Low-carbon regulation and control method, system and device for calculation load of data center and storage medium

Abstract

The invention relates to the technical field of low-carbon regulation and control of data centers, finally achieves the purpose of industry carbon neutralization, provides a low-carbon regulation and control method, system and equipment for load calculation of a data center and a storage medium, and solves the problems of low accuracy and low effectiveness of low-carbon regulation and control. According to the invention, the real-time computing power of each enterprise and the real-time carbon price data of the carbon trading market are collected; a digital twinborn model is constructed based on the two types of data, and the power consumption, the cold energy consumption and the carbon emission of the data center can be mapped in real time; simulating the carbon emission change under a preset computing power scheduling strategy by means of a digital twinborn model, and generating virtual environment feature data; designing a double-layer game model: receiving the basic data and the virtual environment characteristic data by an upper-layer unit, and determining a final carbon quota distribution proportion; and the lower-layer unit dynamically adjusts the power supply voltage and frequency of the electric equipment, so that low-carbon regulation and control of the calculation load of the data center are realized, and the accuracy and effectiveness of the low-carbon regulation and control of the calculation load of the data center are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of low-carbon regulation of data centers, and in particular to a low-carbon regulation method, system, device and storage medium for computing load of a data center. BACKGROUND

[0002] With the rapid development of digital economy, data centers, as the core infrastructure supporting computing power, have computing loads that are characterized by dynamic allocation of multiple enterprise subjects and large fluctuation amplitudes of peak values. At the same time, according to relevant policies, data centers need to achieve coordinated regulation of computing load and carbon emissions on the basis of guaranteeing the computing power demand of each enterprise. In the current low-carbon regulation scenario of data center computing load, the core technical requirements are reflected in three aspects: first, real-time enterprise computing power data and carbon price data in the carbon trading market need to be obtained to ensure the timeliness of the regulation basis; second, differentiated allocation of carbon quotas needs to be implemented according to the differences in computing power use of different enterprises to avoid imbalance in computing power supply caused by single regulation; and third, a correlation mechanism between computing load, equipment energy consumption and carbon emissions needs to be established to ensure that the adjustment measures can directly act on the carbon neutralization target, taking into account the regulation accuracy and computing power service quality.

[0003] At present, the targeted solution adopted by the industry for the above technical requirements is a "carbon quota static regulation scheme based on fixed energy consumption standards". This scheme first sets the unit computing power energy consumption standard of each type of electrical equipment according to the historical operation data of the data center; then, the total carbon quota is evenly distributed to each enterprise according to the enterprise contracted computing power, forming a fixed carbon quota; during the regulation process, the actual energy consumption of the equipment corresponding to each enterprise is monitored, and if the energy consumption exceeds the upper limit of the energy consumption corresponding to the fixed carbon quota, the power supply of the equipment corresponding to the enterprise is reduced to control carbon emissions while ensuring the normal use of computing power of enterprises that have not exceeded the quota.

[0004] This existing scheme has obvious defects: first, the allocation of carbon quotas lacks dynamic adaptability, and the quotas are allocated only according to the contracted computing power of enterprises rather than real-time computing power demand, so when the computing power demand of an enterprise temporarily surges, the fixed quota will force the equipment power to be reduced, affecting the stability of computing power services, while when the demand decreases, the quota is idle, reducing the efficiency of carbon resource utilization; second, real-time carbon price data is not associated, and the regulation decision only relies on fixed energy consumption standards, which cannot adjust the regulation strategy according to carbon price fluctuations - when the carbon price is low, the low-cost carbon resources are not fully utilized to ensure computing power; when the carbon price is high, the regulation intensity is not strengthened to reduce carbon costs, resulting in insufficient regulation economy; third, a dynamic mapping relationship between computing power, energy consumption and carbon emissions is not established, and energy consumption is only controlled by single equipment power adjustment, ignoring the synergistic effect of data center cold consumption and power consumption, for example, reducing equipment power may lead to imbalance between cold supply and demand, thereby increasing overall carbon emissions, and real carbon neutralization precision regulation cannot be achieved. SUMMARY

[0005] The application aims to provide a low-carbon regulation method, system, device and storage medium for data center computing load, so as to solve the problem of low precision and effectiveness of low-carbon regulation of data center computing load caused by the lack of dynamic adaptation, economic inefficiency and neglect of cold and electricity collaboration in carbon quota allocation in the prior art.

[0006] To solve the above technical problems, in a first aspect, the application provides a low-carbon regulation method for data center computing load, comprising:

[0007] Collecting real-time computing power data of each enterprise and obtaining real-time carbon price data of a carbon trading market, wherein the real-time computing power data and the real-time carbon price data are used as basic data for low-carbon regulation of data center computing load;

[0008] Based on the real-time computing power data and the real-time carbon price data, a digital twin model is constructed, which maps the power consumption, cooling consumption and carbon emission of the data center in real time;

[0009] Based on the digital twin model, the change in carbon emission under a preset computing power scheduling strategy is simulated to generate virtual environment feature data;

[0010] A double-layer game model is designed, wherein the upper unit of the double-layer game model receives the basic data and the virtual environment feature data, determines the final carbon quota allocation ratio according to the computing power demand of each enterprise, and the lower unit dynamically adjusts the power supply voltage and power frequency of the power-consuming equipment in the data center according to the final carbon quota allocation ratio, so as to realize low-carbon regulation of the data center computing load.

[0011] Optionally, the double-layer game model is designed, wherein the upper unit of the double-layer game model receives the basic data and the virtual environment feature data, determines the final carbon quota allocation ratio according to the computing power demand of each enterprise, and the lower unit dynamically adjusts the power supply voltage and power frequency of the power-consuming equipment in the data center according to the final carbon quota allocation ratio, so as to realize low-carbon regulation of the data center computing load, comprising:

[0012] The upper unit of the double-layer game model receives the basic data and the virtual environment feature data, divides the computing power usage interval of different enterprises according to the computing power demand declared by each enterprise, and generates an initial carbon quota allocation ratio according to the proportion of the computing power usage interval;

[0013] The upper unit compares the initial carbon quota allocation ratio with the carbon emission trend in the virtual environment feature data, adjusts the initial carbon quota allocation ratio of each enterprise according to the comparison result, and forms a final carbon quota allocation ratio;

[0014] The lower layer unit determines a power supply adjustment threshold of the power consumption device according to the final carbon quota allocation ratio;

[0015] Based on the power supply adjustment threshold, the power supply voltage of the power consumption device is adjusted to be within a preset threshold range, and the power consumption frequency is adjusted according to the adjusted power supply voltage;

[0016] Based on the adjusted power supply voltage and the adjusted power consumption frequency, the data center computing load is low-carbon regulated.

[0017] In a second aspect, the application provides a low-carbon regulation system for data center computing load, comprising:

[0018] The acquisition module is configured to acquire real-time computing power data of each enterprise and obtain real-time carbon price data of a carbon trading market, wherein the real-time computing power data and the real-time carbon price data are used as basic data for low-carbon regulation of the data center computing load;

[0019] The construction module is configured to construct a digital twin model based on the real-time computing power data and the real-time carbon price data, wherein the digital twin model is used to map the power consumption, cooling consumption and carbon emission of the data center in real time;

[0020] The generation module is configured to simulate changes in carbon emissions under a preset computing power scheduling strategy based on the digital twin model to generate virtual environment feature data;

[0021] The adjustment module is configured to design a double-layer game model, wherein an upper layer unit of the double-layer game model receives the basic data and the virtual environment feature data, determines a final carbon quota allocation ratio according to the computing power demand of each enterprise, and a lower layer unit dynamically adjusts the power supply voltage and power consumption frequency of the power consumption device in the data center according to the final carbon quota allocation ratio, so as to realize low-carbon regulation of the data center computing load.

[0022] In a third aspect, the application provides an electronic device, comprising:

[0023] A memory is configured to store a computer program;

[0024] A processor is configured to execute the computer program to realize the steps of the low-carbon regulation method for data center computing load according to the first aspect.

[0025] In a fourth aspect, the application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the steps of the low-carbon regulation method for data center computing load according to the first aspect.

[0026] In the present application, a low-carbon regulation method for data center computing load is provided, comprising: collecting real-time computing power data of each enterprise, and simultaneously acquiring real-time carbon price data of the carbon trading market, the real-time computing power data and the real-time carbon price data being used as basic data for low-carbon regulation of the data center computing load; based on the real-time computing power data and the real-time carbon price data, a digital twin model is constructed, the digital twin model real-time mapping the power consumption, cooling consumption and carbon emission of the data center; based on the digital twin model, the change of carbon emission under a preset computing power scheduling strategy is simulated to generate virtual environment feature data; a double-layer game model is designed, the upper unit of the double-layer game model receives the basic data and the virtual environment feature data, determines the final carbon quota allocation ratio according to the computing power demand of each enterprise, and the lower unit dynamically adjusts the power supply voltage and power frequency of the electrical equipment in the data center according to the final carbon quota allocation ratio, so as to realize low-carbon regulation of the data center computing load. The present application has the following advantages: by collecting real-time computing power data of each enterprise, and simultaneously acquiring real-time carbon price data of the carbon trading market, and using the real-time computing power data and the real-time carbon price data as basic data for low-carbon regulation of the data center computing load, a core basis based on real-time working conditions and market carbon price can be provided for subsequent low-carbon regulation, avoiding that the regulation decision deviates from the actual operation and market environment; by constructing a digital twin model based on the real-time computing power data and the real-time carbon price data, and the digital twin model real-time mapping the power consumption, cooling consumption and carbon emission of the data center, a dynamic correlation between computing power, carbon price and data center energy consumption and carbon emission can be established, and the key operation indicators of the data center can be real-time perceived; by simulating the change of carbon emission under a preset computing power scheduling strategy based on the digital twin model to generate virtual environment feature data, the carbon emission trend under different computing power scheduling scenarios can be predicted in advance, providing prediction support for subsequent carbon quota allocation and equipment adjustment; by designing a double-layer game model, the upper unit of the double-layer game model receives the basic data and the virtual environment feature data and determines the final carbon quota allocation ratio according to the computing power demand of each enterprise, and the lower unit dynamically adjusts the power supply voltage and power frequency of the electrical equipment in the data center according to the final carbon quota allocation ratio, the carbon quota allocation and equipment parameter adjustment can be realized in a coordinated manner, and a hierarchical execution path is provided for low-carbon regulation.

[0027] Further, in designing the double-layer game model, the upper layer unit first receives the basic data and the virtual environment feature data, divides the computing power use interval of different enterprises according to the computing power demand declared by each enterprise, generates an initial carbon quota allocation ratio according to the interval proportion, compares and adjusts the initial carbon quota allocation ratio with the carbon emission change trend in the virtual environment feature data to form a final carbon quota allocation ratio; the lower layer unit first associates the final carbon quota allocation ratio with the corresponding power consumption equipment of each enterprise to clarify the carbon quota proportion of each equipment, sorts the equipment according to the proportion from high to low and sets the corresponding reference adjustment coefficient, obtains the current running power of each equipment based on the real-time computing power data, calculates the preliminary voltage adjustment range and the preliminary frequency adjustment range from the running power and the reference adjustment coefficient, limits the preliminary adjustment range to determine the upper limit of voltage fluctuation, the lower limit of voltage fluctuation and the frequency fluctuation interval in combination with the carbon emission change trend in the virtual environment feature data, integrates to form the power supply adjustment threshold of the power consumption equipment, and then adjusts the power supply voltage of the power consumption equipment to the preset threshold range based on the power supply adjustment threshold, and adjusts the power frequency according to the adjusted power supply voltage, finally realizes the low-carbon regulation and control of the data center computing load. Through the dynamic generation and adjustment of the carbon quota allocation ratio by the upper layer unit, the carbon quota can match the real-time computing power demand and the carbon emission trend of the enterprise, solving the adaptability problem of static allocation of carbon quota; through the lower layer unit, the carbon quota is associated with the specific power consumption equipment, the power supply adjustment threshold is calculated and optimized in combination with the computing power data, and then the voltage and frequency of the equipment are accurately adjusted, establishing an accurate correspondence between the carbon quota and the equipment operation parameters, avoiding the imbalance of cold and electricity caused by single adjustment, and at the same time improving the accuracy of equipment adjustment, ensuring the effect of low-carbon regulation and control and the stability of computing power service.

[0028] These aspects or other aspects of the present application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0030] Figure 1 A flowchart of a low-carbon regulation and control method of data center computing load provided by an embodiment of the present application;

[0031] Figure 2 A specific implementation schematic diagram of a low-carbon regulation and control method of data center computing load provided by an embodiment of the present application;

[0032] Figure 3A structural schematic diagram of a low-carbon regulation system for data center computing load provided by an embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to solve the problems of no dynamic adaptation, insufficient economy and ignoring cold and electricity synergy in the allocation of carbon quota in the prior art, an embodiment of the present application provides a low-carbon regulation method for data center computing load, which adopts the following design concept: first, collect the real-time computing power data used by each enterprise, and obtain real-time price data of the carbon trading market, and the two types of data are used together as the basis for regulation; then, using the two types of basic data, a model is built which can correspond to the actual operation of the data center in real time and reflect the real state, and the model can display the power, cold and carbon emissions of the data center in real time; then, using the model, simulate how the carbon emissions will change under different computing power arrangement modes, and obtain data that can reflect the trend of carbon emissions in advance; finally, design a two-layer working regulation model, the upper layer determines the final proportion of carbon quota allocated to each enterprise according to the previous basic data and the simulated carbon emission trend data, combined with the computing power demand of the enterprise, and the lower layer flexibly adjusts the power supply voltage and power frequency of the power consumption equipment in the data center. Through such a method, the carbon quota can be adjusted according to the real-time demand of the enterprise and the trend of carbon emissions, solving the problem of fixed quota; the real-time carbon price can be referred to optimize the regulation strategy, improving the economy; the dynamic correlation between computing power, energy consumption and carbon emissions can be established to avoid problems caused by single adjustment of equipment power, and finally the low-carbon regulation of data center computing load is accurately realized.

[0034] In order to make the personnel in the technical field better understand the present application, the present application will be further described in detail below in combination with the drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0035] The core of the present application is to provide a low-carbon regulation method for data center computing load, and a flowchart of one specific embodiment thereof is shown in Figure 1 The method comprises the following steps:

[0036] S11, collect real-time computing power data of each enterprise, and obtain real-time carbon price data of the carbon trading market, and the real-time computing power data and the real-time carbon price data are used together as the basic data for low-carbon regulation of data center computing load.

[0037] The real-time computing power data is related information generated by each enterprise when using the computing power of the data center at the moment, including the amount of computing power occupied by the enterprise when processing various tasks, the task processing speed and the like; the real-time carbon price data is related information that each unit of carbon emission right corresponds to the transaction price of the carbon trading market at the current time point; the basic data is a data set formed by integrating the collected real-time computing power data and the obtained real-time carbon price data, which is used to provide a basis for subsequent low-carbon regulation of the computing load of the data center, and the finally generated basic data will be the core reference for subsequent model construction and strategy making.

[0038] In the embodiment of the application, first, the computing task processing system of each enterprise in the data center is connected through a data collection tool to collect the computing power usage of each enterprise in real time to obtain real-time computing power data, such as enterprise A, enterprise B and enterprise C in a data center, enterprise A needs to occupy 200 units of computing power per hour to process real-time order generation tasks, enterprise B needs to occupy 150 units of computing power per hour to process real-time file storage tasks, and enterprise C needs to occupy 100 units of computing power per hour to process real-time data statistics tasks, and the computing power occupation of these enterprises per hour is recorded in detail; secondly, through the official information release interface of the carbon trading market, the current carbon price data of the market is obtained in real time within the same hour of collecting the computing power data of the above enterprises, for example, the transaction price of each unit of carbon emission right in the carbon trading market is 10 yuan at this time, and the price data is recorded; finally, the recorded real-time computing power data and real-time carbon price data are arranged in the same data document to serve as the basic data for low-carbon regulation of the computing load of the data center.

[0039] S12, based on the real-time computing power data and the real-time carbon price data, a digital twin model is constructed, and the digital twin model maps the power consumption, cooling consumption and carbon emission of the data center in real time.

[0040] The digital twin model is a model that can correspond to the actual running state of the data center in real time, which can reflect the power consumption, cooling consumption and carbon emission of the data center in the running process like a mirror, wherein the power consumption is the total power consumed by all electrical equipment in the data center when running, the cooling consumption is the total cooling consumed by the data center to maintain the normal running temperature of the equipment, and the carbon emission is the total amount of greenhouse gases such as carbon dioxide generated in the running process of the data center due to energy consumption.

[0041] S13, based on the digital twin model, simulating the change of carbon emission under the preset computing power scheduling strategy to generate virtual environment feature data.

[0042] The virtual environment feature data is a data set formed by simulating the carbon emission changes of the data center under different computing power scheduling strategies. This data set contains carbon emission fluctuation information of each time period under different strategies. The digital twin model is consistent with the concept in S12, which is a model that can real-time map the power consumption, cooling consumption and carbon emission of the data center. The preset computing power scheduling strategy is to allocate computing power to each enterprise in different time periods according to the operation requirements and business characteristics of the data center in advance, such as allocating different amounts of computing power to enterprises during peak and trough periods of business. The finally generated virtual environment feature data will be an important reference for subsequent determination of carbon quota allocation ratio, helping to develop more reasonable regulation and control strategies.

[0043] S14, a double-layer game model is designed. The upper unit of the double-layer game model receives basic data and virtual environment feature data, determines the final carbon quota allocation ratio according to the computing power demand of each enterprise, and the lower unit dynamically adjusts the power supply voltage and power frequency of the power consumption equipment in the data center according to the final carbon quota allocation ratio, so as to realize low-carbon regulation and control of the computing load of the data center, and further realize carbon neutral regulation and control.

[0044] The double-layer game model is a regulation and control model that works in two layers. The upper unit is mainly responsible for receiving relevant data and determining the final ratio of carbon quota allocation to each enterprise. This ratio is the final carbon quota allocation ratio. The lower unit is mainly responsible for adjusting the power supply voltage and power frequency of the power consumption equipment in the data center according to the final carbon quota allocation ratio. The power supply voltage is the voltage value used by the power consumption equipment during operation, and the power frequency is the frequency value of the current during operation of the power consumption equipment.

[0045] In the embodiment of the present application, first, the upper unit of the double-layer game model is designed to be able to receive the basic data of S11 and the virtual environment feature data of S13, and then the upper unit calculates the proportion of the computing power of each enterprise in the total computing power according to the computing power demand of each enterprise. The calculation process is that the total computing power is equal to 220+140+110, the result is 470 units, the computing power proportion of enterprise A is 220÷470≈44%, the computing power proportion of enterprise B is 140÷470≈29.8%, and the computing power proportion of enterprise C is 110÷470≈23.4%, and the initial carbon quota allocation proportion is generated according to the proportion; then the upper unit compares the initial carbon quota allocation proportion with the carbon emission change in the virtual environment feature data, for example, it is found that the carbon emission of the data center will exceed the standard during the business peak period if the carbon quota is allocated according to the initial proportion, so the initial proportion is adjusted, the proportion of enterprise A is reduced to 42%, the proportion of enterprise B is increased to 30%, and the proportion of enterprise C is increased to 28%, forming the final carbon quota allocation proportion; secondly, the lower unit of the double-layer game model is designed to receive the final carbon quota allocation proportion determined by the upper unit, first, the proportion is associated with the corresponding power equipment of each enterprise, for example, the carbon quota of 42% of enterprise A corresponds to server group 1, the carbon quota of 30% of enterprise B corresponds to server group 2, and the carbon quota of 28% of enterprise C corresponds to server group 3, and the carbon quota proportion of each device group is determined; then the device groups are sorted from high to low according to the carbon quota proportion, the sorting result is that server group 1 is greater than server group 2 which is greater than server group 3 (28%), and the corresponding reference adjustment coefficient is set for each device group after sorting, for example, server group 1 is 1.2, server group 2 is 1.0, and server group 3 is 0.8; then the current running power of each device group is obtained by using the real-time computing power data of S11, for example, the running power of server group 1 is 2000 watts, the running power of server group 2 is 1500 watts, and the running power of server group 3 is 1000 watts, the running power of each device group is multiplied by the corresponding reference adjustment coefficient to obtain the basic adjustment value, the calculation process is that the basic adjustment value of server group 1 is 2000×1.2=2400, the basic adjustment value of server group 2 is 1500×1.0=1500, and the basic adjustment value of server group 3 is 1000×0.8=800, and the preliminary voltage adjustment range and the preliminary frequency adjustment range are obtained according to the basic adjustment value; then according to the carbon emission change trend in the virtual environment feature data, the preliminary voltage adjustment range and the preliminary frequency adjustment range are subjected to amplitude limiting processing, for example, the upper limit of the voltage of server group 1 is reduced to 225 volts, and the upper limit of the frequency is reduced to 50.8 hertz, the upper limit of the voltage of server group 2 is reduced to 215 volts, and the upper limit of the frequency is reduced to 50.3 hertz, and the lower limit of the voltage of server group 3 is increased to 195 volts, and the lower limit of the frequency is increased to 49.8 Hz, the upper limit of voltage fluctuation, the lower limit of voltage fluctuation and the frequency fluctuation interval of each device group are determined, and the power supply adjustment threshold of each device group is determined; finally, the lower unit adjusts the power supply voltage of each device group to the preset threshold range according to the power supply adjustment threshold, for example, server group 1 is adjusted to 220 volts, server group 2 is adjusted to 210 volts, and server group 3 is adjusted to 200 volts, and then the power frequency is adjusted according to the adjusted power supply voltage, for example, server group 1 is adjusted to 50.5 Hz, server group 2 is adjusted to 50.2 Hz, and server group 3 is adjusted to 50 Hz, the energy consumption and carbon emission of each device group are controlled through the adjusted power supply voltage and power frequency, and finally the low-carbon regulation and control of the data center computing load is realized.

[0046] Figure 2 The specific implementation diagram of the low-carbon regulation and control method of the data center computing load provided by the embodiment of the application is similar to the process in Figure 1 , which will not be repeated here, based on Figure 2The method flow provided, the present application provides the following specific examples: when a certain data center carries out low-carbon regulation and control of computing load, first, execute S11 step: through the deployed computing power monitoring software, automatically collect the real-time computing power data of enterprises A, B and C every 15 minutes, at a certain time of collection, enterprise A increases the order processing volume, and the computing power occupation increases from 200 units to 220 units, enterprise B reduces the file storage volume, and the computing power occupation decreases from 150 units to 140 units, and the computing power of enterprise C is stable at 110 units; at the same time, through data connection with the carbon trading platform, real-time carbon price data of 11 yuan per unit at the same time point is obtained, and these data are summarized to a regulation and control data folder as basic data. Then execute S12 step: associate the real-time computing power data in the basic data with the equipment, determine that A corresponds to server group 1 of 3 servers, B corresponds to server group 2 of 2 servers, and C corresponds to server group 3 of 1 server, and test finds that server group 1 increases 10 degrees of electricity per 20 units of computing power, so when A is 220 units of computing power, the electricity consumption calculation process of server group 1 is 110 degrees; combined with the real-time carbon price of 11 yuan, it is determined that 0.45 units of carbon emissions are generated per degree of electricity, and 4 units of cold are required per 10 degrees of electricity, and then it is calculated that the cold of server group 1 is 110 / 10*4=44 units, and the carbon emissions are 53.3 units, and the data of server group 2 and server group 3 are calculated in the same way, and are integrated into a digital twin model. Then execute S13 step: combined with the business characteristics of enterprises (A peak in the morning, B peak in the afternoon, C no peak), set two computing power scheduling strategies, strategy one is A 230 units, B 160 units and C 120 units from 9 to 17 o'clock, and A 180 units, B 130 units and C 90 units in the remaining period; input strategy one into the digital twin model, simulate that A 230 units of computing power from 9 to 17 o'clock corresponds to 115 degrees of electricity consumption of server group 1 ((230-180) / 20*10+90), 46 units of cold and 59.9 units of carbon emissions, B 160 units corresponds to 80 degrees of electricity consumption of server group 2, 32 units of cold and 42.4 units of carbon emissions, and C 120 units corresponds to 60 degrees of electricity consumption of server group 3, 24 units of cold and 31.8 units of carbon emissions, the total carbon emissions in this period are 134.1 units, and the total carbon emissions in the remaining period are 106.3 units; simulate the data of strategy two in the same way, and arrange the virtual environment characteristic data. Finally, execute S14 step: design a double-layer game model, the upper unit receives the basic data and the virtual environment characteristic data, calculates the initial carbon quota proportion, adjusts to A 42%, B 30% and C 28% as the final proportion after comparing the virtual data; the lower unit associates the proportion with the equipment group, sets a reference adjustment coefficient according to the order, obtains the current power of the equipment group, calculates the basic adjustment value, determines the preliminary adjustment range, limits the amplitude combined with the carbon emission trend to obtain the power supply adjustment threshold, finally adjusts A group to 220 volts and 50.6 hertz, B group to 210 volts and 50.3 hertz, and C group to 205 volts and 50.2 hertz, to realize low-carbon regulation and control.

[0047] By performing S11-S14, the embodiment of the application collects enterprise real-time computing power data and carbon trading market real-time carbon price data through S11 step and integrates them into basic data, which provides a basis for subsequent regulation that fits actual operation and market dynamics, avoiding the regulation deviating from demand due to outdated data; S12 step constructs a digital twin model based on the basic data, which maps power, cold consumption and carbon emissions in real time, allowing staff to intuitively understand the impact of computing power changes on energy consumption and carbon emissions, solving the problem of being unable to clearly perceive the internal operation state of the data center in the past; S13 step simulates the change of carbon emissions under different computing power scheduling strategies through the digital twin model and generates virtual environment feature data, which predicts the trend of carbon emissions in advance and provides forward-looking reference for carbon quota allocation, avoiding the situation of exceeding carbon emissions or insufficient computing power after quota allocation; S14 step designs a double-layer game model, the upper unit dynamically adjusts the carbon quota allocation ratio, solving the problem of poor adaptability of traditional fixed quota, and the lower unit adjusts the voltage and frequency of equipment based on the quota, avoiding the imbalance of cold and electricity caused by single power adjustment, ensuring the computing power demand of the enterprise while effectively controlling carbon emissions. The synergistic effect of each step ultimately realizes the precision, flexibility and forward-looking of low-carbon regulation of data center computing load, ensuring that the regulation work can match the real-time demand of the enterprise and meet the carbon neutralization target and market carbon price dynamics.

[0048] In a possible embodiment, S14, a double-layer game model is designed, the upper unit of the double-layer game model receives the basic data and the virtual environment feature data, determines the final carbon quota allocation ratio according to the computing power demand of each enterprise, and the lower unit dynamically adjusts the power supply voltage and power frequency of the power consumption equipment in the data center according to the final carbon quota allocation ratio, to realize the low-carbon regulation of the data center computing load, including:

[0049] Step 141, the upper unit of the double-layer game model receives the basic data and the virtual environment feature data, divides the computing power use interval of different enterprises according to the computing power demand declared by each enterprise, and generates an initial carbon quota allocation ratio according to the proportion of the computing power use interval.

[0050] The upper unit of the double-layer game model is a module responsible for receiving relevant data and preliminarily determining the carbon quota allocation ratio, mainly used for processing information related to carbon quota allocation; the basic data is a collection of real-time computing power data of each enterprise and real-time carbon price data of the carbon trading market collected previously, used to provide a basis for quota allocation; the virtual environment feature data is the carbon emission change related data obtained by simulating different computing power scheduling strategies through the digital twin model previously, used to assist in judging the rationality of quota allocation; the computing power demand declared by the enterprise is the computing power usage amount application proposed by the enterprise according to its own business needs; the computing power usage interval is the exclusive computing power usage range of each enterprise according to the computing power demand declared by the enterprise; the initial carbon quota allocation ratio is the proportion of the carbon quota that each enterprise should obtain preliminarily determined by the upper unit according to the proportion of the computing power usage interval of each enterprise in the total computing power usage interval of all enterprises, which is the basis for subsequent adjustment of carbon quota, and the initial carbon quota allocation ratio generated will be used as the initial basis for further adjustment by the upper unit.

[0051] Step 142, the upper unit compares the initial carbon quota allocation ratio with the carbon emission change trend in the virtual environment feature data, adjusts the initial carbon quota allocation ratio of each enterprise according to the comparison result, and forms the final carbon quota allocation ratio.

[0052] Wherein, the carbon emission change trend in the virtual environment feature data is the trend of carbon emission of the data center under different computing power scheduling strategies recorded in the data, such as the situation that carbon emission rises in peak business period and falls in trough period; the final carbon quota allocation ratio is the determined quota ratio obtained by adjusting the initial carbon quota allocation ratio after comparing with the carbon emission change trend by the upper unit, which will be used as the direct basis for adjusting the power consumption equipment parameters by the lower unit, to ensure that the subsequent equipment adjustment can meet the carbon neutralization target.

[0053] Step 143, the lower unit determines the power supply adjustment threshold of the power consumption equipment according to the final carbon quota allocation ratio.

[0054] The lower unit is a module responsible for adjusting the parameters of the power consumption equipment according to the carbon quota proportion in the double-layer game model, mainly processing adjustment operations related to power consumption equipment; the power consumption equipment is equipment in the data center that provides computing power support for enterprises, such as a server group composed of multiple servers, and each enterprise corresponds to dedicated power consumption equipment; the reference adjustment coefficient is a coefficient set by the lower unit for calculating the adjustment parameters of the equipment according to the carbon quota proportion corresponding to the power consumption equipment, and the coefficient is usually different for different proportions; the operating power is the power value of the power consumption equipment when it is currently working, which can be obtained from the basic data; the preliminary voltage adjustment range and the preliminary frequency adjustment range are the initial adjustable ranges of the voltage and frequency of the equipment calculated by the lower unit based on the operating power and the reference adjustment coefficient; the power supply adjustment threshold is the final adjustable range of the power supply voltage and power consumption frequency of the power consumption equipment after the lower unit limits the preliminary adjustment range, including the upper limit of voltage fluctuation, the lower limit of voltage fluctuation, and the frequency fluctuation interval, which is used to regulate the subsequent adjustment of equipment parameters.

[0055] Step 144, based on the power supply adjustment threshold, adjust the power supply voltage of the power consumption equipment to a preset threshold range, and adjust the power consumption frequency according to the adjusted power supply voltage.

[0056] Wherein, the power supply voltage is the voltage value required when the power consumption equipment is running, and the preset threshold range is the range between the upper limit of voltage fluctuation and the lower limit of voltage fluctuation in the power supply adjustment threshold; the power consumption frequency is the frequency value of the current when the power consumption equipment is running, and its adjustment needs to be determined according to the adjusted power supply voltage to ensure that the adjusted frequency and voltage match each other and ensure stable operation of the equipment. The final adjusted power supply voltage and power consumption frequency will be used as the parameters for the operation of the equipment.

[0057] Step 145, based on the adjusted power supply voltage and the adjusted power consumption frequency, low-carbon regulation and control of the data center computing load.

[0058] Wherein, the data center computing load is the total amount of computing tasks undertaken by all power consumption equipment in the data center when it is running, which is related to the voltage and frequency of the equipment. When the voltage and frequency are matched, the computing load can better meet the needs of enterprises; low-carbon regulation and control is a process of balancing carbon emissions and emission reduction measures in the data center by controlling equipment operating parameters, ultimately achieving carbon emissions that meet target requirements. The adjusted power supply voltage and power consumption frequency are key parameters for achieving this regulation and control.

[0059] The application provides the following specific examples: first, step 141 is performed: the upper unit of the double-layer game model receives the basic data containing the declaration of 180-230 units of computing power of enterprise A, the declaration of 130-180 units of computing power of enterprise B, and the declaration of 80-130 units of computing power of enterprise C, and the real-time carbon price data of 11 yuan per unit, and simultaneously receives the virtual environment characteristic data containing the rising trend of carbon emissions during the business peak period; the computing power use interval is divided according to the declaration demand, enterprise A is 180-230 units, enterprise B is 130-180 units, and enterprise C is 80-130 units, the interval sum is 50+50+50=150 units, and the initial carbon quota allocation proportion of enterprise A, B and C is generated according to the proportion 50 / 150≈33.3%. Then step 142 is performed: the upper unit calls the initial proportion, extracts the trend of “allocating according to the initial proportion during 9-17 of workday (enterprise A peak period), A corresponding device carbon emission exceeds the standard, B and C quota remains” from the virtual environment characteristic data; after comparison, the proportion is adjusted, A is reduced to 28%, B is increased to 36%, and C is increased to 36%, to form the final carbon quota allocation proportion. Then step 143 is performed: the lower unit receives the final proportion, and associates enterprise A corresponding server group 1 (28%), enterprise B corresponding server group 2 (36%), and enterprise C corresponding server group 3 (36%); the proportion is sorted as server group 2=server group 3>server group 1, the group 2 and 3 reference adjustment coefficient is set to 1.1, and the group 1 is 0.9; the running power is obtained from the basic data, the basic adjustment value is calculated, the preliminary voltage range and the preliminary frequency range are obtained; the carbon emission trend is limited, the upper limit of the voltage of group 2 and 3 is reduced to 230 volts, the upper limit of the frequency is reduced to 51 hertz, the lower limit of the voltage of group 1 is increased to 200 volts, and the lower limit of the frequency is increased to 49.9 hertz, to form the power supply adjustment threshold. Then step 144 is performed: the lower unit calls the threshold, and adjusts the voltage of group 1 to 210 volts and the voltage of group 2 and 3 to 215 volts; according to the voltage adjustment frequency, the frequency of group 1 is 50 hertz, and the frequency of group 2 and 3 is 50.3 hertz. Finally, step 145 is performed: each server group runs according to the adjusted parameters, group 1 meets the computing power demand of A at 210 volts and 50 hertz and has low energy consumption and low emission, group 2 and 3 match the demand of B and C at 215 volts and 50.3 hertz and have no invalid energy consumption, finally the data center computing load flexibly adapts to the enterprise demand, the carbon emission is controlled within the target range, and low-carbon regulation is realized.

[0060] By performing steps 141-145, the embodiment of the application receives basic data and virtual environment feature data through step 141, divides the computing power interval according to the enterprise reporting demand and generates the initial carbon quota ratio, ensures that the initial quota is related to the actual demand of the enterprise, and avoids blind allocation; step 142 compares and adjusts the initial ratio with the carbon emission trend to form the final quota ratio adapted to the carbon neutralization target, solving the problem of carbon emission exceeding the standard or quota waste caused by the initial quota; step 143 associates the final ratio with the power equipment, combines the running power and the benchmark adjustment coefficient to calculate and limit the power supply adjustment threshold, so that the adjustment range of the equipment adapts to the carbon quota target and avoids deviation in the adjustment direction; step 144 adjusts the voltage and frequency according to the threshold to ensure that the equipment parameters are adapted and stably operated, providing a reliable equipment basis for regulation and control; step 145 makes the equipment run according to the adjusted parameters to realize the coordinated control of the calculated load and carbon emission. The above steps work together to not only guarantee the computing power demand of the enterprise, but also precisely control the carbon emission of the data center, effectively solving the problems of poor adaptability of traditional quota allocation, blind equipment adjustment, and out-of-control carbon emission, and finally realizing the precision, stability, and rationality of low-carbon regulation and control of the computing load of the data center.

[0061] In one possible embodiment, step 143, the lower unit determines the power supply adjustment threshold of the power equipment according to the final carbon quota allocation ratio, including:

[0062] a1, the lower unit associates the final carbon quota allocation ratio with the power equipment corresponding to each enterprise, and determines the carbon quota proportion corresponding to each power equipment.

[0063] Wherein, the lower unit is a module responsible for processing power equipment adjustment related operations according to the final carbon quota allocation ratio in the double-layer game model; the final carbon quota allocation ratio is the carbon quota proportion that each enterprise should obtain determined by the upper unit in combination with the enterprise computing power demand and carbon emission trend; the power equipment corresponding to each enterprise is the equipment in the data center that provides computing power support for each enterprise, such as a server group composed of multiple servers; the carbon quota proportion corresponding to each power equipment is the proportion of the final carbon quota allocation ratio associated with the power equipment, which is the key basis for determining the equipment adjustment parameters, and finally generates the carbon quota proportion corresponding to each power equipment through the association operation.

[0064] a2, sort each power equipment according to the carbon quota proportion from high to low, and set the corresponding benchmark adjustment coefficient for each sorted power equipment.

[0065] Wherein, the order from high to low is arranged in descending order according to the proportion of the carbon quota of each device, and if the proportion is the same, it is arranged in parallel; the benchmark adjustment coefficient is a coefficient set according to the order of the proportion of the carbon quota of each device, which is used to calculate the adjustment range of the device subsequently, and generally the higher the proportion, the larger the coefficient, so as to match the adjustment needs of different devices, and finally generate the benchmark adjustment coefficient corresponding to each device.

[0066] a3, based on the real-time computing power data, the current running power of each electrical equipment is obtained, and the running power is calculated with the benchmark adjustment coefficient to obtain the preliminary voltage adjustment range and the preliminary frequency adjustment range.

[0067] Wherein, the real-time computing power data is the data collected previously, reflecting the current computing power usage of each enterprise, which contains the running load information of the electrical equipment, from which the running power of the equipment can be extracted; the current running power of each electrical equipment is the power value of the equipment in the current working state, which can reflect the actual load size of the equipment; the preliminary voltage adjustment range is the initial range of the adjustable voltage of the equipment calculated by the running power of the equipment and the benchmark adjustment coefficient; the preliminary frequency adjustment range is also the initial range of the adjustable frequency of the equipment calculated by the above calculation, which together serves as the basis for the subsequent optimization adjustment range, and finally generates the preliminary voltage and frequency adjustment range of each device.

[0068] a4, according to the carbon emission change trend in the virtual environment characteristic data, the preliminary voltage adjustment range and the preliminary frequency adjustment range are respectively subjected to limiting amplitude processing to determine the upper limit of voltage fluctuation, the lower limit of voltage fluctuation and the frequency fluctuation interval.

[0069] Wherein, the virtual environment characteristic data is the data containing carbon emission change information obtained by simulating different computing power scheduling scenarios through the digital twin model previously; the carbon emission change trend is the trend of carbon emission change with time or computing power extracted from the virtual environment characteristic data, such as the increase of carbon emission in peak period and the decrease of carbon emission in trough period; the limiting amplitude processing is to adjust the upper and lower limits of the preliminary adjustment range according to the carbon emission change trend, so that it is more suitable for the operation of carbon neutralization target; the upper limit of voltage fluctuation and the lower limit of voltage fluctuation are the highest value and the lowest value of the adjustable voltage of the device determined after limiting amplitude processing; the frequency fluctuation interval is the range of the adjustable frequency of the device determined after limiting amplitude processing, and finally generates the voltage fluctuation upper and lower limits and the frequency fluctuation interval of each device.

[0070] a5, the upper and lower limits of the voltage fluctuation upper limit, the voltage fluctuation lower limit and the frequency fluctuation interval are integrated into the power supply adjustment threshold of the electrical equipment.

[0071] The upper limit of voltage fluctuation and the lower limit of voltage fluctuation are the voltage adjustment boundaries of each device determined after the a4 step limiting processing; the frequency fluctuation interval is the frequency adjustment range of each device determined after the a4 step limiting processing; the power supply adjustment threshold of the power-using device is the final adjustment standard formed by integrating the upper limit of voltage fluctuation, the lower limit of voltage fluctuation and the frequency fluctuation interval of each device, which clearly defines the adjustable range of the power supply voltage and frequency of the device, and is the direct basis for subsequent accurate adjustment of device parameters, and finally generates the power supply adjustment threshold of each device.

[0072] The present application provides the following specific examples: a data center needs to determine the power supply adjustment threshold of the power-using device, and the specific process is as follows: first, perform a1 step, the lower unit receives the final carbon quota allocation ratio (enterprise A 28%, enterprise B 36%, enterprise C 36%) from the upper unit, calls the “enterprise-device” association record, finds that enterprise A corresponds to server group 1, enterprise B corresponds to server group 2, and enterprise C corresponds to server group 3, binds the quota ratio with the device, and clearly defines that server group 1 accounts for 28%, server group 2 accounts for 36%, and server group 3 accounts for 36%; then perform a2 step, sort the carbon quota proportion from high to low as server group 2=server group 3>server group 1, set the reference adjustment coefficient 1.1 for server groups 2 and 3, and set 0.9 for server group 1; then perform a3 step, extract the current running power of each device from the real-time computing power data, calculate the basic adjustment value, and refer to the device manual to determine the preliminary adjustment range; then perform a4 step, analyze the virtual environment characteristic data, extract the trend of “daily 10-16 carbon emission peak, which will exceed the standard when running in the preliminary range”, limit the preliminary range (server group 1: voltage lower limit is raised to 200 volts, frequency lower limit is raised to 49.9 hertz; server groups 2 and 3: voltage upper limit is reduced to 230 volts, frequency upper limit is reduced to 51 hertz), and clearly define the upper and lower limits of voltage fluctuation and the frequency fluctuation interval of each device; finally, perform a5 step, integrate the device parameters to form the power supply adjustment threshold, and complete the entire threshold determination process.

[0073] By performing a1-a5, the embodiment of the present application realizes the accurate binding of carbon quota and power utilization equipment through the a1 step, avoids the subsequent device adjustment from deviating from the carbon quota target; the a2 step sets different reference adjustment coefficients according to the proportion, solves the problem of "unified coefficient leading to mismatch between adjustment intensity and quota demand"; the a3 step combines the real-time running power to calculate the preliminary adjustment range, ensures that the adjustment range fits the actual load of the equipment, and avoids blind adjustment that deviates from the running state; the a4 step limits the processing according to the carbon emission trend, so that the adjustment range can adapt to the low-carbon regulation and control demand at different times, and prevent carbon emission from exceeding the standard or the equipment from running inefficiently; the a5 step integrates parameters to form clear thresholds, providing unified and executable standards for subsequent device voltage and frequency adjustment. The whole process deeply integrates carbon quota, device load, carbon emission trend and adjustment range, ensures that the final determined power supply adjustment threshold not only meets the carbon neutralization target of the data center, but also adapts to the actual operation demand of the equipment, lays a reliable foundation for subsequent accurate regulation and control of the equipment and realizes the carbon neutralization of the calculated load.

[0074] In a possible embodiment, a3, based on real-time computing power data, obtains the current running power of each power utilization equipment, calculates the running power and the reference adjustment coefficient to obtain the preliminary voltage adjustment range and the preliminary frequency adjustment range, including:

[0075] b1, based on real-time computing power data, extracts the computing power usage records corresponding to each power utilization equipment, and obtains the current running power of each power utilization equipment from the computing power usage records.

[0076] Among them, the real-time computing power data is the data collected before reflecting the current use of data center computing power of each enterprise, which contains the running related information of the enterprise to the power utilization equipment; the computing power usage record is the usage record directly associated with each power utilization equipment extracted from the real-time computing power data, which contains the details of computing power consumption when the equipment is running; the current running power of each power utilization equipment is the power value of the power utilization equipment in the current working state, which can reflect the actual running load of the equipment, and can be extracted from the computing power usage record according to the corresponding relationship between computing power consumption and power; finally, by extracting the computing power usage record and obtaining the current running power, the actual load fitting basic data is provided for subsequent calculation of equipment adjustment parameters.

[0077] b2, based on the function type of the power utilization equipment, sets corresponding voltage conversion ratio and frequency conversion ratio for different types of power utilization equipment.

[0078] The function type of the power-using equipment is a category divided according to the core task undertaken by the power-using equipment in the data center, such as equipment specially for enterprise to process computing tasks, equipment specially for the data center to provide cold quantity to maintain the temperature of the equipment, etc.; the voltage conversion ratio is a proportional value set according to the function type of the equipment, which is used for subsequent conversion of the basic adjustment value into an adjustable voltage range of the equipment; the frequency conversion ratio is also a proportional value set according to the function type of the equipment, which is used for subsequent conversion of the basic adjustment value into an adjustable frequency range of the equipment; the two proportions will differ due to different functions of the equipment, and finally through setting differentiated proportions, a basis is provided for subsequent calculation of adjustment ranges for different function equipment.

[0079] b3, the current running power of each power-using equipment is calculated with the corresponding reference adjustment coefficient to obtain the basic adjustment value of each power-using equipment.

[0080] The reference adjustment coefficient is a value set in advance according to the carbon quota proportion corresponding to each equipment, which is used to reflect the difference in adjustment intensity of different equipment; the basic adjustment value is a result value obtained by multiplying the current running power of each equipment with the corresponding reference adjustment coefficient, which integrates the actual load of the equipment and the adjustment intensity demand, and is the core basic data for subsequent calculation of the preliminary voltage and frequency adjustment range. Finally, the basic adjustment value of each equipment is obtained through calculation.

[0081] b4, the basic adjustment value is multiplied by the corresponding voltage conversion ratio to obtain the preliminary voltage adjustment range of each power-using equipment, and the basic adjustment value is multiplied by the corresponding frequency conversion ratio to obtain the preliminary frequency adjustment range of each power-using equipment.

[0082] The preliminary voltage adjustment range is obtained by multiplying the basic adjustment value of each equipment with the corresponding voltage conversion ratio, and then combining the current voltage of the equipment to obtain the voltage adjustable range; the preliminary frequency adjustment range is obtained by multiplying the basic adjustment value of each equipment with the corresponding frequency conversion ratio, and then combining the current frequency of the equipment to obtain the frequency adjustable range; the two ranges are the initial basis for subsequent amplitude limiting processing according to the carbon emission trend, and finally the preliminary voltage and frequency adjustment ranges of each equipment are generated.

[0083] The application provides the following specific examples: when a data center carries out preliminary adjustment range calculation of power-consuming equipment, the overall process is as follows: first, execute b1 step, the staff extracts the computing power use records of each power-consuming equipment from real-time computing power data through data processing software, wherein the records of server group 1 corresponding to enterprise A show that “the current hourly computing power consumption is 210 units”, the records of server group 2 corresponding to enterprise B show that “the current hourly computing power consumption is 230 units”, and the records of the refrigeration equipment group corresponding to enterprise C show that “the current hourly computing power consumption is 180 units”; then, according to the preset “computing power consumption-operation power” corresponding rule, the current operation power of server group 1 is 2100 watts, the current operation power of server group 2 is 2300 watts, and the current operation power of the refrigeration equipment group is 1800 watts; next, execute b2 step, divide server groups 1 and 2 into “computing equipment” and divide the refrigeration equipment group into “cold equipment”, and refer to the equipment characteristics to set the voltage conversion ratio of the computing equipment to 0.1, the frequency conversion ratio of the computing equipment to 0.002, the voltage conversion ratio of the cold equipment to 0.08, and the frequency conversion ratio of the cold equipment to 0.0015; then, execute b3 step, call the reference adjustment coefficient of each equipment, and calculate the basic adjustment value: the basic adjustment value of server group 1 is 2100*0.9=1890, the basic adjustment value of server group 2 is 2300*1.1=2530, and the basic adjustment value of the refrigeration equipment group is 1800*1.0=1800; finally, execute b4 step, confirm the current voltage and frequency of each equipment, calculate the preliminary voltage adjustment range: the preliminary voltage adjustment range of server group 1 is 200±(1890*0.1)=181.1 to 218.9 volts, the preliminary voltage adjustment range of server group 2 is 235.3 volts, and the preliminary voltage adjustment range of the refrigeration equipment group is 190±(1800*0.08)=175.6 to 204.4 volts; calculate the preliminary frequency adjustment range: the preliminary frequency adjustment range of server group 1 is 50±(1890*0.002)=49.622 to 50.378 hertz, the preliminary frequency adjustment range of server group 2 is 50±(2530*0.002)=49.494 to 50.506 hertz, and the preliminary frequency adjustment range of the refrigeration equipment group is 50±(1800*0.0015)=49.73 to 50.27 hertz, and the entire preliminary adjustment range calculation process is completed.

[0084] By performing b1~b4, the embodiment of the application extracts the device computing power usage record from the real-time computing power data and obtains the current running power through the b1 step, establishes a direct correlation between the computing power data and the actual load of the device, avoids the invalidation of parameters due to the disconnection of the device state in subsequent calculations; the b2 step sets differentiated voltage and frequency conversion ratios according to the device function type, solves the problem of mismatch between the unified ratio and the device function, and ensures that the adjustment range can adapt to the operating characteristics of different devices; the b3 step multiplies the current power of the device by the reference adjustment coefficient to obtain the basic adjustment value, which combines the adjustment intensity demand of the correlation between the device load and the carbon quota, and provides core data for subsequent calculations that take into account dual requirements; the b4 step calculates the preliminary adjustment range based on the basic adjustment value and the function adaptation ratio, in combination with the current voltage and frequency of the device, and the generated range not only fits the actual operating state of the device, but also matches the carbon neutralization target and the device function demand, avoiding the problem of blindly setting the adjustment range. The entire fusion of computing power, device load, carbon quota adjustment intensity and device function depth provides a reasonable and accurate initial basis for subsequent optimization of the adjustment range according to the carbon emission trend, ensuring that subsequent control work can be orderly carried out around the actual device and the carbon neutralization target.

[0085] In a possible embodiment, S13 simulates, based on the digital twin model, the change in carbon emissions under the preset computing power scheduling strategy to generate virtual environment feature data, including:

[0086] Step 131, based on the digital twin model, simulates, according to the allocation time period in the preset computing power scheduling strategy, the change data of power consumption and the change data of cold consumption corresponding to the enterprise computing power usage in each allocation time period.

[0087] The digital twin model is a virtual model that can simulate the actual data center operating state, and can reflect the device operating parameters, resource allocation, and consumption changes corresponding to enterprise computing power usage; the preset computing power scheduling strategy is a plan for allocating computing power in advance, which includes dividing a day or a scheduling period into multiple specific time intervals; the allocation time period is a specific time interval in the computing power scheduling strategy for phased arrangement of enterprise computing power allocation; the enterprise computing power usage refers to the scale and change of the actual use of computing power by each enterprise in each allocation time period; the power consumption change data is real-time change information of the power consumption of the data center electrical equipment as the enterprise computing power usage changes; the cold consumption change data is real-time change information of the cold consumption of the refrigeration system as the computing power usage changes to maintain the normal operating temperature of the device; and finally, the digital twin model is used to simulate the change data of power consumption and the change data of cold consumption corresponding to the enterprise computing power usage in each allocation time period.

[0088] Step 132, according to the power consumption change data and the cold consumption change data of each allocation period, calculate the carbon emission change value corresponding to the allocation period.

[0089] Wherein, the carbon emission change value is calculated according to the power and cold consumption change data of each period, combined with the conversion coefficient of "unit power / cold corresponding carbon emission" determined in advance, the real-time change value of carbon emission in each allocation period; finally, the carbon emission change value corresponding to each allocation period is calculated to provide the core content for the subsequent integration to form complete environmental data.

[0090] Step 133, integrate the carbon emission change value of each allocation period to form virtual environment characteristic data.

[0091] Wherein, the carbon emission change value of each allocation period is the specific value of the change of carbon emission with time in each allocation period calculated in step 132, which contains the carbon emission data of the key time points such as the initial, peak and end of the period; the virtual environment characteristic data is formed by integrating the carbon emission change value of all allocation periods in time sequence, supplementing the key node data in the period, which can completely reflect the carbon emission change trend in the whole scheduling period; finally, by integrating the data of each period, the virtual environment characteristic data covering the whole scheduling period and the data is coherent, which provides the trend basis for the subsequent optimization of carbon quota allocation and equipment adjustment.

[0092] The application provides the following specific examples: first, step 131 is performed, a digital twin model is called, and a preset computing power scheduling strategy is called; a 8:00-12:00 period is simulated, computing power of enterprise A 200 units, B 150 units, and C 100 units is allocated according to the strategy, according to the rule that “10 units of computing power correspond to 10 watts of power and 6 units of cooling capacity”, power is calculated to increase from 500 watts to 500+(200+150+100) / 10*10=1000 watts, and cooling capacity is calculated to increase from 300 units to 300+(200+150+100) / 10*6=600 units; similarly, a 12:00-18:00 period is simulated, power of 1000-1600 watts and cooling capacity of 600-900 units are obtained; a 18:00-8:00 period of the next day is simulated, power of 1600-600 watts and cooling capacity of 900-360 units are obtained. Then step 132 is performed, the power carbon emission coefficient is set to 0.5 units of carbon / watt, and the cooling capacity is set to 0.3 units of carbon / unit of cooling capacity; carbon emissions in the 8:00-12:00 period are calculated, the initial value is 500*0.5+300*0.3=340 units, and the end value is 1000*0.5+600*0.3=680 units; carbon emissions in the 12:00-18:00 period are calculated, the initial value is 1000*0.5+600*0.3=680 units, and the end value is 1600*0.5+900*0.3=1070 units; carbon emissions in the 18:00-8:00 period of the next day are calculated, the initial value is 1600*0.5+900*0.3=1070 units, and the end value is 600*0.5+360*0.3=408 units. Finally, step 133 is performed, carbon emission data of each period is connected in time sequence to form virtual environment feature data.

[0093] By performing steps 131-133, the embodiment of the application converts the abstract algorithm scheduling strategy into specific resource consumption data through step 131 by simulating the change of power and cold consumption corresponding to enterprise algorithm usage by time period using the digital twin model, providing basic information that fits the actual operation scenario for subsequent carbon emission calculation, and avoiding the deviation of carbon emission calculation caused by the lack of consumption data; step 132 converts the consumption data into carbon emission change values by setting a conversion coefficient, establishes a direct correlation between resource consumption and carbon emission, and makes the originally dispersed consumption data become a core index that can directly reflect the environmental impact, providing key content for subsequent trend analysis; step 133 integrates carbon emission data of each period and supplements key node information to form virtual environmental feature data covering the entire scheduling period and data is coherent, clearly presenting the overall trend of carbon emission, and avoiding the one-sidedness of trend analysis caused by scattered data. The whole process from consumption simulation to carbon emission conversion, and then to data integration forms a complete environmental feature data chain, providing comprehensive and accurate basis for subsequent optimization of carbon quota allocation ratio and adjustment of equipment operation parameters based on carbon emission trend, ensuring that the subsequent regulation and control work can fit the carbon emission change law of the data center, and helping to achieve the carbon neutralization goal.

[0094] In a possible embodiment, step 132 calculates the carbon emission change value of the corresponding allocation period according to the power consumption change data and the cold consumption change data of each allocation period, comprising:

[0095] c1, for the power consumption change data of each allocation period, the power carbon emission change amount of the corresponding period is calculated according to the carbon emission proportion of power generated in the allocation period.

[0096] Wherein, the power consumption change data of each allocation period is obtained by simulating the digital twin model in advance, the change information of the power consumption of the data center equipment in each allocation period, including the power consumption value at the beginning of the period, the power consumption value at the end of the period and the change in the intermediate process; the carbon emission proportion of power generated in the allocation period is the carbon emission amount generated by consuming one unit of power in the period, which will be different due to the different proportions of energy types (such as green energy and conventional energy) used in the period; the power carbon emission change amount is calculated by the power consumption change data of each allocation period and the power carbon emission proportion of the corresponding period, representing the carbon emission change value generated by the change of power consumption in the period; the finally generated power carbon emission change amount of each allocation period will provide carbon emission data related to power consumption for subsequent summary of the total carbon emission change value of the period.

[0097] c2, for the cold consumption change data of each allocation period, the cold carbon emission change amount of the corresponding period is calculated according to the carbon emission proportion of cold generated in the allocation period.

[0098] The power consumption change data of each allocation period is simulated by the digital twin model in advance, and the change information of the power consumption of the data center electrical equipment in each allocation period includes the power consumption value at the beginning of the period, the power consumption value at the end of the period, and the change in the intermediate process; the carbon emission proportion of the power generated in the allocation period is the carbon emission amount generated by consuming one unit of power in the period, and this proportion will be different due to the different proportions of the types of energy used in the period (such as green energy, conventional energy); the power carbon emission change amount is calculated by the power consumption change data of each allocation period and the corresponding period power carbon emission proportion, which represents the carbon emission change value generated by the change of power consumption in the period; the finally generated power carbon emission change amount of each allocation period will provide carbon emission data related to power consumption for the subsequent total carbon emission change value of the summary period.

[0099] c3, the power carbon emission change amount and the cold carbon emission change amount of the same allocation period are added to obtain the carbon emission change value of the corresponding allocation period.

[0100] The carbon emission change value is the result of adding the power carbon emission change amount and the cold carbon emission change amount of the same allocation period, which represents the total carbon emission change value in the period; the finally generated carbon emission change value of each allocation period will provide complete period carbon emission change information for the subsequent integration of virtual environment feature data in time sequence, and ensure that the overall carbon emission change of the data center in the period is reflected.

[0101] The application provides the following specific examples: a data center carries out carbon emission change value calculation work in each allocation period, and the overall process is as follows: first, c1 step is performed, and the power consumption change data of three allocation periods are called, which are 500 watts at 8:00 to 1000 watts at 12:00, 1000 watts at 12:00 to 1600 watts at 18:00, and 1600 watts at 18:00 to 600 watts at 8:00 the next day; the power carbon emission proportion of each period is determined according to the energy use record, the 8:00-12:00 period is set to 0.4 because of the high proportion of green energy, the 12:00-18:00 period is set to 0.6 because of the high proportion of conventional energy, and the 18:00-8:00 period the next day is set to 0.3 because of the rebound of green energy; the power carbon emission change of each period is calculated, which is (1000-500) x 0.4=200 units of carbon, (1600-1000) x 0.6=360 units of carbon, and (600-1600) x 0.3=-300 units of carbon. Then, c2 step is performed, and the cold consumption change data of each period is called, which are 300 units at 8:00 to 600 units at 12:00, 600 units at 12:00 to 900 units at 18:00, and 900 units at 18:00 to 360 units at 8:00 the next day; the cold carbon emission proportion of each period is determined according to the refrigeration system operating efficiency, the 8:00-12:00 period is set to 0.3 because of the high efficiency, the 12:00-18:00 period is set to 0.4 because of the efficiency drop, and the 18:00-8:00 period the next day is set to 0.2 because of the efficiency rebound; the cold carbon emission change of each period is calculated, which is (600-300) x 0.3=90 units of carbon, (900-600) x 0.4=120 units of carbon, and (360-900) x 0.2=-108 units of carbon. Finally, c3 step is performed, and the power and cold carbon emission change of the same period is added, which is 200+90=290 units of carbon at 8:00-12:00, 360+120=480 units of carbon at 12:00-18:00, and-300+(-108)=-408 units of carbon at 18:00-8:00 the next day, to obtain the carbon emission change value of each allocation period.

[0102] By performing c1~c3, the embodiment of the application calculates the power carbon emission change through the c1 step combined with the differences of energy types in different periods, avoids the calculation deviation of power carbon emission caused by ignoring the influence of energy types, and ensures that the influence of power consumption on carbon emission is accurately reflected; the c2 step calculates the cold carbon emission change combined with the differences of refrigeration system operation efficiency, supplements the influence of cold consumption, which is the core energy consumption of the data center, on carbon emission, and avoids the incompleteness of carbon emission calculation due to the lack of cold factors; the c3 step adds and integrates the power and cold carbon emission changes in the same period to form complete data reflecting the total carbon emission change of the period, avoiding the loss of overall carbon emission trend caused by splitting data. The whole process from accurate calculation of factors to integration of overall data provides accurate and complete period carbon emission basis for subsequent integration of virtual environment feature data in chronological order, ensuring that the analysis of data center carbon emission trend can truly reflect the overall operation, and providing reliable data support for formulating low-carbon regulation strategies that fit the actual situation.

[0103] In a possible embodiment, S12, based on real-time computing power data and real-time carbon price data, constructs a digital twin model, including:

[0104] Step 121, associate the real-time computing power data with the operation records of each electrical equipment in the data center to determine the association between the computing power usage of different enterprises and the electrical equipment used.

[0105] Wherein, the real-time computing power data is a set of information reflecting the current computing power usage of each enterprise in the data center, including enterprise name, used computing power scale, use time, etc.; the operation records of each electrical equipment in the data center are a set of information recording the working state of each electrical equipment (such as server group providing computing service, refrigeration equipment maintaining equipment temperature), including equipment number, operation time, load condition, served enterprise, etc.; the association between the computing power usage of different enterprises and the electrical equipment used is the corresponding relationship that clearly indicates which electrical equipment provides support for the computing power usage of a certain enterprise; finally, through the association operation of real-time computing power data and electrical equipment operation records, this corresponding relationship is generated, laying a foundation for the subsequent establishment of the relationship between computing power and energy consumption.

[0106] Step 122, based on the association relationship, set the corresponding proportion of computing power usage and power consumption of electrical equipment used to make the computing power change reflect the power consumption change, and set the conversion rule of power consumption and cold consumption.

[0107] The corresponding relationship is the correspondence between the enterprise computing power usage and the power equipment determined in step 121, and is the basis for setting subsequent proportions and rules; the corresponding proportion of the computing power usage and the power consumption of the power equipment is a numerical relationship that specifies how much power will be consumed by the power equipment for each use of a certain scale of computing power, and through the proportion, the change of the computing power can be directly reflected as the change of the power consumption; the conversion rule of the power consumption and the cold consumption is a rule that specifies how much cold needs to be supplemented to maintain the normal operating temperature of the equipment for each consumption of a certain amount of power, and through the rule, the change of the power consumption can be reflected as the change of the cold consumption; the two proportions and rules are finally generated, and the corresponding relationship between the computing power, the power consumption and the cold consumption is established, so that the changes of the three are related to each other and can be deduced from each other.

[0108] Step 123, introducing real-time carbon price data, setting the conversion method of power consumption, cold consumption and carbon emission according to the energy carbon emission level corresponding to the carbon price.

[0109] The real-time carbon price data is the transaction price information of each unit of carbon emission right in the current carbon trading market, which will affect the carbon emission level of the energy used by the data center; the energy carbon emission level corresponding to the carbon price is the carbon emission amount generated in the process of producing unit power or generating unit cold under the current real-time carbon price; the conversion method of power consumption, cold consumption and carbon emission is a calculation method that specifies how much carbon emission will be generated for each consumption of a certain amount of power or cold; finally, by introducing the real-time carbon price data, combined with the corresponding energy carbon emission level, the conversion method is set up, and the direct relationship between the power consumption, the cold consumption and the carbon emission is established, so that the energy consumption data can be converted into carbon emission data.

[0110] Step 124, integrating the corresponding proportion, the conversion rule and the conversion method to generate a digital twin model that can reflect the power consumption, the cold consumption and the carbon emission in real time.

[0111] The digital twin model is a virtual model that can simulate the actual operating state of the data center. By integrating the above-mentioned corresponding proportion, conversion rule and conversion method, the model can automatically calculate and display the corresponding power consumption, cold consumption and carbon emission after inputting the real-time computing power data of the enterprise; finally, the digital twin model that can reflect the changes of the three in real time is generated, providing a visual and calculable virtual tool for subsequent operation monitoring and control of the data center.

[0112] The present application provides the following specific examples: the overall process of constructing a digital twin model by a certain data center is as follows: first, step 121 is performed, real-time computing power data and power equipment operation records are called through a data management system, and the association relationship of enterprise A with server group 1, enterprise B with server group 2, and enterprise C with server group 3 is determined by matching enterprise names. Then step 122 is performed, based on the association relationship, server group 1 is continuously counted for 3 days, and power consumption and power consumption at different time periods are recorded, it is found that every 10 units of computing power corresponds to 100 watts of power, and the ratio of “10 units of computing power = 100 watts of power” is set; the power and cooling capacity consumption are also counted, it is found that every 100 watts of power needs 50 units of cooling capacity to maintain the equipment temperature, and the rule of “100 watts of power = 50 units of cooling capacity” is set, and server groups 2 and 3 set the same way of setting the ratio and the rule. Then step 123 is performed, the real-time carbon price of 10 yuan per unit of carbon emission is obtained through the official interface of the carbon trading market, the energy carbon emission level under the carbon price is queried: 100 watts of power generates 0.05 units of carbon emission, and 50 units of cooling capacity generates 0.03 units of carbon emission, and the conversion method is set accordingly: power carbon emission = power consumption ÷ 100 × 0.05, and cooling capacity carbon emission = cooling capacity consumption ÷ 50 × 0.03. Finally, step 124 is performed, the ratio, rule and conversion method are input into a virtual modeling tool, and the linkage logic of “computing power → power → cooling capacity → carbon emission” is written; when testing, 200 units of computing power of enterprise A is input, the model first calculates the power: 200 ÷ 10 × 100 = 2000 watts, then calculates the cooling capacity: 2000 ÷ 100 × 50 = 1000 units, and finally calculates the carbon emission: (2000 ÷ 100 × 0.05) + (1000 ÷ 50 × 0.03) = 1 + 0.6 = 1.6 units, the real-time display result is displayed, and the construction of the digital twin model is completed.

[0113] By performing steps 121-124, the embodiment of the application establishes the association between enterprise computing power and electrical equipment through step 121, solves the problem of lack of clear equipment attribution in subsequent rule setting, and lays a precise data corresponding foundation for the entire process; step 122 establishes the linkage relationship between computing power and electricity, electricity and cold quantity, converts abstract computing power demand into specific energy consumption data, fills in the gap between "computing power demand-actual energy consumption", and makes energy consumption changes derivable through computing power changes; step 123 introduces real-time carbon price and sets the conversion method of energy consumption and carbon emissions, so that energy consumption data can directly reflect environmental impact, meet the core goal of low-carbon regulation, and avoid the disconnection between energy consumption control and carbon emission target; step 124 integrates all rules to build a digital twin model, realizes the visualization linkage of "computing power input-multi-index output", and avoids the difficulty of regulation caused by data fragmentation. The model generated by closely combining enterprise demand, equipment operation, energy consumption control and carbon emission target can intuitively reflect the running state of the data center, provide a reliable virtual tool for subsequent precise regulation of computing power allocation and control of carbon emissions, ensure that the regulation decision fits the actual operation, and help the data center gradually achieve the carbon neutralization target.

[0114] Figure 3 The structure diagram of a low-carbon regulation system for data center computing load provided by the embodiment of the application is shown in Figure 3 The system comprises:

[0115] The acquisition module 31 is used to acquire real-time computing power data of each enterprise, and acquire real-time carbon price data of the carbon trading market, and the real-time computing power data and the real-time carbon price data are used as basic data for low-carbon regulation of the data center computing load.

[0116] The construction module 32 is used to construct a digital twin model based on the real-time computing power data and the real-time carbon price data, and the digital twin model is used to map the power consumption, cold consumption and carbon emission of the data center in real time.

[0117] The generation module 33 is used to simulate the change of carbon emission under a preset computing power scheduling strategy based on the digital twin model, to generate virtual environment feature data.

[0118] The adjustment module 34 is used to design a double-layer game model, the upper unit of the double-layer game model receives the basic data and the virtual environment feature data, determines the final carbon quota allocation ratio according to the computing power demand of each enterprise, and the lower unit dynamically adjusts the power supply voltage and power frequency of the electrical equipment in the data center according to the final carbon quota allocation ratio, to realize the low-carbon regulation of the data center computing load.

[0119] The low-carbon regulation system for data center computing load of the embodiments of the present application is used to implement the low-carbon regulation method for data center computing load described above, and therefore the specific embodiments in the low-carbon regulation system for data center computing load can be seen from the embodiments of the low-carbon regulation method for data center computing load described above, and the specific embodiments can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0120] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of the low-carbon regulation method for data center computing load described above when executing the computer program.

[0121] The present application also provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the low-carbon regulation method for data center computing load described above when executed by a processor.

[0122] In an exemplary embodiment, the computer readable storage medium described above can include but is not limited to: a U disk, a read-only memory, a random access memory, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0123] The embodiments of the present application also provide a computer program product, which comprises a computer program, and the computer program implements the steps in the low-carbon regulation method for data center computing load described above when executed by a processor.

[0124] The skilled person can further realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the above description. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0125] The low-carbon regulation method, system, device and storage medium for data center computing load provided by the present application are described in detail above. The principles and implementation manners of the present application are described by specific examples in this paper, and the above description of the examples is only used to help understand the method and core idea of the present application. It should be pointed out that, for ordinary skilled person in the technical field, without departing from the principles of the present application, some improvements and modifications can be made to the present application, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A low-carbon regulation method for data center computing load, characterized in that, The method comprises the following steps: Collecting real-time computing power data of each enterprise and real-time carbon price data of a carbon trading market, wherein the real-time computing power data and the real-time carbon price data are used as basic data for low-carbon regulation of computing load of a data center; Based on the real-time computing power data and the real-time carbon price data, a digital twin model is constructed, which real-time maps power consumption, cooling consumption and carbon emissions of the data center; Based on the digital twin model, the change of carbon emissions under a preset computing power scheduling strategy is simulated to generate virtual environment feature data; A double-layer game model is designed, wherein an upper unit of the double-layer game model receives the basic data and the virtual environment feature data, determines a final carbon quota allocation ratio according to computing power demand of each enterprise, and a lower unit dynamically adjusts power supply voltage and power frequency of power-consuming equipment in the data center according to the final carbon quota allocation ratio, so as to realize low-carbon regulation of computing load of the data center. 2.The method of claim 1, wherein, The design of the double-layer game model, wherein an upper unit of the double-layer game model receives the basic data and the virtual environment feature data, determines a final carbon quota allocation ratio according to computing power demand of each enterprise, and a lower unit dynamically adjusts power supply voltage and power frequency of power-consuming equipment in the data center according to the final carbon quota allocation ratio, so as to realize low-carbon regulation of computing load of the data center, comprises: The upper unit of the double-layer game model receives the basic data and the virtual environment feature data, divides computing power use intervals of different enterprises according to computing power demand declared by each enterprise, and generates an initial carbon quota allocation ratio according to a proportion of the computing power use intervals; The upper unit compares the initial carbon quota allocation ratio with a carbon emission change trend in the virtual environment feature data, adjusts the initial carbon quota allocation ratio of each enterprise according to a comparison result, and forms a final carbon quota allocation ratio; The lower unit determines a power supply adjustment threshold of the power-consuming equipment according to the final carbon quota allocation ratio; Based on the power supply adjustment threshold, the power supply voltage of the power-consuming equipment is adjusted to a preset threshold range, and the power frequency is adjusted according to the adjusted power supply voltage; Based on the adjusted power supply voltage and the adjusted power frequency, low-carbon regulation is performed on the computing load of the data center. 3.The method of claim 2, wherein, The lower unit determines a power supply adjustment threshold of the power-consuming equipment according to the final carbon quota allocation ratio, which comprises: The lower unit associates the final carbon quota allocation ratio with the power-consuming equipment corresponding to each enterprise, and determines a carbon quota proportion corresponding to each power-consuming equipment; Each power-consuming equipment is sorted according to the carbon quota proportion from high to low, and a corresponding reference adjustment coefficient is set for each sorted power-consuming equipment; Based on the real-time computing power data, the current operating power of each power-consuming equipment is obtained, the operating power and the reference adjustment coefficient are calculated to obtain a preliminary voltage adjustment range and a preliminary frequency adjustment range; According to the carbon emission trend in the virtual environment feature data, the preliminary voltage adjustment range and the preliminary frequency adjustment range are respectively subjected to amplitude limiting processing to demarcate the upper limit of voltage fluctuation, the lower limit of voltage fluctuation and the frequency fluctuation interval; The upper and lower limits of the voltage fluctuation upper limit, the voltage fluctuation lower limit and the frequency fluctuation interval are integrated into the power supply adjustment threshold of the power consumption equipment. 4.The method of claim 3, wherein, Based on the real-time computing power data, the current running power of each power consumption equipment is obtained, and the running power and the reference adjustment coefficient are calculated to obtain a preliminary voltage adjustment range and a preliminary frequency adjustment range, including: Based on the real-time computing power data, the computing power usage records corresponding to each power consumption equipment are extracted, and the current running power of each power consumption equipment is obtained from the computing power usage records; Based on the function type of the power consumption equipment, corresponding voltage conversion ratio and frequency conversion ratio are set for different types of power consumption equipment; The current running power of each power consumption equipment and the corresponding reference adjustment coefficient are calculated to obtain the basic adjustment value of each power consumption equipment; The basic adjustment value is multiplied by the corresponding voltage conversion ratio to obtain the preliminary voltage adjustment range of each power consumption equipment, and the basic adjustment value is multiplied by the corresponding frequency conversion ratio to obtain the preliminary frequency adjustment range of each power consumption equipment. 5.The method of claim 1, wherein, Based on the digital twin model, the carbon emission under the preset computing power scheduling strategy is simulated to generate virtual environment feature data, including: Based on the digital twin model, according to the distribution period in the preset computing power scheduling strategy, the power consumption change data and the cold consumption change data corresponding to the enterprise computing power usage in each distribution period are simulated in turn; According to the power consumption change data and the cold consumption change data of each distribution period, the carbon emission change value of the corresponding distribution period is calculated; Integrate the carbon emission change values of each distribution period to form virtual environment feature data. 6.The method of claim 5, wherein, According to the power consumption change data and the cold consumption change data of each distribution period, the carbon emission change value of the corresponding distribution period is calculated, including: For the power consumption change data of each distribution period, the power carbon emission change amount of the corresponding period is calculated according to the carbon emission proportion of power in the distribution period; For the cold consumption change data of each distribution period, the cold carbon emission change amount of the corresponding period is calculated according to the carbon emission proportion of cold in the distribution period; The power carbon emission change amount and the cold carbon emission change amount of the same distribution period are added to obtain the carbon emission change value of the corresponding distribution period.

7. The method of claim 1, wherein the data center computing load is a low-carbon regulated. Based on the real-time computing power data and the real-time carbon price data, a digital twin model is constructed, including: The real-time computing power data is associated with the running records of each power consumption equipment in the data center to determine the association between the computing power usage of different enterprises and the power consumption equipment; Based on the association relationship, set the corresponding proportion of computing power usage and power consumption of the electric device, so that the change of computing power reflects the change of power consumption, and set the conversion rule of power consumption and cold consumption; Introducing the real-time carbon price data, setting the conversion mode of power consumption, cold consumption and carbon emission according to the energy carbon emission level corresponding to the carbon price; Integrate the corresponding proportion, the conversion rule and the conversion mode to generate a digital twin model that can reflect the power consumption, cold consumption and carbon emission in real time.

8. A low-carbon regulation system for data center computing load, characterized in that, Comprise: The acquisition module is used for acquiring real-time computing power data of each enterprise, and acquiring real-time carbon price data of carbon trading market, wherein the real-time computing power data and the real-time carbon price data are used as basic data for low-carbon regulation of data center computing load; The construction module is used for constructing a digital twin model based on the real-time computing power data and the real-time carbon price data, wherein the digital twin model maps the power consumption, cold consumption and carbon emission of the data center in real time; The generation module is used for simulating the change of carbon emission under a preset computing power scheduling strategy based on the digital twin model to generate virtual environment feature data; The adjustment module is used for designing a double-layer game model, wherein the upper unit of the double-layer game model receives the basic data and the virtual environment feature data, determines the final carbon quota allocation ratio according to the computing power demand of each enterprise, and the lower unit dynamically adjusts the power supply voltage and power frequency of the electric device in the data center according to the final carbon quota allocation ratio, so as to realize the low-carbon regulation of the data center computing load.

9. An electronic device, comprising: Comprise: The memory is used for storing computer programs; The processor is used for executing the computer programs to realize the steps of the low-carbon regulation method of the data center computing load according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer programs, and the computer programs can realize the low-carbon regulation method of the data center computing load according to any one of claims 1 to 7 when executed by the processor.

Images