power stations, uninterruptible power supply (UPS) system rooms, and fire equipment rooms. The administrative area is the location for daily administrative management and customer management of hosted equipment, including offices, lobbies, duty rooms, restrooms, changing rooms, and user workspaces.

[0057] In defining physical boundaries, all building spaces and their internal equipment within the data center's ownership area should be included in the accounting scope. For multi-tenant shared data centers, the carbon emissions from each tenant's equipment electricity consumption should be allocated according to the responsibility-sharing mechanism described in subsequent steps.

[0058] S13, the time boundaries for the building materials acquisition stage, construction stage, equipment procurement stage, and demolition stage are the actual duration of the stage, from the start time to the end time of the stage; the time boundary for the operation and maintenance stage is usually a complete calendar year (i.e., January 1 to December 31). For data centers that are put into operation for the first time, if the first calendar year is less than one year, it will be proportionally calculated based on the actual number of operating days.

[0059] It should be noted that carbon emission accounting is an ongoing process, not a one-time event. For newly built data centers, carbon emissions at each stage should be calculated and transferred to the data center's carbon emission account promptly after the completion of that stage. For data centers already in operation, carbon emission accounting for the operation and maintenance phase should be conducted annually, and carbon emission data from the equipment procurement and dismantling phases should be updated promptly when events such as equipment replacement or expansion occur.

[0060] S14. In the five life cycle stages mentioned above, it is necessary to comprehensively identify and systematically classify the important greenhouse gas emission points in each stage. The carbon emission breakdown for each stage is as follows: Carbon emissions during the building materials acquisition phase include carbon emissions during the building materials production phase, carbon emissions during the building materials transportation phase, and carbon emissions during the building materials packaging recycling and processing phase. Carbon emissions during the construction phase include carbon emissions from the use of fossil fuels during construction, carbon emissions from electricity consumption during construction, and carbon emissions from the transportation and disposal of solid waste generated during construction. Carbon emissions during the equipment procurement phase include carbon emissions during the equipment production phase, carbon emissions during the equipment transportation phase, carbon emissions during the equipment installation phase, and carbon emissions during the equipment packaging material recycling and disposal phase. Carbon emissions during the operation and maintenance phase include carbon emissions from fossil fuel combustion, carbon emissions from electricity consumption, carbon emissions from water use, carbon emissions from refrigerant emissions, carbon emission reductions from waste heat utilization, and carbon emissions from new or replaced equipment. Carbon emissions during the demolition phase include carbon emissions generated during the demolition construction process and carbon emissions generated during the recycling and treatment of solid waste (construction waste, etc.) during the demolition phase.

[0061] By classifying and defining carbon emissions at each stage, a clear boundary foundation is laid for subsequent quantitative calculations.

[0062] S2 establishes a hierarchical mathematical model and calculation rules for carbon footprint accounting. Based on the principles of life cycle assessment, a unified calculation path is established for each carbon emission source identified in S1. The total carbon emissions of the data center throughout its entire life cycle are calculated according to the following formula:

[0063] in, Carbon emissions during the building materials acquisition phase, Carbon emissions during the building construction phase. Carbon emissions during the equipment procurement phase. Carbon emissions during the operation and maintenance phase Carbon emissions during the demolition phase.

[0064] In this embodiment, the unit for carbon emissions is kgCO2e. Specifically, the calculation of carbon emissions at each stage includes the following steps: S21, Carbon emissions during the building materials acquisition phase Calculate according to the following formula:

[0065] in, Carbon emissions during the building materials production stage. Carbon emissions during the transportation of building materials. This refers to the carbon emissions during the packaging material recycling and processing stage of building materials.

[0066] Carbon emissions during building materials production Calculate according to the following formula:

[0067] in, For the first The consumption of major building materials, expressed in units of kg, t, m³, or m². For the first Carbon emission factors of major building materials, expressed in kgCO2e / functional unit, where n is the number of major building materials. Major building materials include, but are not limited to, steel, concrete, cement, aluminum, glass, and insulation materials. For example, in this embodiment, the construction of a data center consumed 5,000 tons of concrete (emission factor 280 kgCO2e / t) and 1,200 tons of steel (emission factor 2,100 kgCO2e / t). Calculate the carbon emissions during the building material production stage. It is 3,920,000 kg CO2e.

[0068] The consumption of building materials is obtained from the design drawings and the list of construction materials. The carbon emission factor of building materials shall be based on the default value provided in Appendix D of the "Standard for Calculation of Carbon Emissions in Buildings" (GB / T 51366). If there is an Environmental Product Declaration (EPD) verified by a third party, the data in the declaration shall be used first.

[0069] Carbon emissions during building material transportation Calculate according to the following formula:

[0070] in, For the first The transportation distance between the production site and the construction site of various building materials, in km. For the first The carbon emission factor per unit weight of building materials transported over a given distance using the transportation method employed, expressed in kgCO2e / (t·km); For example, in this embodiment, the transportation distance for concrete used in the construction of a data center is 100km, and the transportation distance for steel is 500km. Road transportation is used, and the road transportation factor is 0.05kgCO2e / (t·km). Calculate the carbon emissions during the transportation of building materials. It is 55,000 kg CO2e.

[0071] The transportation distance is determined based on the geographical location of the building material supplier and the actual transportation route; the carbon emission factor of the transportation mode adopts different default values ​​according to different transportation modes such as road transportation, rail transportation, and waterway transportation.

[0072] Carbon emissions during the packaging material recycling process Calculate according to the following formula:

[0073] in, This refers to the carbon emissions during the packaging material transportation stage. This refers to carbon emissions during the packaging material disposal stage. Packaging materials include, but are not limited to, wooden pallets, paper boxes, and plastic films. Carbon emissions from packaging material transportation are calculated similarly to those from building material transportation, except that the variables related to "building materials" are replaced with those related to "packaging materials". Carbon emissions from packaging material disposal are calculated based on the emission factors corresponding to different disposal methods (recycling, incineration, landfill), with recycling being able to offset the corresponding carbon emissions.

[0074] S22, In a preferred embodiment of the present invention, the carbon emissions during the building construction phase... The following formula can be used for calculation:

[0075] in, This refers to the carbon emissions generated from the use of fossil fuels during construction. This refers to the carbon emissions generated by electricity consumption during construction. This refers to the carbon emissions generated during the transportation and disposal of solid waste produced during the construction process.

[0076] Carbon emissions from the combustion of fossil fuels during construction Calculate according to the following formula:

[0077] For the first time during the construction process The consumption of various fossil fuels, expressed in kg. For the first The carbon emission factor of various fossil fuels is expressed in kgCO2e / kg. Fossil fuels include, but are not limited to, diesel (used in construction machinery and transport vehicles) and gasoline (used in vehicles). For example, in this embodiment, 8000L of No. 0 diesel oil was consumed during the construction of a data center. The emission factor was approximately 2.718kgCO2e / L, and the carbon emissions generated by the diesel consumption during the construction process were approximately 21680kgCO2e.

[0078] Carbon emission factors of fossil fuels Calculate using the following formula:

[0079] Where is the first The calorific value of fossil fuels is expressed in kJ / kg and is taken from the "General Rules for Calculation of Integrated Energy Consumption" (GB / T2589). For the first The oxidation rate of the fossil fuels was conservatively estimated. For the first The emission coefficients for fossil fuels based on calorific value are taken from the default values ​​in IPCC-2006; Carbon emissions from electricity consumption during construction Calculate according to the following formula:

[0080] in, This represents the electricity consumption during construction, expressed in kWh. The carbon footprint factor for electricity is expressed in kgCO2e / kWh. For example, in this embodiment, the construction of a data center consumes 50,000 kWh of electricity, and the emission factor of the area is 0.5703 kgCO2e / kWh. Calculate the carbon emissions generated by the electricity consumption during the construction process. The figure is 28,515 kg CO2e, using the average carbon footprint factor of electricity in the corresponding region.

[0081] Carbon emissions from solid waste treatment during construction The accounting method is consistent with the building material packaging recycling and processing stage, and the relevant variables are replaced with construction waste-related variables.

[0082] S23, Carbon emissions during the equipment procurement phase Calculate according to the following formula:

[0083] in This refers to carbon emissions during the equipment manufacturing phase. This refers to carbon emissions during the equipment transportation phase. This refers to carbon emissions during the equipment installation phase. This refers to the carbon emissions during the recycling and processing of equipment packaging materials.

[0084] Carbon emissions during equipment production Calculate according to the following formula:

[0085] For the first The number of such devices For the first The carbon footprint factor of this equipment is expressed in kgCO2e / functional unit.

[0086] If carbon footprint factor data for the equipment cannot be obtained, preferably, the carbon footprint factor of the equipment is estimated according to the following formula:

[0087] in, For the first stage of equipment production The unit consumption of raw materials or energy in each unit process To correspond to the emission factors of raw materials or energy, To correspond to the global warming potential of greenhouse gases, This refers to the carbon dioxide equivalent directly emitted during the equipment's production process.

[0088] Preferably, carbon emissions during equipment transportation The calculation method is the same as that for the building materials transportation stage, that is, replacing the variables related to "building materials" with variables related to "equipment", including the transportation distance of the equipment from the manufacturing plant to the data center, the transportation mode and its corresponding carbon emission factor.

[0089] Preferably, carbon emissions during the equipment installation phase The calculation method is the same as that for the construction phase, that is, the variables related to the "construction phase" are replaced with the variables related to the "installation phase", including the energy consumption such as electricity and fuel used during the installation process.

[0090] Preferably, the carbon emissions during the equipment packaging material recycling and processing stage The calculation method is the same as that for the building material packaging recycling stage, that is, replacing the variables related to "building material packaging" with the variables related to "equipment packaging".

[0091] S24, Carbon emissions during the operation and maintenance phase Calculate according to the following formula:

[0092] in, This refers to the carbon emissions generated from the combustion of fossil fuels during the operation and maintenance phase. This refers to the carbon emissions generated by electricity consumption during the operation and maintenance phase. Carbon emissions generated from water use during the operation and maintenance phase. This refers to the carbon emissions generated by refrigerant leakage during the operation and maintenance phase. This refers to the carbon emission reduction from waste heat utilization during the operation and maintenance phase. Carbon emissions generated by adding or replacing equipment during the operation and maintenance phase.

[0093] Preferably, the carbon emissions generated by using fossil fuels during the operation and maintenance phase The calculation method is the same as that in the construction phase, mainly calculating the diesel consumption and carbon emissions during the periodic testing and emergency operation of the diesel generator.

[0094] Preferably, the carbon emissions generated by using non-fossil fuels during the operation and maintenance phase are... Calculate according to the following formula:

[0095] in, This refers to the total electricity purchased from the grid during the operation and maintenance phase. This refers to the carbon footprint factor for electricity. If the data center uses renewable energy (such as self-built photovoltaic power generation or market-purchased green electricity), the carbon footprint factor for its corresponding electricity consumption is determined according to the carbon footprint factor for the corresponding power generation type, and this portion of electricity consumption is not counted twice with electricity purchased from the grid.

[0096] Specifically, for the calculation of carbon emission reductions from green electricity consumption, when calculating carbon emissions from electricity during the operation and maintenance phase, if the data center has purchased green electricity certificates or purchased green electricity through market transactions, the corresponding electricity volume will be deducted from the total electricity purchased from the grid, and its carbon emissions will be calculated according to the carbon footprint factor of the corresponding power generation type; or the difference between the grid purchase carbon emissions and the grid average carbon emissions corresponding to green electricity and the actual carbon emissions of green electricity will be used as the carbon emission reduction.

[0097] Preferably, carbon emissions from water use during the operation and maintenance phase Calculate according to the following formula:

[0098] in, This represents the total water consumption during the operation and maintenance phase, in tons (t). The carbon footprint factor for water use is expressed in kgCO2e / t, reflecting carbon emissions during the acquisition, treatment, and supply of water resources.

[0099] Preferably, the carbon emissions generated during the operation and maintenance phase Calculate according to the following formula:

[0100] in, For the first The number of devices that can produce emissions. For the first The initial refrigerant charge in this type of equipment, expressed in kg. For the first The service life of this type of equipment, in years. For the first The global warming potential of the refrigerant used in this type of equipment.

[0101] It should be noted that, This calculation measures the amount of refrigerant released into the atmosphere annually due to leakage from the equipment. This calculation is based on the assumption of uniform refrigerant leakage throughout the equipment's lifespan. Sources of leakage include, but are not limited to, refrigerant-containing equipment such as chillers, precision air conditioners, and unitary air conditioning units.

[0102] For example, in this embodiment, a data center has 50 precision air conditioners, each filled with 50 kg of refrigerant, with a service life of 10 years and a refrigerant GWP value of 2088. The annual leakage rate is set based on the amortized lifespan, and the carbon emissions generated during the operation and maintenance phase are calculated. It is 522,000 kg CO2e.

[0103] Preferably, the carbon emission reduction generated by waste heat utilization during the operation and maintenance phase Calculate according to the following formula:

[0104] in, The heat sold as steam or hot water to the data center, measured in GJ, is obtained from the readings of the heat meter. The emission factor for steam or hot water, expressed in tCO2e / GJ, characterizes the carbon emissions corresponding to providing the same amount of heat through traditional heating methods. In this embodiment, 0.11 tCO2e / GJ is used. For example, if a data center supplies 10,000 GJ of waste heat, the emission factor is 0.11 tCO2e / GJ, which is equivalent to 110 kgCO2e / GJ. The carbon emission reduction generated by waste heat utilization during the operation and maintenance phase is calculated. It is 1,100,000 kg CO2e.

[0105] If steam is measured in mass units (t), it needs to be converted to heat units using the following formula:

[0106] in, The heat of steam, For the quality of steam, 83.74 represents the enthalpy of steam per kilogram at the corresponding temperature and pressure, and 83.74 represents the enthalpy of feedwater at a temperature of 20°C. The unit is kJ / kg.

[0107] If hot water is measured in mass units (t), it needs to be converted to heat units using the following formula:

[0108] in, For the heat of hot water, For the quality of hot water, 20 represents the temperature of hot water, 20 represents the temperature of water at room temperature, and 4.1868 represents the specific heat of water at room temperature and pressure, in kJ / (kg·℃).

[0109] Preferably, the carbon emissions generated by adding or replacing equipment during the operation and maintenance phase are... The calculation method is consistent with the equipment procurement stage, and is calculated based on the number of newly added or replaced equipment and the actual procurement activities that occur each year.

[0110] Carbon emissions during the S25 dismantling phase Calculate according to the following formula:

[0111] in, This refers to the carbon emissions generated during the demolition and construction process. Carbon emissions generated during the recycling and treatment of solid waste (construction waste, etc.) in the demolition phase.

[0112] Preferably, the carbon emissions generated during the demolition process The calculation method is the same as that for the construction phase, except that the variables related to the "construction phase" are replaced with the variables related to the "demolition phase", including the fuel consumption and electricity consumption of the demolition machinery.

[0113] Preferably, the carbon emissions generated from the recycling and treatment of solid waste during the demolition phase are... The calculation method is the same as that for the building material packaging recycling and processing stage, that is, replacing the variables related to "packaging materials" with the variables related to "demolition solid waste", including the entire process of transportation, classification, recycling and final disposal of construction waste.

[0114] S3, Construct a regionalized emission factor data system and data collection standards. Figure 3 This diagram illustrates the S3 process of a data center carbon emission optimization method based on the entire lifecycle of the present invention. Please refer to [link / reference]. Figure 3 Specifically, it includes the following steps: S31, Systematically assess, screen, and integrate emission factor sources for data centers. Emission factors are categorized by type as follows: The first category is the electricity carbon footprint factor. This corresponds to the carbon emissions generated by electricity consumption during the building construction, equipment installation, operation and maintenance, and demolition phases. For the operation and maintenance, construction, and demolition phases, the average electricity carbon footprint factor for the corresponding region is used. For scenarios involving the use of specific types of electricity generation, the electricity carbon footprint factor for the corresponding type of electricity generation is used for calculation.

[0115] The second category is fossil fuel emission factors. These correspond to carbon emissions from the combustion of fossil fuels such as diesel and gasoline during construction, operation and maintenance (e.g., diesel generators), and demolition phases. Based on the fuel calorific value data provided in the "General Rules for Calculating Integrated Energy Consumption" (GB / T 2589) and the emission coefficients based on calorific value provided in the IPCC-2006 default values, carbon emission factors for various fossil fuels are calculated. .

[0116] The third category is the implicit carbon factor of building materials. During the acquisition stage of building materials, the implicit carbon emissions generated in the production and manufacturing process of major building materials such as steel, concrete, cement, aluminum, and glass are calculated using the default values ​​of building material carbon emission factors provided in Appendix D of the "Building Carbon Emission Calculation Standard" (GB / T51366).

[0117] The fourth category is the implicit carbon factor of equipment. This refers to the implicit carbon emissions generated during the manufacturing process of newly added or replaced equipment (such as servers, chillers, UPS, etc.) during the equipment procurement and operation and maintenance phases. Data from third-party verification product carbon footprint reports or EPDs based on LCA are used. If these are not available, preferably, carbon footprint factor data of similar or related equipment in the domestic Life Cycle Assessment Database (CLCD) can be used, and cross-validated with the international mainstream Life Cycle Assessment (LCA) background database (Ecoinvent).

[0118] The fifth category is the transportation emission factor. This corresponds to the carbon emissions generated during the transportation of building materials and equipment from the production site to the construction site, as well as the transportation of waste and packaging materials to the disposal point. It includes the carbon emission factor per unit weight of transportation distance for road, rail, and water transport. For road, rail, and water transport, the corresponding unit weight, transportation distance, and default carbon emission factor values ​​are used for calculation, respectively.

[0119] Category 6 is the fugitive emission factor. This refers to the non-carbon dioxide greenhouse gas emissions from refrigerants (such as R134a, R410A, R407C, etc.) that escape into the atmosphere due to equipment leaks during the operation and maintenance phase. It is calculated using the corresponding Global Warming Potential (GWP) value provided in the IPCC assessment report.

[0120] The seventh category is water consumption emission factors. This corresponds to carbon emissions generated by water consumption during the operation and maintenance phase of data centers, including carbon emissions from water acquisition, treatment, and supply.

[0121] Category 8 is the emission factor for waste and packaging disposal. This corresponds to carbon emissions from the recycling and treatment of packaging materials, and the disposal of construction waste and solid waste. Different emission factors are used to calculate emissions for different disposal methods (including recycling, incineration, and landfill); recycling can offset corresponding carbon emissions (i.e., a negative value is taken), while incineration and landfill generate positive emissions.

[0122] The ninth category is the waste heat utilization emission reduction factor. This corresponds to the carbon emission factor provided by traditional heating methods, and is used to calculate the carbon emission reduction generated by the utilization of waste heat from the data center. In this embodiment, this factor is set to 0.11tCO2e / GJ.

[0123] S32 specifies the statistical scope, measurement methods, and data source requirements for activity level data for each carbon emission source, as detailed below: Building material consumption data: The statistical scope is the total amount of various building materials actually consumed during the construction of the data center. The unit of measurement is kg, t, m³ or m². The data source is the material list in the design drawings and the material procurement and requisition records during the construction process.

[0124] Equipment quantity and specifications data: The statistics cover the actual quantity and specifications of various types of equipment purchased and installed at each stage of the data center's lifecycle (equipment procurement stage or new equipment added during operation and maintenance stage). The units of measurement are functional units such as units, kW, and kVA. The data sources are procurement contracts, equipment lists, and technical specifications.

[0125] Energy consumption data: The statistical scope is the total amount of fossil fuels and electricity actually consumed in each stage. The unit of measurement for fossil fuels is kg, and the unit of measurement for electricity is kWh. The data sources are fuel purchase records, fuel consumption logs, electricity bills and smart meter records.

[0126] Transportation distance and transportation mode data: The statistical scope is the actual transportation distance and transportation mode used between the production site of building materials or equipment and the construction site. The unit of measurement for transportation distance is km. The data source is the geographical location information of the supply location and the actual transportation route planning.

[0127] Water consumption data: The statistical scope is the total amount of water resources actually consumed during the operation and maintenance phase, and the unit of measurement is t. The data source is water bills and water meter records.

[0128] Refrigerant filling and leakage data: The statistical scope includes the number of equipment containing refrigerant, the original refrigerant filling amount, and the design service life of the equipment. The data sources are equipment nameplate information, equipment technical manuals, and maintenance records.

[0129] Waste heat utilization data: The statistical scope is the heat of steam or hot water supplied by the data center to the outside world, the unit of measurement is GJ, and the data source is the reading of heat metering instruments and sales records.

[0130] S33. To ensure the accuracy and consistency of the accounting results, the following quality control measures are implemented during the collection process of activity level data: Integrity check: Verify that activity level data for all emission sources within the accounting boundary are covered, and identify and record any missing data items.

[0131] Accuracy verification: Cross-validate key activity level data (such as total electricity consumption and total fuel consumption), for example, by comparing electricity bill data with smart meter readings to ensure data consistency and accuracy.

[0132] Data traceability: All activity level data should record its source, collection time, and collection method to ensure data traceability.

[0133] S4. Achieve dynamic carbon footprint characterization and carbon emission responsibility allocation. Addressing the dynamic changes in data center load rate, energy efficiency, and external environmental conditions, this step establishes a dynamic carbon footprint characterization method based on time-varying data, evolving from the traditional "annual static average value" calculation to a "dynamic value under typical load scenarios" or "carbon intensity curve" characterization.

[0134] Specifically, this paper defines and standardizes the calculation method for carbon usage effectiveness (CUE) based on IT load rate. Dynamic carbon usage effectiveness. At any moment The value is calculated according to the following formula:

[0135] in, For a moment Data center power consumption For a moment The carbon footprint factor of electricity For a moment Total power of data center IT equipment. Read data in real time from the smart meter at a set frequency (such as every 15 minutes or every hour); Dynamically calculated based on real-time / hourly power generation structure data released by national or regional power grid dispatching agencies; Obtained in real time from the IT equipment management system.

[0136] The above formula can be used to calculate the dynamic carbon usage efficiency of a data center at any given time, forming a "carbon intensity curve" that reflects the changes in the data center's carbon performance over time. This curve can be statistically analyzed and visualized daily, weekly, and monthly, and used to identify the operational characteristics of the data center during high-carbon and low-carbon periods, providing data support for operational management decisions.

[0137] In a preferred embodiment of the present invention, data center managers can identify periods of high carbon utilization efficiency based on the carbon intensity curve, i.e. periods when a unit of IT load corresponds to higher carbon emissions. During these periods, low-load operation or latency-tolerant tasks can be prioritized and postponed to low-carbon periods. Conversely, during periods of low carbon utilization efficiency, i.e. periods when a unit of IT load corresponds to lower carbon emissions, the computing load can be appropriately increased to make full use of green electricity.

[0138] For scenarios where multiple users share the same physical infrastructure under models such as hybrid cloud and managed services, a carbon emission responsibility sharing mechanism should be established. The core of this sharing mechanism is to allocate the total carbon emissions of the data center to each tenant or business unit according to the principles of fairness, transparency, and operability.

[0139] Specifically, the total carbon emissions of data centers (referring to the total carbon emissions of data centers within a specific accounting period) are allocated according to the following allocation model:

[0140] in, For the first Carbon emissions allocated to each tenant or business unit To calculate the total carbon emissions of the data center during the accounting period, the carbon emissions from electricity consumption, refrigerant leaks, backup generator fuel consumption, and operational activities such as purchased heat and water treatment are included. For the first rack power quota for each tenant Power allocation for the total server racks in the data center. For the first Actual energy consumption of each tenant's IT equipment. Total energy consumption of data center IT equipment. For the first The computing power output of each tenant, For the total computing power output of the data center, ~ Let be the weighting coefficient, satisfying .

[0141] The weighting coefficient is determined based on the data center's service model and contractual agreements. For data centers primarily engaged in rack leasing, the weighting coefficient is increased. The value can be set to 0.6 to 0.8; for data centers that primarily provide computing power services, increase the value. The value can be set to 0.5 to 0.7; for data centers that are billed based on actual electricity consumption, increase the value. The value can be set to 0.7 to 0.9.

[0142] It should be noted that while there is a correlation between the total energy consumption of IT equipment, rack power quota, and computing power output in the above-mentioned cost-sharing model, these three factors reflect the tenant's resource usage characteristics from different dimensions: rack power quota reflects the infrastructure capacity occupied by the tenant, actual IT equipment energy consumption reflects the tenant's actual electricity consumption behavior, and computing power output reflects the tenant's consumption of computing resources. A weighted fusion of these three factors can more comprehensively reflect the tenant's actual contribution to the data center's carbon emissions.

[0143] Through the above-mentioned sharing mechanism, each tenant can clearly understand the carbon emissions corresponding to its business activities, providing transparent basic data for carbon footprint tracing and green procurement in the upstream and downstream of the supply chain.

[0144] S5 generates a carbon emission optimization strategy based on the accounting results and implements feedback adjustments. Specifically, it includes the following steps: S51, based on the calculation results of steps S1 to S4, firstly, identify the items that contribute the most to carbon emissions in each stage of the data center's life cycle and in each emission source, and designate them as key targets for emission reduction. Specifically, by calculating the proportion of carbon emissions in each stage to the total carbon emissions, and the proportion of carbon emissions from each emission source to the carbon emissions in each stage, a carbon emission structure analysis matrix is ​​formed, from which carbon emission hotspots are identified.

[0145] Based on the carbon emission structure analysis results, targeted carbon emission optimization strategies are generated. These optimization strategies include, but are not limited to, the following types: Optimization strategies for the procurement of building materials and equipment: Based on carbon emission factor data of various building materials and equipment, generate a list of preferred low-carbon options; during the design phase, compare the differences in the carbon emissions of different building structure schemes (such as steel structure and concrete structure) throughout their entire life cycle to provide a quantitative basis for the selection of low-carbon schemes.

[0146] Construction phase optimization strategies: Based on data on fuel and electricity consumption of construction machinery, identify high-carbon emission processes during construction and recommend low-carbon construction measures such as replacing diesel machinery with electric construction machinery and optimizing construction transportation routes to reduce transportation distances.

[0147] Energy structure optimization strategy during operation and maintenance phase: Based on the time-of-use electricity carbon footprint factor and data center load curve, it is recommended to reduce non-critical loads during high-carbon periods and increase computing tasks during low-carbon periods; assess the carbon emission levels under different green electricity procurement ratios and generate green electricity procurement ratio optimization suggestions; evaluate the carbon emission reduction effect of energy storage system configuration and generate energy storage capacity configuration suggestions.

[0148] Equipment decommissioning optimization strategy: Based on carbon emission factor data of equipment recycling and treatment, it is recommended to prioritize equipment recycling and reuse rather than landfill disposal, and to quantitatively compare the carbon emission differences of different disposal methods.

[0149] S52 transforms the above optimization strategies into executable operation instructions and sends them to the data center's equipment control system or operations management system for execution. Specifically: For energy structure optimization strategies, the generated specific operation instructions include, but are not limited to: adjusting the set temperature of chiller units, adjusting the frequency of cooling pump inverters, starting or stopping specific server nodes, migrating latency-tolerant computing tasks to low-carbon periods, and adjusting the charging and discharging strategies of energy storage systems. These operation instructions are sent to the corresponding equipment controllers or management systems via standard industrial communication protocols (such as Modbus TCP, BACnet, SNMP, etc.).

[0150] For low-carbon procurement strategies, the generated specific operational instructions include, but are not limited to: setting carbon emission thresholds for building materials and equipment as screening criteria in the procurement system, and adding carbon emission indicators as a basis for scheme comparison in design reviews. These operational instructions are executed through an enterprise resource planning system or a procurement management platform.

[0151] During the optimization strategy execution process, operational data from the data center is continuously collected, and the actual carbon emissions after execution are compared with the carbon emissions predicted when the strategy was generated. Specifically, at the end of each accounting cycle (usually a calendar month or a calendar year), the actual carbon emissions within that cycle are calculated and compared with the carbon emissions predicted based on the optimization strategy before the start of that cycle. The deviation between the actual and predicted values ​​is calculated, and when the deviation exceeds a preset threshold (e.g., ±10%), the strategy correction process is automatically triggered.

[0152] Figure 4 This diagram illustrates a strategy correction process for a data center carbon emission optimization method based on the entire lifecycle of the present invention. Please refer to [link / reference]. Figure 4 Specifically, it includes the following steps: S521, Deviation Source Location: Decompose the actual carbon emissions by stage and emission source, and identify the specific stage and emission source type of the main source of deviation.

[0153] S522, Deviation Cause Analysis: Based on the deviation source location results, analyze the causes of the deviation. Possible causes include, but are not limited to: a large difference between the actual operating load and the predicted load, a large deviation between the actual power carbon footprint factor and the predicted value (such as due to changes in grid dispatch), a difference between the actual energy efficiency and the theoretical energy efficiency of the equipment, and omissions or errors in the data acquisition process.

[0154] S523, Strategy Parameter Adjustment: Based on the analysis of deviation causes, adjust the relevant parameters in the optimization strategy. For example, if the actual load is significantly lower than the predicted load, lower the load reduction target value during high-carbon periods; if the actual electricity carbon footprint factor is significantly higher than the predicted value, raise the green electricity procurement ratio target.

[0155] S524, Strategy Redeployment: The adjusted optimization strategy is converted back into operational instructions and issued for execution, entering the next round of the "execution-monitoring-comparison-correction" cycle.

[0156] Through the above process, a closed-loop management mechanism is formed, upgrading carbon emission management from static one-time accounting to dynamic continuous optimization.

[0157] S6. Generate a standardized carbon footprint accounting report. Based on the accounting and optimization results of steps S1-S5, generate a data center full lifecycle carbon footprint accounting report according to a unified report template. The report shall include at least the following: Accounting boundary description: including life cycle stage division (five stages: building material acquisition, building construction, equipment procurement, operation and maintenance, and demolition), physical boundary (the main server room, auxiliary area, support area, and administrative management area within the data center's property rights) and time boundary (the start and end time of each stage).

[0158] Carbon emission source inventory and activity data summary: The carbon emission source type, activity level data (including data source description), emission factor value (including factor source description), and carbon emission calculation results for each emission source at each stage are listed in tabular form.

[0159] Carbon emissions and percentages at each stage: The carbon emissions at each stage of the life cycle and their percentage of total carbon emissions are displayed in a pie chart or bar chart, along with detailed data on carbon emissions from different sources within each stage.

[0160] Key performance indicators include total carbon emissions over the entire life cycle, carbon intensity per unit IT load (the total carbon emissions over the entire life cycle corresponding to each kW of IT load), annual average CUE and dynamic curve, and the proportion of green electricity used.

[0161] Carbon emission optimization strategies and implementation recommendations: Based on the accounting results, emission reduction recommendations are proposed for each stage, including the expected emission reduction effects of each strategy and the priority ranking of implementation.

[0162] Example 2 Figure 5 This diagram illustrates the overall architecture of a data center carbon emission optimization system based on the entire lifecycle of the present invention. Please refer to [link / reference]. Figure 5 As shown, this embodiment provides a data center carbon emission optimization system based on the entire life cycle, including a data acquisition layer, a data fusion and processing layer, an accounting engine layer, an optimization decision layer, an execution feedback layer, and a visualization reporting layer.

[0163] The data acquisition layer is configured to acquire raw data at each stage of the data center's entire lifecycle; The data fusion and processing layer communicates with the data acquisition layer and is configured to perform timestamp unification, data cleaning, format standardization and fusion processing on the acquired multi-source heterogeneous data to build a time-series basic dataset. The accounting engine layer is connected to the data fusion and processing layer and is configured to calculate the carbon emissions of each stage of the data center's entire life cycle based on the basic dataset and according to the hierarchical carbon footprint accounting mathematical model. Optimize the communication connection between the decision-making layer and the accounting engine layer, and configure it to generate carbon emission optimization strategies based on the accounting results; The execution feedback layer and the optimization decision layer are connected and configured to convert optimization strategies into equipment operation instructions and issue them for execution. At the same time, the execution data is collected and fed back to the accounting engine layer for effect verification and strategy correction. The visualization report layer communicates with the accounting engine layer and the optimization decision layer respectively, and is configured to display carbon emission accounting results and optimization strategies in a graphical interface, and generate standardized carbon footprint accounting reports.

[0164] Preferably, the system runs on an industrial control computer or server, and interacts with various sensors, smart meters, equipment controllers, and management systems in the data center via industrial communication protocols. Communication between different layers of the system can be achieved through internal API interface calls or message queues.

[0165] Specifically, the data acquisition layer includes a building materials and equipment data acquisition unit, a construction data acquisition unit, an operation data acquisition unit, and a decommissioning data acquisition unit.

[0166] Preferably, the building materials and equipment data acquisition unit is configured to obtain information on the types, specifications, and quantities of building materials and equipment from design drawings, procurement contracts, equipment lists, and technical specifications. This unit automatically extracts the names, specifications, and design quantities of various building materials by importing Building Information Modeling (BIM) files or structured data tables from the design phase; and automatically extracts the names, specifications, and quantities of various equipment by importing contract and order data from the procurement management system. For information that cannot be obtained through structured data, this unit provides a manual input interface for data entry personnel to manually input information from paper or electronic documents. Preferably, the construction data acquisition unit is configured to obtain data such as fuel consumption of construction machinery, material transportation mileage and mode, and construction power consumption from construction logs, fuel purchase records, transportation documents, and temporary construction power meters. This unit extracts the operating time and fuel refueling records of construction machinery by importing log data from the construction management system; extracts the transportation distance and mode of building materials and equipment by importing waybill data from the transportation management system; and automatically reads power consumption data during construction by interfacing with the communication interface of the temporary construction power smart meter. Preferably, the operational data acquisition unit is configured to collect real-time data on the data center's power consumption, water consumption, fuel consumption (diesel generator testing and emergency operation), refrigerant filling and leak detection, and waste heat utilization through sensors such as smart meters, water meters, heat meters, temperature sensors, and flow sensors, as well as the Data Center Infrastructure Management System (DCIM). This unit establishes communication connections with the smart meters via standard industrial communication protocols such as Modbus TCP, BACnet, and SNMP, reading metering data from various meters at a set sampling frequency (e.g., every 15 minutes or hour); and obtains the total power of the data center's IT equipment and the load distribution data for each area through API interface with the DCIM system. Preferably, the decommissioning data acquisition unit is configured to acquire energy consumption and solid waste disposal data during the demolition process from demolition construction records and solid waste disposal documents. The data acquisition method of this unit is similar to that of the construction data acquisition unit, achieved by importing logs and document data from the demolition management system.

[0167] Specifically, the data fusion and processing layer includes a time synchronization unit, a data cleaning unit, a feature extraction unit, and a time series modeling unit.

[0168] The time synchronization unit is configured to unify data from different sources and with different time granularities to the same time reference. This unit uses the annual accounting cycle of the data center operation and maintenance phase as the reference time axis, mapping data from the building material acquisition phase, construction phase, equipment procurement phase, and demolition phase to their corresponding positions on this time axis according to their actual start and end times, ensuring that all data is comparable under the same time coordinate system. For operational data with different sampling frequencies (electricity data sampled every 15 minutes and water meter data sampled hourly in this embodiment), this unit uniformly interpolates them into a data sequence with the same time granularity. The data cleaning unit is configured to identify and process missing values, outliers, and noisy data in the original data. For missing values, the unit employs different processing strategies based on the proportion and pattern of missing data: if the proportion of missing values ​​is small (e.g., less than 5%), linear interpolation or adjacent value imputation methods are used; if the proportion of missing values ​​is large (e.g., greater than 5% but less than 20%), seasonal decomposition or substitution with the mean of similar data is used; if the proportion of missing values ​​is greater than 20%, the data item is marked as unusable and the data quality limitation is noted in the report. For outliers, the unit identifies data points deviating from the normal range based on statistical methods (e.g., box plots or Z-scores), and determines whether they are genuine abnormal events (e.g., power spikes caused by sudden load peaks) or measurement errors based on the data source and type, retaining data from genuine abnormal events while discarding measurement error data. The feature extraction unit is configured to extract energy consumption features, carbon emission features, and supply chain features from the cleaned data. Energy consumption features include, but are not limited to: total electricity consumption, consumption of various fossil fuels, and water resource consumption; carbon emission features include, but are not limited to: carbon emissions at each stage, carbon emissions from each emission source, and carbon utilization efficiency values; supply chain features include, but are not limited to: the geographical location of building materials and equipment suppliers, transportation methods and distances, and the proportion of green electricity procurement. The time-series modeling unit is configured to construct a time-series runtime dataset based on the extracted feature data. Specifically, this unit reorganizes the energy consumption and carbon emission characteristics data of the runtime phase at a uniform time granularity (hourly in this embodiment) to form a continuous time-series data sequence for subsequent dynamic carbon footprint calculation and trend analysis.

[0169] Specifically, the accounting engine layer includes a life cycle stage accounting unit, an emission factor management unit, a dynamic carbon footprint calculation unit, a responsibility allocation calculation unit, and a carbon emission reduction accounting unit.

[0170] The lifecycle stage accounting unit is configured to calculate the carbon emissions for the building material acquisition stage, construction stage, equipment procurement stage, operation and maintenance stage, and demolition stage according to step S2 of Embodiment 1. This unit receives a basic dataset from the data fusion and processing layer, calls the corresponding accounting sub-modules based on the carbon emission source type (in this embodiment, this includes a building material production accounting sub-module, an electricity carbon emission accounting sub-module, a refrigerant emission accounting sub-module, etc.), and summarizes the calculation results of each sub-module to generate the carbon emissions for each stage. The emission factor management unit is configured to store, update, and retrieve various carbon emission factor data. Specifically, this unit has a built-in emission factor database that stores data such as electricity carbon footprint factors, fossil fuel emission factors, building material embodied carbon factors, equipment embodied carbon factors, transportation emission factors, and refrigerant GWP values. This unit provides emission factor version management functionality. When national authorities release new electricity carbon footprint factor data or other authoritative sources update emission factor data, the system management interface updates the data and records the version change history. When the accounting engine calls emission factor data, this unit automatically selects the version data that matches the accounting time boundary. The dynamic carbon footprint calculation unit is configured to calculate the dynamic carbon utilization efficiency of the data center based on time-of-use electricity consumption data and real-time grid carbon emission factors. This unit receives real-time electricity consumption data and electricity carbon footprint factor data from the operational data acquisition unit, calculates the CUE value at a set time granularity, and generates a carbon intensity curve. The responsibility allocation calculation unit is configured to allocate the total carbon emissions of the data center to each tenant or business unit based on rack power quotas, IT equipment energy consumption, or computing power output ratios. This unit receives rack power quota data, actual IT equipment energy consumption data, and computing power output data from the operational data acquisition unit, as well as the total carbon emissions of the data center calculated by the accounting engine layer, and outputs the carbon emissions allocated to each tenant. The carbon emission reduction calculation unit is configured to calculate the carbon emission reduction from green electricity consumption (including market-based green electricity purchases, self-built photovoltaic power generation, and green electricity certificates) and the carbon emission reduction from waste heat utilization. Specifically, this unit calculates the carbon emission reduction of the corresponding amount of green electricity (i.e., the carbon emissions calculated based on the grid average carbon footprint factor minus the carbon emissions calculated based on the corresponding power generation type carbon footprint factor) based on green electricity purchase contracts and green electricity certificate transaction records; and calculates the carbon emission reduction from waste heat utilization based on heat meter readings.

[0171] Specifically, the optimization decision-making layer includes an emission reduction potential analysis unit, a strategy generation unit, and a strategy evaluation unit.

[0172] The emission reduction potential analysis unit is configured to identify carbon emission hotspots by calculating the proportion of carbon emissions in each stage and the proportion of carbon emissions from each emission source in the stage, based on the carbon emission data of each stage and each emission source. The unit outputs the identification results in the form of a carbon emission structure matrix, where rows correspond to each life cycle stage, columns correspond to each emission source type, and matrix elements represent the carbon emission amount and proportion of the corresponding stage and emission source. The strategy generation unit is configured to generate low-carbon procurement plans, low-carbon construction plans, energy structure optimization plans, and resource recovery plans based on the identification results of the emission reduction potential analysis unit. Specifically: The low-carbon procurement plan includes: generating a list of preferred low-carbon options based on carbon emission factor data of various building materials and equipment, while meeting technical specifications; comparing the differences in carbon emissions throughout the entire life cycle of different building structure schemes during the design phase, and outputting recommendations for low-carbon scheme selection.

[0173] The low-carbon construction plan includes: based on data on fuel and electricity consumption of construction machinery and carbon emissions from transportation, it generates suggestions for replacing electric construction machinery, optimizing construction transportation routes, and optimizing construction sequence.

[0174] The energy structure optimization plan includes: generating a load reduction plan for high-carbon periods and a load increase plan for low-carbon periods based on the time-of-use electricity carbon footprint factor and data center load curve; generating green electricity procurement ratio optimization suggestions based on carbon emission simulation calculations under different green electricity procurement ratios; and generating energy storage capacity configuration suggestions based on the evaluation of the carbon emission reduction effect of energy storage system configuration schemes.

[0175] Resource recovery solutions include: generating recommendations for priority disposal methods for decommissioned equipment and construction waste based on the carbon emission factors of various equipment and building material recycling methods (such as prioritizing recycling over landfill).

[0176] The strategy evaluation unit is configured to quantitatively evaluate and compare the expected emission reduction effects and economic costs of different optimization strategies. Based on the schemes output by the strategy generation unit, this unit calculates the expected carbon emissions after implementation of each scheme and compares them with the carbon emissions of the original scheme to obtain the expected emission reduction. Simultaneously, based on the implementation costs of each scheme (including equipment investment costs, increases or savings in operating costs, etc.), it calculates the cost-benefit ratio per unit of emission reduction, serving as a quantitative basis for scheme comparison.

[0177] Specifically, the execution feedback layer includes an instruction conversion unit, an execution monitoring unit, a deviation detection unit, and a strategy correction unit.

[0178] The instruction conversion unit is configured to convert optimization strategies into operation instructions that can be recognized by the equipment control system or operation management system. For energy structure optimization strategies, this unit converts load adjustment plans, temperature setpoint adjustments, energy storage charging and discharging strategies, etc., in the strategy into control instructions in Modbus TCP, BACnet, or SNMP protocol formats; for low-carbon procurement strategies, this unit converts procurement target screening conditions, carbon emission thresholds, etc., in the strategy into data screening rules that can be recognized by the procurement management system. The execution monitoring unit is configured to monitor the execution status and effect of the optimization strategy in real time. This unit obtains the execution status of each operation command (including executed, in progress, and failed statuses in this embodiment) through a communication interface with the equipment control system and management system, and continuously collects operational data (such as power consumption, temperature, and equipment operating status) during the strategy execution process. The deviation detection unit is configured to compare the actual carbon emission results after execution with the expected values ​​to determine whether the deviation exceeds a preset threshold. At the end of each accounting cycle, this unit calculates the actual carbon emissions for that cycle and compares them with the predicted carbon emissions when the strategy was generated, calculating the relative deviation. When the absolute value of the relative deviation exceeds a preset threshold (e.g., 10%), the strategy correction process is triggered. The strategy correction unit is configured to automatically trigger the correction and adjustment of the optimization strategy when the deviation exceeds a preset threshold, following the four sub-steps described in S521~S524 of Embodiment 1: deviation source location, deviation cause analysis, strategy parameter adjustment, and strategy redeployment.

[0179] Specifically, the visualization report layer includes a dashboard display unit, a report generation unit, and a permission management unit.

[0180] The dashboard display unit is configured to display key indicators such as carbon emissions, carbon emission structure, and carbon use efficiency at each stage of the life cycle in real time using charts. Specifically, the unit displays the dynamic CUE curve using a line graph, the carbon emission percentage at each stage using a pie chart, the carbon emission comparison of each emission source using a bar chart, and the carbon emission allocation details for each tenant using a table. The report generation unit is configured to automatically generate a data center full lifecycle carbon footprint accounting report according to the standardized template described in step S6 of Embodiment 1. This unit supports online previewing of the report, exporting it to PDF or Word format, and automatically sending it to a designated recipient via email. The access control unit is configured to set differentiated data access and operation permissions for different user roles. Specifically, system administrators have read and write permissions for all data and system configuration permissions; data center operations and management personnel have permissions to view operational data and execute optimization strategies; tenant users only have permissions to view their own carbon emission allocation data; and auditors have read-only permissions for all accounting data.

[0181] In summary, the data center carbon emission optimization system based on the entire lifecycle provided in this embodiment transforms the method steps described in Embodiment 1 into a deployable and runnable technical entity through the coordinated operation of a data acquisition layer, a data fusion and processing layer, an accounting engine layer, an optimization decision layer, an execution feedback layer, and a visualization reporting layer. The system uses the data acquisition layer to automatically acquire multi-source heterogeneous data throughout the entire lifecycle; the accounting engine layer to accurately calculate and dynamically characterize carbon emissions at each stage; the optimization decision layer to generate targeted emission reduction strategies; the execution feedback layer to form a closed-loop adaptive adjustment capability; and the visualization reporting layer to transparently present the accounting results and optimization suggestions. Through the organic collaboration of each layer, an end-to-end automated process is achieved from data acquisition, carbon emission accounting, strategy generation to execution feedback, improving the real-time performance, accuracy, and intelligence of data center carbon management. It can also be flexibly embedded into actual business processes such as data center planning, procurement, construction, operation and maintenance, and certification.

[0182] The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for optimizing carbon emissions from data centers based on their entire lifecycle, characterized in that, Includes the following steps: S1 constructs a classification system for the system boundary and carbon emission sources of the entire life cycle of data centers, dividing the entire life cycle of data centers into five stages: building material acquisition, construction, equipment procurement, operation and maintenance, and demolition, and identifying the carbon emission sources in each stage; S2, Establish a hierarchical mathematical model and calculation rules for carbon footprint accounting; S3 establishes a regionalized emission factor data system and data collection standards, specifying the statistical scope, measurement methods, and data source requirements for calculating carbon emissions at each stage of the data center's life cycle. S4 enables dynamic carbon footprint characterization and carbon emission responsibility allocation. Based on time-of-use electricity consumption data and real-time grid carbon emission factors, it calculates the dynamic carbon usage efficiency of the data center and allocates the total carbon emissions of the data center to each tenant or business unit. S5 generates carbon emission optimization strategies based on the accounting results and performs feedback adjustments. It identifies the main carbon emission sources and emission reduction potential points in each life cycle stage, generates carbon emission optimization strategies and issues them for execution. During the execution process, it continuously collects operational data and compares the actual carbon emission results with the predicted values. When the deviation exceeds the preset threshold, it automatically triggers strategy correction, forming a closed-loop management mechanism. S6 generates standardized carbon footprint accounting reports, producing a data center full lifecycle carbon footprint accounting report according to a unified report template.

2. The data center carbon emission optimization method based on the entire life cycle as described in claim 1, characterized in that, In step S2, the total carbon emissions of the data center throughout its entire lifecycle are calculated using the following formula: in, Carbon emissions during the building materials acquisition phase, Carbon emissions during the building construction phase. Carbon emissions during the equipment procurement phase. Carbon emissions during the operation and maintenance phase Carbon emissions during the dismantling phase.

3. The data center carbon emission optimization method based on the entire life cycle as described in claim 1, characterized in that, Step S4 allocates the total carbon emissions of the data center to each tenant or business unit according to the following allocation model: in, For the first Carbon emissions allocated to each tenant or business unit This represents the total carbon emissions of the data center during the accounting period. For the first rack power quota for each tenant Power allocation for the total server racks in the data center. For the first Actual energy consumption of each tenant's IT equipment. Total energy consumption of data center IT equipment. For the first The computing power output of each tenant, For the total computing power output of the data center, ~ Let be the weighting coefficient, satisfying ; The weighting coefficient , and The value is determined based on the data center's service model and contractual agreement.

4. The data center carbon emission optimization method based on the entire life cycle as described in claim 1, characterized in that, The dynamic carbon utilization efficiency mentioned in step S4 is at time... The value is calculated according to the following formula: in, For a moment Data center power consumption For a moment The carbon footprint factor of electricity For a moment Total power of data center IT equipment.

5. The data center carbon emission optimization method based on the entire life cycle as described in claim 1, characterized in that, The automatic triggering strategy correction process described in S5 is as follows: when the absolute value of the deviation exceeds the preset threshold, the correction process of deviation source location, deviation cause analysis, strategy parameter adjustment and strategy redeployment is executed.

6. The data center carbon emission optimization method based on the entire life cycle as described in claim 1, characterized in that, The report described in step S6 includes a description of the accounting boundaries, a carbon emission source inventory and activity data summary, carbon emissions and their proportions at each stage, key performance indicators, and carbon emission optimization strategies and implementation recommendations.

7. A data center carbon emission optimization system based on the entire life cycle, characterized in that, The entity responsible for implementing the data center carbon emission optimization method based on the entire lifecycle includes: The data acquisition layer is configured to acquire raw data at each stage of the data center's entire lifecycle, including building materials and equipment data acquisition units, construction data acquisition units, operation data acquisition units, and decommissioning data acquisition units. The data fusion and processing layer is communicatively connected to the data acquisition layer and is configured to perform timestamp unification, data cleaning, format standardization and fusion processing on the acquired multi-source heterogeneous data to construct a time-series basic dataset, including a time synchronization unit, a data cleaning unit, a feature extraction unit and a time-series modeling unit. The accounting engine layer is connected to the data fusion and processing layer and is configured to calculate the carbon emissions of each stage of the data center's entire life cycle based on the basic dataset and according to the hierarchical carbon footprint accounting mathematical model. This includes a life cycle stage accounting unit, an emission factor management unit, a dynamic carbon footprint calculation unit, a responsibility allocation calculation unit, and a carbon emission reduction accounting unit. The optimization decision layer is connected to the accounting engine layer and configured to generate carbon emission optimization strategies based on the accounting results, including an emission reduction potential analysis unit, a strategy generation unit, and a strategy evaluation unit. The execution feedback layer is connected to the optimization decision layer and is configured to convert the optimization strategy into equipment operation instructions and issue them for execution. At the same time, it collects the running data after execution and feeds it back to the accounting engine layer for effect verification and strategy correction. It includes an instruction conversion unit, an execution monitoring unit, a deviation detection unit and a strategy correction unit. The visualization report layer communicates with both the accounting engine layer and the optimization decision layer. It is configured to display carbon emission accounting results and optimization strategies in a graphical interface and generate standardized carbon footprint accounting reports, including a dashboard display unit, a report generation unit, and an access control unit.

Images