37 results about "Green data center" patented technology

The invention discloses a green data center scheduling optimization method under computing power and electric power coupling, data load types of a data center comprise a transferable load, a reducible load and a rigid load, and the method comprises the following steps: constructing a double-target scheduling optimization model by taking the minimum power consumption cost of the data center and the maximum green power fitting degree as targets; the power consumption cost of the data center is used for representing the sum of expenses generated by the data center for purchasing power from a power grid in order to meet the power required for processing the data load in each time period, and the green power fitting degree is used for evaluating the matching degree between the power consumption demand of the data center and the renewable energy source supply; solving the dual-target scheduling optimization model by using a multi-target dogvessel swarm algorithm to obtain a Pareto frontier; and determining an optimal scheduling scheme based on the Pareto frontier, wherein the optimal scheduling scheme comprises a data center power consumption curve, an energy storage power station charging and discharging curve and a transferable load scheduling strategy. According to the invention, the power consumption cost of the data center can be reduced, and the renewable energy consumption capability can be improved.

The invention relates to the technical field of computing power resource scheduling, and particularly discloses a green data center computing power demand prediction and energy consumption control method, system and device. Firstly, computing power information and environmental parameters of a data center are collected in real time based on edge equipment; afterwards, a time sequence trend is predicted by using an LSTM neural network, a CNN network is fused to extract computing power features, an LSTM-CNN hybrid model is constructed, repeated training is performed through an Adam optimization algorithm, and the model can accurately pre-judge a future computing power demand; and in combination with analysis of historical energy consumption and real-time monitoring data, an energy consumption analysis model is established to evaluate power consumption conditions under different computing power requirements. Based on the prediction result and the energy consumption analysis, the working state of the server cluster is dynamically adjusted, the load is scheduled, the energy use is optimized, the energy efficiency ratio of the data center is continuously improved by verifying and adjusting the energy consumption strategy, the overall operation cost and the energy consumption are remarkably reduced, and the development requirements of green energy conservation are met.

An information handling system executing a software test suite green data center policy generation system comprising a network interface device receiving data center hardware utilization analytics and a software test suite identifier for a selected software test suit and a processor to identify a reserved capacity of the data center processing system reserved for general execution of software test suites according to general quality of service requirements for input / output commands and to predict a future data center processor utilization rate and execution duration for the selected software test suite by forecasting actual use of data center resources during a future execution of the selected software test suite that is below the reserved capacity, and the network interface device to transmit an instruction to throttle power supplied to an over-allocated portion of the data center resources from the reserved capacity duration execution of the selected software test suite.

The invention relates to a processor low-power-consumption state switching method based on an RISC-V server platform, and belongs to the technical field of computers. The method comprises the following steps: (1) configuring multi-stage low-power-consumption state information of the RISC-V processor; (2) initializing a CPU idle state management module by utilizing the multi-stage low-power-consumption state information; (3) the CPU idle state management module selects an optimal low-power-consumption state for the RISC-V processor based on a control strategy of a reinforcement learning SARSA algorithm; and (4) enabling the RISC-V processor to enter a specified low-power-consumption state. According to the invention, efficient power management of the RISC-V processor on a server platform is realized, the energy consumption can be obviously reduced while the system performance is ensured, the energy efficiency ratio of the server is effectively improved, and the method is particularly suitable for power management of a large-scale server cluster and construction of a green data center.

The invention discloses a green data center key cooling system, and relates to the technical field of data center energy saving and heat management. The system comprises a cabinet, a cold and hot channel, a refrigeration device, a waste heat recovery device and a control system. The refrigeration device comprises an indoor refrigeration unit, an outdoor heat dissipation unit and a fluorine pump energy-saving unit, and heat exchange and cooperative operation are achieved through pipelines. The cold and hot channel is provided with a guide plate and an adjustable air valve to optimize airflow distribution; the waste heat recovery device comprises a fluorine-fluorine heat exchanger, a heat pump system and a circulating medium box, and recovery and reutilization of heat energy are achieved. The control system integrates a lightweight edge AI module, constructs a multi-stage temperature sensing and load prediction mechanism, dynamically adjusts the power output of each cooling unit, and supports redundant fault-tolerant switching. The system intelligently switches the operation modes of the fluorine pump and the heat dissipation unit based on the external temperature and the thermal load prediction result, energy consumption is effectively reduced, and the energy efficiency ratio is increased.

The present invention relates to a method for switching the low-power state of a processor based on a RISC-V server platform, belonging to the field of computer technology. The steps include: (1) configuring multi-level low-power state information of the RISC-V processor; (2) initializing the CPU idle state management module using the multi-level low-power state information; (3) the CPU idle state management module selects the optimal low-power state for the RISC-V processor based on the control strategy of the reinforcement learning SARSA algorithm; (4) the RISC-V processor enters the specified low-power state. The present invention realizes efficient power management of the RISC-V processor on the server platform, can significantly reduce energy consumption while ensuring system performance, effectively improves the energy efficiency ratio of the server, and is particularly suitable for the power management of large-scale server clusters and the construction of green data centers.

The invention discloses a method and system for configuring a refrigerating system switching threshold value based on a large reasoning model, and relates to the technical field of green data center artificial intelligence energy saving in the mobile communication technology. The method for configuring the switching threshold value of the refrigerating system based on the reasoning large model comprises the steps that normalization scoring is conducted on collected CPU power consumption of a threshold value adjusting module, air cooling power consumption of a precise air conditioner and liquid cooling power consumption of a cooling cold plate, and an air cooling and liquid cooling switching threshold value model evaluation score is obtained; inputting the evaluation score and the cue word of the air cooling and liquid cooling switching threshold value model into a reasoning large model to obtain an air cooling and liquid cooling reasoning switching threshold value; and dividing the air cooling and liquid cooling reasoning switching threshold into a plurality of accuracy levels according to the evaluation score of the air cooling and liquid cooling switching threshold model, and updating the corresponding switching threshold when the threshold adjustment module executes the prediction of the corresponding accuracy by adopting a preset prediction model by using the plurality of accuracy levels. According to the invention, the switching accuracy and the energy utilization efficiency of the cabinet cooling system are improved.

The invention relates to the technical field of geothermal energy development and utilization, in particular to a superconducting pipe mediated geothermal transmission ground power generation system which comprises a geothermal collection and transmission module, a heat energy receiving and separating module, a power generation module and an energy gradient utilization module. The geothermal collection and transmission module collects heat energy from a geothermal rock stratum in a phase change mode through a closed circulation superconducting pipe and transmits the heat energy to the ground. The heat energy receiving and separating module receives the heat energy and separates the heat energy to form a high-temperature heat source and a low-temperature cold source; the power generation module uses a high-temperature heat source to drive power generation; the power input end of the data computing power center is connected with the power generation module to receive power, and the cooling input end of the data computing power center is connected with the heat energy receiving and separating module to receive a low-temperature cold source to cool internal equipment. Efficient transmission and gradient utilization of geothermal energy and direct energy supply of the green data center are achieved, and the system has the advantages of being high in efficiency, good in environmental protection property and high in resource utilization rate.

The invention discloses a non-exchange data center internal network architecture based on free space optical communication, and belongs to the technical field of optical communication and data center networks. According to the framework, a server with a tunable light emitter, a light distribution beam combiner and the like are arranged in a cabinet, and non-line-of-sight communication is realized through a spatial light element; all-optical routing is completed through wavelength, spatial mode and time slot joint multiplexing, and a traditional electric / optical switch is not needed. The optical path routing in the architecture adopts passive devices, optical fiber wiring is avoided, online expansion of a cabinet and a server is supported, deployment and energy consumption cost is remarkably reduced, high-bandwidth, low-delay and low-power-consumption green data center communication is achieved, and the system can be widely applied to scenes such as cloud computing and big data.

The invention relates to the technical field of cloud computing and green data centers, in particular to a cloud computing data center energy efficiency optimization and carbon reduction method and system based on computing power collaboration. The system comprises a multi-source data acquisition module, a computing power collaborative arrangement engine, a multi-target real-time optimizer, a carbon market linkage interface, a hierarchical collaborative scheduling architecture and an interpretable decision output unit. A unit effective cost model is constructed by fusing electricity prices, carbon emission factors and energy efficiency indexes, a rolling time domain optimization and hierarchical scheduling mechanism is combined, cross-center computing power resources are dynamically coordinated on the premise of guaranteeing SLA, multi-target optimization of electric power cost, carbon emission and migration overhead is achieved, carbon market signal linkage and scheduling decision traceability are supported, and the scheduling efficiency is improved. And the energy efficiency and the low-carbon operation level of the data center are effectively improved.

The invention discloses a multi-energy green data center task scheduling method and device, a medium and a product, and relates to the field of task scheduling, and the method comprises the steps: converting a multi-energy green data center task scheduling process into a Markov process; constructing a multi-agent reinforcement learning model according to the Markov process; acquiring the generating capacity of a renewable energy generator, demand information of an energy green data center and electricity price information; based on an XGBoost prediction method, determining a renewable energy power generation amount, an energy green data center demand quantity sequence and a current state of each energy green data center; according to the current state, adopting a multi-agent reinforcement learning model to determine an action; the root action obtains a real reward from the environment; and updating the multi-agent reinforcement learning model according to the expected reward and the real reward. According to the invention, lower SLO violation rate, total energy currency cost and total carbon emission are realized.

This application discloses a hybrid energy storage optimization configuration method for data centers that considers carbon emissions throughout the entire lifecycle, comprehensively taking into account the coupling relationship between the data center's thermodynamic environment, battery life degradation, and carbon footprint. The method first constructs a thermodynamic-based dynamic PUE model and a unified matrix model for batch processing loads. Second, it introduces an active frequency division mechanism for the hybrid energy storage system, utilizing supercapacitors to smooth high-frequency fluctuations and extend lithium battery life. Then, it establishes a dynamic implicit carbon emission model to quantify the impact of equipment degradation on the carbon footprint throughout the entire lifecycle. Based on this, a two-layer multi-objective optimization framework is proposed: the upper layer optimizes capacity configuration using the NSGA-II algorithm, and the lower layer performs rolling scheduling across multiple typical days using the CPLEX solver. This not only achieves decoupled optimization of economic costs and carbon emissions but also effectively improves the lifespan and environmental benefits of the energy storage system, making it suitable for low-carbon planning and operation of green data centers.

The application discloses a flywheel energy storage cooling system based on a green data center, relates to the flywheel energy storage cooling technical field, and solves the technical problem that the prior art cannot infer heat detection according to the heat state of different heat scenes, specifically, an energy storage operation detection module is used for performing operation detection on a flywheel energy storage system, numerical values are set for corresponding matching stages according to the matching stage analysis of internal components of the flywheel energy storage system, the operation time period of the flywheel energy storage system is counted at each moment, and the operation time period is divided into an abnormal operation time period and a normal operation time period according to the statistical values; a heat scene analysis unit is used for performing heat scene analysis on the flywheel energy storage system in different operation time periods, and each type of operation time period is obtained according to the scene analysis; and a cooling layout analysis optimization module is used for performing analysis and optimization on the cooling layout of the flywheel energy storage system.

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

The invention relates to the technical field of data center heat management, and provides a green data center liquid cooling heat management method and system, and the method comprises the steps: monitoring the temperature of a cooling liquid outlet of a liquid cooling loop, and obtaining a fluctuation characteristic; according to the fluctuation characteristics, the accumulated thermal load of the cold plate key area is evaluated, and the hardware thermal capacity margin is obtained; comparing the hardware thermal capacity margin with a preset thermal capacity threshold to obtain an intervention judgment result; when the intervention judgment result shows that the intervention mechanism is triggered, a local heat dissipation bottleneck is manufactured; through a local heat dissipation bottleneck, the core temperature of the processor in the hot spot area rises to an internal temperature protection threshold value, and the processor is driven to reduce the working frequency or power consumption; an upper-layer workload scheduler is controlled to adjust a resource allocation strategy; and when the recovery of the hardware thermal capacity margin is completed, stopping manufacturing the local heat dissipation bottleneck. The method has the advantages that an intelligent intervention and feedback mechanism is provided, the system performance and the energy efficiency are balanced, and hardware damage and unnecessary energy consumption are avoided.

The application discloses a data center hybrid energy storage optimization configuration method considering whole life cycle carbon emission, which can comprehensively consider the coupling relationship among the thermodynamic environment of the data center, the battery life attenuation and the carbon footprint. The method firstly constructs a dynamic PUE model based on thermodynamics and a batch processing load uniform matrix model; secondly, an active frequency division mechanism of the hybrid energy storage system is introduced, and a super capacitor is used to smooth high-frequency fluctuations to prolong the service life of a lithium battery; then, a dynamic implicit carbon emission model is established to quantify the influence of equipment attenuation on the whole life cycle carbon footprint. On this basis, a double-layer multi-objective optimization framework is proposed, the upper layer optimizes the capacity configuration through an NSGA-II algorithm, and the lower layer solves multiple typical days rolling scheduling through a CPLEX solver. In this way, not only the decoupling optimization of economic cost and carbon emission is realized, but also the service life and environmental benefits of the energy storage system are effectively improved, and the method is suitable for low-carbon planning and operation of green data centers.

The invention discloses a distributed intelligent calculation center coupled with land wind power generation and an energy consumption control system and method. An integrated system comprising a wind power generation system, an energy storage system and a distributed intelligent calculation center system is constructed; the wind power generation system converts wind energy to supply power, and the energy storage system stores redundant electric energy and releases the redundant electric energy when needed to guarantee stable operation of the intelligent calculation center. The distributed intelligent calculation center systems are in one-to-one correspondence with fan site selection positions, and the natural environment is utilized for heat dissipation; a heat dissipation system is further arranged and comprises an air cooling system and a liquid cooling system, the air cooling system dissipates heat through natural low-temperature air, the liquid cooling system dissipates heat through natural water circulation in a water storage pool on the mountain and under the mountain, the heat dissipation efficiency is improved, and energy consumption is reduced; according to the wind power generation power and the real-time power utilization demand of the intelligent calculation center, the charging and discharging strategy of the energy storage system is dynamically adjusted, stable power supply is ensured, resources of the intelligent calculation center are flexibly scheduled according to the task load, efficient utilization is achieved, energy consumption of the intelligent calculation center is remarkably reduced, the wind energy utilization rate is increased, and development of the green data center is promoted.

The invention provides a multi-network fusion heterogeneous complementary power supply system of a zero-carbon park / green data center, which is a highly integrated and innovative micro-grid architecture, introduces a'single-line looped network 'technology as a core, and is composed of a new energy power generation cluster single-line looped network, a park / data center direct-current power distribution single-line looped network and an alternating-current power distribution three-phase looped network. The new energy power generation cluster is connected in series with n power generation systems through a single-line looped network, and the looped network DC / DC converter, the DC / AC inverter and the DC input end of the multi-port bidirectional energy storage energy router form a closed loop. The DC power distribution single-line looped network is connected in series with the n DC / DC power distribution devices, the AC / DC rectifier and the DC / DC end of the energy storage router; the AC power distribution three-phase looped network is connected in parallel with the AC / AC power distribution equipment, the transformer and the DC / AC end of the energy storage router. According to the invention, the high-efficiency consumption of new energy is realized through a single-line looped network architecture, and the high-reliability and high-efficiency power supply requirements of a zero-carbon park / data center in a 100% new energy supply scene are met.

The invention relates to the technical field of data center energy conservation, in particular to a data center energy conservation and consumption reduction method and system.The data center energy conservation and consumption reduction system comprises a solar power generation module, a roof planting module, a bidirectional flow fresh air equipment module and a control module. The fresh air system operates depending on electric energy converted by the solar photovoltaic module, clean air on the planting roof is continuously conveyed into a room, the indoor environment is more comfortable, a good operation environment is provided for various precise instruments and equipment, the failure rate of the precise instruments and equipment is reduced, and therefore the service life of the precise instruments and equipment is prolonged. Self-sufficient supply of partial electricity of the data center is achieved, the energy efficiency is maximized, the energy consumption of the data center is effectively reduced, and the development concept of a green data center is met.

The invention provides a data center energy storage standby battery economic operation method based on multi-stage robust optimization. The method comprises the following steps: S101, setting a dynamic adjustable capacity range of an energy storage standby battery based on a data center standby capacity demand; s102, constructing a multi-stage robust optimization model considering multiple uncertainties of load, electricity price and renewable energy output based on the dynamic adjustable capacity range of the energy storage standby battery; and S103, converting the multi-stage robust optimization model into a two-stage robust optimization model through an affine transformation method, and solving by adopting a nested column and constraint generation algorithm to obtain a data center energy storage real-time optimal scheduling strategy. The operation method provided by the invention improves the operation robustness of the data center energy storage standby battery while ensuring the economical efficiency, and provides a theoretical basis for building a green data center.

The invention provides a data center efficient cooling energy supply system and method based on artificial intelligence, and belongs to the technical field of efficient cooling energy supply. The data acquisition module acquires multi-dimensional data of a data center and extracts core feature data. The multi-target optimization configuration module adopts a grey wolf optimization algorithm to set a multi-target cooling energy supply strategy and decomposes the multi-target cooling energy supply strategy into sub-targets. And the intelligent regulation and control module generates a cooling equipment operation parameter control instruction based on the sub-target, dynamically adjusts a cooling system combination mode and optimizes an energy distribution scheme. And the monitoring feedback module collects operation data of the cooling energy supply system in real time, performs deviation analysis with the sub-targets, evaluates the target achievement degree and forms a feedback adjustment signal. The cooling energy supply strategy is subjected to multi-objective optimization through the artificial intelligence algorithm, the equipment operation parameters and the energy distribution scheme are dynamically adjusted, the overall energy consumption of the cooling system is effectively reduced, the energy waste is reduced, the development trend of a green data center is met, and the operation cost is reduced for users.

The present application belongs to the technical field of cooling improvement of electrical equipment, and discloses a two-phase liquid cooling-Carnot cell system for data center cooling and energy storage, which comprises a boiling pool, a compressor, an expander, a pump, a water tank, a stop valve, an electronic expansion valve, a condenser and an evaporator; the boiling pool is used for cooling data center servers, the compressor and the water tank are used for recycling waste heat generated in the data center cooling process, and the expander is used for generating power by utilizing the waste heat. The system can switch among the natural cooling mode, the energy storage mode and the power generation mode according to the electricity price in the operation period. In the conventional period, the natural cooling mode is operated to cool the data center by air cooling; in the valley electricity period, the energy storage mode is operated to store waste heat of the data center cooling loop; and in the peak electricity period, the power generation mode is operated to convert the stored waste heat into electric energy. The present application combines the two-phase immersion liquid cooling technology with the Carnot cell energy storage technology, and provides a solution for the coupling of green data center cooling and energy storage systems.

The invention discloses a green data center power supply system large-model intelligent adjustment terminal load equipment power supply method, medium and equipment, and the method comprises the steps: collecting the reading of an intelligent electric meter in a cabinet corresponding to different terminal load equipment, obtaining the power consumption data per minute of the cabinet, and carrying out the serialization per minute; the method comprises the following steps: acquiring average power consumption of cabinets within a minutes after a certain time point, acquiring power consumption data within b minutes before the time point as initial features, adding timestamp features, acquiring power consumption data from the initial features of each cabinet, combining the power consumption data with the timestamp features to form input feature vectors, and gathering z input feature vectors into a sample data set; inputting the input feature vector of the zth cabinet at the current time point into the prediction model, and outputting an average power consumption prediction value within a minutes in the future; and obtaining the future total predicted power consumption of all cabinets, obtaining the net demand according to the power generation predicted value of the power supply system, comparing the energy storage of the energy storage module, and judging whether to access the power grid. According to the invention, the problem of pre-judging the energy consumption condition of the terminal load equipment is solved.

The invention discloses a sustainable management optimization method and system for an interconnected data center of a power system, and the method comprises the steps: carrying out the modeling of the migration distribution and processing of a data center network operation in an interconnected data center network model, and optimizing the resource distribution and energy consumption during demand response; establishing a green data center mathematical model integrating uncertain renewable energy sources based on the job migration requirement of the data center network job and the job processing model; the green data center mathematical model comprises a dynamic price signal of demand response, a load scheduling strategy and a user demand; solving the green data center mathematical model, and optimizing the response capability and benefit of the data center in demand response. According to the invention, the flexibility and energy efficiency of the data center in the demand response activity can be effectively improved, and the sustainable development of the green data center is promoted.

The invention relates to a green data center facade facing material aided design system and method. The system is developed based on a Visual Basic programming language, adopts a Windows application program architecture, and comprises a material data management module used for collecting and sorting basic data of exterior facing materials and providing dynamic classification and keyword retrieval functions of the materials; the material combination and cost calculation module is used for obtaining comprehensive cost information by combining material market price information according to the material combination configured by the user, the proportion of each material and the selection condition of the thermal insulation material; the visual display module is used for configuring a high-definition picture and detailed attribute description for each material so that a user can click the picture to check detailed attribute information of the material, and providing three-dimensional effect display of the material; and the design auxiliary module is used for providing a rapid comparison function of design schemes, and comparing the cost and effect of different schemes based on the real-time adjustment of material combination and proportion or thermal insulation material selection of a user.

The invention relates to the technical field of energy consumption management and resource scheduling optimization, and discloses a green data center energy-saving control method and system, and the method comprises the steps: obtaining real-time equipment operation data and real-time load data, and constructing an initial energy consumption model; dynamically adjusting model parameters according to the initial energy consumption model and the real-time load data to obtain a dynamic energy consumption model; according to the dynamic energy consumption model, optimizing a resource scheduling strategy to obtain a preliminary resource allocation scheme; dynamically adjusting the preliminary resource allocation scheme according to the real-time load data to obtain a dynamic resource allocation scheme; according to the dynamic resource allocation scheme and the real-time equipment operation data, performing allocation scheme optimization to obtain an optimized resource allocation scheme; and deploying and implementing the optimized resource allocation scheme, and performing real-time dynamic fine adjustment to obtain a real-time resource allocation scheme. According to the method, self-adaptive adjustment of dynamic conditions of load surge or equipment isomerism enhancement can be realized.

The invention discloses a flywheel energy storage cooling system based on a green data center, relates to the technical field of flywheel energy storage cooling, solves the technical problem that heating detection cannot be performed according to heating states of different heating scenes in the prior art, and particularly relates to an energy storage operation detection module for performing operation detection on a flywheel energy storage system, and a control module for controlling the flywheel energy storage system. According to matching stage analysis of internal parts of the flywheel energy storage system, numerical value setting is carried out on the corresponding matching stage, statistics is carried out on the running time period of the flywheel energy storage system at all moments, and the running time period is divided into an abnormal running time period and a normal running time period according to the statistical numerical value; the heating scene analysis unit is used for performing heating scene analysis on the flywheel energy storage system in different operation periods, and obtaining various types of the operation periods according to the scene analysis; and the cooling layout analysis and optimization module is used for analyzing and optimizing the cooling layout of the flywheel energy storage system.

The invention discloses a green data center end-to-end intelligent operation and maintenance voltage load adjusting method, system and device and a medium, and belongs to the field of data center energy saving, aiming at the problem that the CPU current frequency of a using terminal is influenced by large fluctuation of input voltage and input current caused by natural factors such as wind power and solar energy. The method comprises the following steps of: subtracting computing power required by a running service in a CPU from total computing power of the CPU to obtain current idle computing power of the CPU, when the power supply voltage suddenly changes, if the idle computing power of the CPU is lower than a set abnormal computing power threshold value, judging the sudden change of the power supply voltage as abnormal sudden change, sending a voltage regulation instruction, and regulating the voltage of the power supply voltage; evaluating the performance of all the CPUs, and dividing the CPUs into a plurality of groups according to the performance scores of the CPUs; and each group of CPUs respectively updates an abnormal computing power threshold value for the next round of voltage regulation. According to the invention, the energy utilization rate can be optimized, the false alarm rate of abnormality diagnosis is reduced, and the accuracy of voltage regulation is improved.

The present invention discloses a scenario-based short-term wind speed prediction method and system for green data centers, the method comprising the following steps: collecting historical weather data, clustering the historical weather data, and obtaining several weather scenarios; constructing Copula functions of wind speed and other meteorological factors in each scenario under different weather scenarios, and determining the optimal input variables of wind speed and other meteorological factors based on the Copula function; constructing an LSTM neural network model, and training the LSTM neural network model based on historical weather data and optimal input variables to obtain a wind speed prediction model; obtaining weather factor data of the scenario to be predicted, and using the wind speed prediction model to predict the wind speed data of the next period. The present invention effectively solves the problem that a single scenario prediction method is difficult to accurately capture the law of wind speed changes when facing complex and changeable weather parameters by introducing technical means such as fuzzy clustering, cluster number evaluation, and Copula function correlation analysis.

The invention provides an intelligent computing power cluster energy consumption prediction modeling method and system, and a load prediction method and system, and the method comprises the steps: screening out high-correlation main influence factors, such as temperature, humidity, air pressure, load, electricity price and the like, according to the initial load data of cluster computing power; carrying out IMF feature fluctuation extraction and dimension reduction processing on the high influence factor time series data, and then constructing an EMD-t-SNE-LSTM load prediction model; and then an EMD-t-SNE-LSTM load prediction model is used to carry out energy consumption prediction and performance optimization management in stages of day-ahead, intra-day and real-time for the intelligent computing power cluster so as to realize accurate load prediction under the high-performance computing cluster, promote computing power allocation and resource scheduling of an intelligent computing center, improve energy efficiency and reduce operation cost. According to the method, the flexible response capability of a power system is enhanced and the construction of a green data center and the development of an intelligent power grid are promoted through the directions of energy consumption prediction, load characteristic analysis, power grid interaction potential and the like which can be suitable for supporting the intelligent computing power cluster.