Data center electric power-thermal scheduling method and system combined with building virtual energy storage

Abstract

The invention belongs to the technical field of data center refrigeration, and discloses a data center electric power-heating power scheduling method combined with building virtual energy storage, and the method comprises the following steps: obtaining the historical electricity consumption, refrigerating capacity, indoor temperature and other data of a data center; according to the correlation between the electrical load and the refrigerating capacity, the future refrigeration demand is predicted by predicting the electricity consumption; the refrigeration demand and the electricity price change are judged, the refrigerating capacity is adjusted in advance through building virtual energy storage and the peak-valley electricity price difference, timely response of the indoor temperature of the data center to the heat dissipation demand change is achieved, energy waste is avoided, and the system stability is improved. From two aspects of electric power and heating power, the electric power-heating power collaborative optimal scheduling method is constructed by taking the lowest operation cost as an objective function and taking building thermal inertia, indoor temperature range, heat dissipating capacity and the like as constraint conditions. According to the method, the system stability is improved, and the operation cost is remarkably reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of data center refrigeration, and in particular to a data center power-heat scheduling method and system combined with building virtual energy storage. Background Art

[0002] The widespread application of big data and artificial intelligence technologies has promoted the development of data centers. In data centers, IT equipment generates a lot of heat during operation, which becomes one of the factors restricting its efficient and stable operation. The large demand for cooling increases the operating costs of data centers. Therefore, the optimization and intelligence of the cooling system is particularly important.

[0003] The power load of a data center usually fluctuates over time, and its cooling demand also changes accordingly. When adjusting the cooling capacity, the indoor temperature changes are often monitored first, and then the cooling capacity is adjusted. However, buildings have dynamic characteristics of cold / heat. When the building obtains cold or heat, the indoor temperature changes do not respond in time, and it often takes a certain amount of time to adjust to the set temperature. In other words, the building's cooling / heating system has inertia. If the cooling system relies on a simple temperature feedback mechanism to adjust the cooling capacity, that is, to increase or decrease the cooling capacity according to the deviation between the indoor temperature and the set value, in actual operation, it is often difficult to quickly and accurately reach and maintain the indoor temperature within the ideal range. This adjustment lag may threaten the performance stability and service life of IT equipment, and may also cause unnecessary energy waste, thereby pushing up the operating costs of the data center.

[0004] Through the above analysis, the problems and defects of the prior art are as follows:

[0005] The cooling systems of existing data centers often fail to take into account the thermal inertia of the building, resulting in untimely and inflexible indoor temperature adjustment, which increases operating costs. Summary of the invention

[0006] In view of the problems existing in the prior art, the present invention provides a data center power-heat scheduling method combined with building virtual energy storage.

[0007] The present invention is implemented as follows: a data center power-heat scheduling method combined with building virtual energy storage includes:

[0008] Step 1: Obtain the historical power load data of the data center and predict the power consumption in the future based on the historical data;

[0009] Step 2, respectively obtain the time-shiftable load and the non-time-shiftable load from the historical total power load of the data center, and redistribute the time-shiftable load to the lowest electricity price time within the time range of the time-shiftable load;

[0010] Step 3: Analyze and model the correlation between the historical power load, cooling demand, and indoor temperature of the data center under different meteorological conditions;

[0011] Step 4, predicting the cooling demand in the future based on the data obtained in steps 1, 2, and 3;

[0012] Step 5, determining whether the cooling demand will increase, if yes, executing step 6; if no, executing step 8;

[0013] Step 6, determine whether the electricity price will rise in the future, if yes, go to step 7; if not, go to step 8;

[0014] Step 7: Increase the cooling capacity of the data center in advance;

[0015] Step 8: Reduce cooling of the data center in advance;

[0016] Step 9: Monitor whether the air temperature in the data center is within the appropriate range; if not, increase or decrease the cooling capacity again according to the actual situation;

[0017] Step 10, collect the data of the above steps to more accurately predict the cooling demand and improve the scheduling strategy in the next scheduling.

[0018] Furthermore, in step 2, when the time-shiftable load is redistributed, the change in electricity load within the allocatable time period is taken into consideration, and the total time-shiftable load is evenly transferred to the time period with lower electricity price and lower total load to avoid untimely heat dissipation due to sudden increase in heat dissipation.

[0019] Furthermore, in step 7, if the cooling demand will increase and the electricity price rises, the cooling capacity is increased in advance by utilizing the thermal inertia of the building and the appropriate indoor temperature range, and the indoor temperature is adjusted to a lower level to ensure that when the heat dissipation demand increases, the indoor temperature is still controlled within an appropriate range; in this step, a portion of the cooling electricity load is transferred by utilizing the peak-valley electricity price difference, thereby reducing costs.

[0020] Furthermore, in step 8, if the cooling demand increases and the electricity price decreases, the cooling capacity is reduced in advance, but still within the indoor temperature range required by the data center; the cooling capacity is increased during the period when the electricity price is lower to meet the heat dissipation demand; if the cooling demand decreases, the cooling capacity is reduced regardless of whether the electricity price increases or decreases.

[0021] Furthermore, in step 10, the situation of each scheduling is recorded, including indoor temperature, cooling capacity, power consumption, and indoor temperature response speed. Combined with big data analysis or machine learning methods, the lowest operating cost is used as the objective function, and the building thermal inertia, indoor temperature range, and heat dissipation are used as constraints to construct an optimized scheduling method for power-heat synergy.

[0022] Another object of the present invention is to provide a data center power-heat dispatching system combined with building virtual energy storage, comprising:

[0023] The power consumption prediction module is used to obtain the historical power load data of the data center and predict the power consumption in the future based on the historical data;

[0024] A transfer module is used to obtain the time-shiftable load and the non-time-shiftable load from the historical total power load of the data center, and redistribute the time-shiftable load to the lowest electricity price moment within the time range of the time-shiftable load;

[0025] Modeling module, used to analyze and model the correlation between historical power load, cooling demand and indoor temperature of data centers under different meteorological conditions;

[0026] The cooling demand forecasting module is used to forecast the cooling demand in the future based on the obtained data and to determine whether the cooling demand will increase;

[0027] The electricity price judgment module is used to judge whether the electricity price will rise in the future; increase the cooling capacity of the data center in advance; reduce the cooling capacity of the data center in advance;

[0028] The monitoring module is used to monitor whether the air temperature in the data center is within the appropriate range; if not, the cooling capacity is increased or decreased again according to the actual situation;

[0029] The data collection module is used to collect data from the above steps to more accurately predict the cooling demand and improve the scheduling strategy in the next scheduling.

[0030] Another object of the present invention is to provide a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the data center power-heat scheduling method combined with building virtual energy storage.

[0031] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the data center power-heat scheduling method combined with building virtual energy storage.

[0032] Another object of the present invention is to provide an information data processing terminal, which is used to implement the data center power-heat scheduling system combined with building virtual energy storage.

[0033] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0034] First, the data center power-heat scheduling method of the present invention is implemented in two ways with the goal of minimum cost. First, from the power aspect, by transferring the time-shiftable load to the period with lower electricity prices, peak load reduction and valley filling are achieved, which reduces the volatility of overall power demand and reduces operating costs. Second, from the thermal aspect, by utilizing the thermal inertia of the building, using the building as an energy storage carrier, and combining the difference in electricity prices, part of the cooling is transferred to the period with low electricity prices, which also reduces costs.

[0035] The present invention uses building virtual energy storage, historical big data analysis and future cooling capacity prediction to achieve a faster response to indoor temperature, ensuring that the temperature remains within the set range when the cooling demand changes. Compared with systems that do not take into account the thermal inertia of the building, the data center of the present invention has increased operational stability and reliability.

[0036] The technical problems of the prior art solved by the present invention are:

[0037] 1. Low temperature control accuracy and speed: In the prior art, a simple temperature feedback mechanism is often used to adjust the cooling capacity, resulting in insufficient temperature control. The present invention uses the building's virtual energy storage, historical big data analysis, and future cooling capacity prediction to quickly respond to indoor temperature changes and ensure that the temperature remains within the set range when the heat dissipation demand changes.

[0038] 2. High operating cost: The present invention transfers part of the work to the off-peak electricity price period through electricity-heat optimization scheduling, fully utilizes the low-cost electricity during the off-peak period, and significantly reduces electricity expenses.

[0039] 3. Low system reliability: Compared with the system that does not consider the thermal inertia of the building, the present invention adjusts the indoor temperature in time through more flexible power-heat scheduling, improves the operating stability and reliability of the data center, and reduces the impact of temperature fluctuations on the performance and life of IT equipment.

[0040] 4. Lack of intelligent management: Combining big data and artificial intelligence technology, the present invention can realize intelligent management of the refrigeration system, monitor and predict the load and heat dissipation requirements of the data center in real time, automatically adjust the refrigeration strategy, and improve management efficiency.

[0041] The significant technical progress achieved by the present invention:

[0042] 1. Reduce operating costs: Through power-heat optimization scheduling, the present invention can transfer part of the workload to the period with lower electricity prices, make full use of low-cost electricity during the period of low electricity prices, and significantly reduce electricity costs.

[0043] 2. Improve energy efficiency: Use buildings as energy storage carriers, consider fluctuations in electricity prices and cooling demand, adjust cooling capacity in a timely manner, stabilize indoor temperature within the set range, improve the efficiency of the refrigeration system, and reduce energy waste.

[0044] 3. Improve operational stability and reliability: Compared with the traditional system that does not consider the thermal inertia of the building, the present invention uses more flexible and intelligent power-heat scheduling, and the temperature control is more accurate and timely, so as to more effectively respond to load changes and reduce equipment failures and maintenance costs caused by temperature fluctuations.

[0045] 4. Intelligent management: Combining big data and artificial intelligence technology, the present invention realizes intelligent management of the refrigeration system. Through the data accumulated in step 10, the power-heat scheduling scheme can be continuously optimized and improved to achieve better control of data center temperature and minimize costs.

[0046] Through these technological advances, the present invention effectively solves key problems in the prior art, improves the response speed of the indoor temperature of the data center, reduces operating costs, and provides an advanced solution for the data center refrigeration system.

[0047] Second, the expected benefits and commercial value of the technical solution of the present invention after transformation are as follows: The present invention uses building virtual energy storage to achieve advance adjustment by predicting the cooling capacity, ensuring that the indoor temperature can respond quickly when the heat dissipation demand changes, thereby realizing intelligent management of the refrigeration system. This improves energy utilization efficiency, reduces operating costs, and has economic benefits. In addition, the effective regulation of cooling capacity avoids waste of cooling capacity and reduces carbon emissions, which meets the society's requirements for sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of a data center power-heat scheduling method combined with building virtual energy storage provided by an embodiment of the present invention.

[0049] Figure 2 It is a detailed flow chart of a data center power-heat scheduling method combined with building virtual energy storage provided by an embodiment of the present invention.

[0050] Figure 3 It is a structural block diagram of a data center power-heat dispatching system combined with building virtual energy storage provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] The power-heat dispatching method of the present invention, combined with the building virtual energy storage system, effectively solves many problems in the prior art in the energy consumption management of the data center and achieves significant technological progress:

[0053] 1. High power consumption and inflexible scheduling

[0054] Technical issues: Existing data centers consume huge amounts of electricity, but load scheduling is inflexible and difficult to adjust in real time according to changes in electricity prices and fluctuations in electricity demand, resulting in high operating costs.

[0055] Technological progress: The present invention intelligently allocates time-shiftable loads through historical load data analysis and future load demand forecasting, and schedules them during the period of low electricity prices, thereby reducing electricity costs and making load scheduling more flexible and efficient.

[0056] 2. The refrigeration system consumes a lot of energy and cannot effectively predict demand

[0057] Technical problem: The cooling demand of data centers varies with external climate and internal load. Existing systems cannot accurately predict future cooling demand, resulting in energy waste.

[0058] Technological progress: The present invention builds models based on meteorological conditions, load data and historical data of indoor temperature to achieve accurate prediction of cooling demand, avoid unnecessary cooling, improve energy efficiency and reduce energy waste.

[0059] 3. Lack of dispatch strategies to cope with electricity price fluctuations

[0060] Technical issues: Existing data centers lack flexible scheduling methods to cope with changes in electricity prices. Cooling costs often increase during peak electricity price periods, and the reserved cooling capacity cannot be fully utilized during the low electricity price period.

[0061] Technological progress: The present invention adjusts the cooling capacity in advance through electricity price forecasting, increases cooling reserves during low electricity price periods, realizes "virtual energy storage", and reduces electricity load during peak electricity price periods, thereby significantly saving operating costs.

[0062] 4. It is difficult to achieve real-time dynamic temperature monitoring

[0063] Technical issues: The existing system does not have sufficient temperature monitoring and is unable to dynamically adjust cooling based on temperature changes, resulting in unsatisfactory temperature control and the risk of equipment overheating.

[0064] Technological progress: The present invention ensures that the air temperature is within an appropriate range by real-time monitoring of the data center temperature and dynamically adjusting the cooling capacity. This not only improves the operating stability of the equipment, but also avoids equipment damage caused by insufficient temperature control.

[0065] 5. Scheduling strategies lack self-optimization capabilities

[0066] Technical issues: Traditional scheduling solutions often lack self-learning and improvement capabilities and are unable to adapt to future demand changes and continuously optimize strategies.

[0067] Technological progress: The present invention collects various data during the scheduling process, supports self-learning and continuous optimization of scheduling strategies, makes prediction and control more accurate, and further improves the intelligence and efficiency of the system.

[0068] 6. Difficult to operate efficiently under fluctuations in demand, weather and electricity prices

[0069] Technical issues: Data centers are affected by multiple factors, including load demand, external weather and electricity price fluctuations. Existing systems are difficult to achieve efficient coordination, resulting in high energy consumption and operating costs.

[0070] Technological progress: This invention comprehensively considers the dynamic changes of multiple factors, uses historical data and real-time feedback to optimize the scheduling plan, ensures that the system can still operate stably and efficiently under complex conditions, and greatly improves the energy-saving effect and cost control ability of the data center.

[0071] like Figure 1 , 2 As shown, an embodiment of the present invention provides a data center power-heat scheduling method combined with building virtual energy storage, comprising the following steps:

[0072] S101, obtaining historical power load data of the data center, and predicting power consumption in the future based on the historical data;

[0073] S102, respectively obtaining the time-shiftable load and the non-time-shiftable load from the historical total power load of the data center, and reallocating the time-shiftable load to the time with the lowest electricity price within the time range of the time-shiftable load;

[0074] S103, analyzing and modeling the correlation between the historical power load, cooling demand and indoor temperature of the data center under different meteorological conditions;

[0075] S104, predicting the refrigeration demand in the future period based on the data obtained in S101, S102, and S103;

[0076] S105, determining whether the cooling demand will increase, if yes, executing S106; if no, executing S108;

[0077] S106, judging whether the electricity price will rise in the future, if yes, executing S107; if no, executing S108;

[0078] S107, increasing the cooling capacity of the data center in advance;

[0079] S108, reduce cooling of the data center in advance;

[0080] S109, monitoring whether the air temperature of the data center is within a suitable range; if not, increasing or decreasing the cooling capacity according to the actual situation;

[0081] S110, collecting data from the above steps to more accurately predict cooling demand and improve scheduling strategies in the next scheduling.

[0082] The power-heat dispatching method provided in the embodiment of the present invention combines the building virtual energy storage with the power load and heat demand of the data center to improve the efficiency of power resource utilization and save operating costs. By accurately predicting the power consumption and cooling demand of the data center, optimizing the dispatching strategy, and flexibly adjusting the cooling system load when the electricity price and climate conditions change, efficient energy saving is achieved. The following is the detailed working principle of the method:

[0083] 1. Historical data collection and load forecasting

[0084] In step S101, the historical power load data of the data center is first collected, and the power consumption in a specific time period in the future is predicted by analyzing the past power consumption records of the data center. Based on the load trends and rules of historical data, a power load prediction model can be generated using time series analysis or machine learning methods. This model provides basic data for the subsequent scheduling process to ensure that the scheduling plan can adapt to future load demand fluctuations.

[0085] 2. Time-shifted load optimization distribution

[0086] In step S102, time-shiftable loads and non-time-shiftable loads are separated from the historical loads of the data center. Time-shiftable loads refer to power demands that can be flexibly adjusted within a certain time range, such as delayed processing of some computing tasks or non-urgent device charging. Reallocating time-shiftable loads to the time period with the lowest electricity price can effectively save energy costs and reduce the pressure on the power grid during peak hours. Non-time-shiftable loads are continuously supplied during their demand period to ensure the basic operational stability of the data center.

[0087] 3. Modeling of meteorological factors on load demand

[0088] In step S103, the correlation between the power load, cooling demand and indoor temperature of the data center under different meteorological conditions is analyzed. As the external temperature changes, the heat dissipation demand of the data center and the difficulty of indoor temperature control also change accordingly. By establishing a model between meteorological conditions and cooling demand and using meteorological forecast data, the cooling demand of the data center in the future time period can be more accurately predicted, so as to deal with the impact of external climate fluctuations in advance in thermal scheduling.

[0089] 4. Refrigeration demand forecast

[0090] In step S104, the future data center cooling demand is predicted based on the historical load, meteorological conditions and temperature regulation demand data. The prediction model takes into account the fluctuation of external temperature, the change of power load and the regulation demand of the internal temperature of the data center to ensure the accuracy of the prediction results. Through this prediction, the system can increase or decrease the cooling capacity in advance during high temperature, low temperature or peak power consumption to balance the heat load.

[0091] 5. Determine changes in cooling demand

[0092] In step S105, the system determines whether the cooling demand will increase in the future. If the forecast shows that the cooling demand will increase, it is necessary to evaluate whether to increase the cooling capacity in advance; if the demand will not increase, there is no need to adjust or reduce the cooling load. The judgment in this step enables the system to flexibly respond to dynamic changes in load, avoid unnecessary cooling costs, and ensure that the operating environment of the data center is always within an appropriate temperature range.

[0093] 6. Impact of electricity price changes on cooling scheduling

[0094] In step S106, the fluctuation of electricity prices is further predicted. If the electricity price has an upward trend in the future, the cooling capacity is increased in advance (S107), and the cooling load is increased during the period of low electricity prices, and the "cold capacity" is stored to meet the cooling needs of the data center during the period of high electricity prices. This "virtual energy storage" strategy reduces electricity consumption during high electricity price periods by optimizing the time allocation of electricity and cooling loads, thereby improving the cost-effectiveness of data center operations.

[0095] 7. Real-time monitoring of temperature control

[0096] In step S109, the system continuously monitors the air temperature in the data center to ensure that it always remains within the predetermined safety range. If the temperature is detected to deviate from the normal range, the cooling capacity is immediately increased or decreased according to the actual situation to prevent the equipment from overheating or low temperature from affecting normal operation. This dynamic adjustment mechanism provides real-time feedback for the cooling system of the data center to ensure that the operating environment is always in the best state.

[0097] 8. Data collection and scheduling strategy optimization

[0098] In step S110, the system collects data from each of the above steps, including load forecasting, time-shifted load allocation, cooling capacity adjustment, and real-time changes in meteorological conditions. These data are used to optimize subsequent scheduling strategies, improve prediction accuracy and control efficiency, and ensure that higher energy utilization and cost savings can be achieved in future scheduling processes, thereby enabling the scheduling system to have the ability to self-learn and continuously improve.

[0099] In S102 provided in the embodiment of the present invention, when the time-shiftable load is reallocated, the change in electricity load within the allocatable time period is taken into consideration, and the total time-shiftable load is evenly transferred to the time period with lower electricity price and lower total load to avoid untimely heat dissipation due to sudden increase in heat dissipation.

[0100] In S107 provided in the embodiment of the present invention, if the cooling demand is about to increase and the electricity price rises, the cooling capacity is increased in advance by utilizing the thermal inertia of the building and the appropriate indoor temperature range, and the indoor temperature is adjusted to a lower level to ensure that when the heat dissipation demand increases, the indoor temperature is still controlled within an appropriate range; in this step, a portion of the cooling electricity load is transferred by utilizing the peak-valley electricity price difference, thereby reducing costs.

[0101] In S108 provided in the embodiment of the present invention, if the cooling demand increases and the electricity price decreases, the cooling capacity is reduced in advance, but still needs to be within the indoor temperature range required by the data center; the cooling capacity is increased during the period when the electricity price is lower to meet the heat dissipation demand; if the cooling demand decreases, the cooling capacity is reduced regardless of whether the electricity price increases or decreases.

[0102] In S110 provided in the embodiment of the present invention, the situation of each scheduling is recorded, including indoor temperature, cooling capacity, power consumption, and indoor temperature response speed. In combination with big data analysis or machine learning methods, the lowest operating cost is used as the objective function, and the building thermal inertia, indoor temperature range, and heat dissipation are used as constraints to construct an optimized scheduling method for power-heat synergy.

[0103] like Figure 3 As shown, an embodiment of the present invention provides a data center power-heat dispatching system combined with building virtual energy storage, including:

[0104] The power consumption prediction module is used to obtain the historical power load data of the data center and predict the power consumption in the future based on the historical data;

[0105] A transfer module is used to obtain the time-shiftable load and the non-time-shiftable load from the historical total power load of the data center, and redistribute the time-shiftable load to the lowest electricity price moment within the time range of the time-shiftable load;

[0106] Modeling module, used to analyze and model the correlation between historical power load, cooling demand and indoor temperature of data centers under different meteorological conditions;

[0107] The cooling demand forecasting module is used to forecast the cooling demand in the future based on the obtained data and to determine whether the cooling demand will increase;

[0108] The electricity price judgment module is used to judge whether the electricity price will rise in the future; increase the cooling capacity of the data center in advance; reduce the cooling capacity of the data center in advance;

[0109] The monitoring module is used to monitor whether the air temperature in the data center is within the appropriate range; if not, the cooling capacity is increased or decreased again according to the actual situation;

[0110] The data collection module is used to collect data from the above steps to more accurately predict the cooling demand and improve the scheduling strategy in the next scheduling.

[0111] Another object of the present invention is to provide a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the data center power-heat scheduling method combined with building virtual energy storage.

[0112] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the data center power-heat scheduling method combined with building virtual energy storage.

[0113] Another object of the present invention is to provide an information data processing terminal, which is used to implement the data center power-heat scheduling system combined with building virtual energy storage.

[0114] The data center power-heat dispatching system combined with building virtual energy storage provided by the embodiment of the present invention aims to achieve accurate dispatching of power and cooling requirements of the data center through the coordination of intelligent modules, save energy and reduce operating costs. The working principle of the system is as follows:

[0115] 1. Power consumption forecasting and dynamic load management

[0116] The power consumption prediction module obtains and analyzes the historical power load data of the data center, and uses the historical data model to predict the power consumption in the future. The predicted load demand serves as the basic data for subsequent scheduling, helping the system to determine in advance when the peak or valley of power consumption may occur. The load prediction results will be used by the transfer module to optimize the scheduling of time-shiftable loads and achieve a balance between peak reduction and valley power consumption.

[0117] 2. Time-shiftable load allocation and electricity price optimization

[0118] The task of the transfer module is to identify time-shiftable loads (such as some computing tasks) and non-time-shiftable loads from historical load data. Based on the predicted electricity price, the time-shiftable loads are allocated to the low-price period, thereby reducing electricity costs and the impact of peak loads on the power grid. This allocation process enables data centers to better cope with electricity price fluctuations and achieve efficient electricity management.

[0119] 3. Meteorological factor modeling and cooling demand forecasting

[0120] The modeling module builds a weather and load forecasting model based on the correlation between the historical power consumption, cooling demand and indoor temperature of the data center under different meteorological conditions. The model combines the impact of external temperature, humidity and other factors on cooling demand, allowing the system to flexibly adjust the cooling capacity under changing climate conditions. The cooling demand forecasting module uses this model to estimate the cooling demand of the data center in a specific time period in the future and schedule the cooling system in advance to cope with the upcoming load changes.

[0121] 4. Linking refrigeration scheduling with electricity price forecasting

[0122] The electricity price judgment module determines whether the electricity price will rise in the future by predicting the electricity price trend, and makes advance adjustments to the load of the refrigeration system based on this judgment. When the electricity price rises, the system increases the cooling capacity reserve during the low-price period and uses "virtual energy storage" to cool the data center in advance, so that it can reduce electricity demand during the high-price period. This strategy effectively reduces the operating costs during the peak electricity price period.

[0123] 5. Real-time monitoring and dynamic temperature control

[0124] The monitoring module monitors the air temperature in the data center in real time to ensure that the temperature is always within the set safety range. When the temperature deviates from the appropriate range, the system adjusts the cooling capacity according to the current temperature control requirements, increases or decreases the cold air output, ensures that the equipment operates in the optimal ambient temperature, and prevents overheating or overcooling from affecting the stability of the data center. This dynamic temperature control mechanism helps improve the operating efficiency and safety of the data center.

[0125] 6. Data collection and scheduling strategy optimization

[0126] The data collection module records the actual operation data of each step of the dispatch system, including load forecasting, meteorological modeling, electricity price adjustment, and temperature control feedback. These data are used to improve the forecast accuracy and strategy optimization in the next round of dispatching, and realize the self-learning and continuous improvement capabilities of the dispatching plan. By collecting and analyzing the effects of previous dispatches, the system can adapt to future load and meteorological changes more intelligently and continuously improve the utilization efficiency of power and cooling resources.

[0127] In this embodiment, the method of inferring future data and the method of optimizing the scheduling scheme using multiple scheduling data include but are not limited to big data analysis and machine learning. It should be understood that the method of predicting future data based on historical data will be changed and adjusted according to actual conditions, such as factors such as data volume and result accuracy.

[0128] It should be noted that the prediction of cooling demand in the future in S104 does not limit the specific time length. The specific prediction time length should be determined based on the data obtained in S110 through continuous debugging to achieve the best effect. This will vary depending on the overall situation of the data center and the specific situation of the cooling system.

[0129] The present invention utilizes building virtual energy storage and considers the synergy of electricity and heat. Compared with a simple system that adjusts the cooling capacity based on a temperature feedback mechanism, the system can respond more quickly, improve the stability of the system, and avoid energy waste. In addition, the present invention takes the lowest operating cost as the goal, fully utilizes the cooling capacity generated by off-peak electricity prices, and the dynamic characteristics of cooling / heat of the building, and significantly reduces the operating cost by predicting and adjusting the cooling capacity in advance and reasonably allocating the power load.

[0130] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0131] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A data center power-heat dispatching method combined with building virtual energy storage, characterized in that: The data center power-heat scheduling method combined with building virtual energy storage comprises the following steps: Step 1: Obtain the historical power load data of the data center and predict the power consumption in the future based on the historical data; Step 2, respectively obtain the time-shiftable load and the non-time-shiftable load from the historical total power load of the data center, and redistribute the time-shiftable load to the lowest electricity price time within the time range of the time-shiftable load; Step 3: Analyze and model the correlation between the historical power load, cooling demand, and indoor temperature of the data center under different meteorological conditions; Step 4, predicting the cooling demand in the future based on the data obtained in steps 1, 2, and 3; Step 5, determining whether the cooling demand will increase, if yes, executing step 6; if no, executing step 8; Step 6, determine whether the electricity price will rise in the future, if yes, go to step 7; if not, go to step 8; Step 7: Increase the cooling capacity of the data center in advance; Step 8: Reduce cooling of the data center in advance; Step 9: Monitor whether the air temperature in the data center is within the appropriate range; if not, increase or decrease the cooling capacity again according to the actual situation; Step 10, collect the data of the above steps to more accurately predict the cooling demand and improve the scheduling strategy in the next scheduling.

2. The data center power-heat dispatching method combined with building virtual energy storage as claimed in claim 1, characterized in that: In step 2, when the time-shiftable load is redistributed, the change in power load within the allocatable time period is considered, and the total time-shiftable load is evenly transferred to the time period with lower electricity price and lower total load to avoid untimely heat dissipation due to sudden increase in heat dissipation.

3. The data center power-heat dispatching method combined with building virtual energy storage as claimed in claim 1, characterized in that: In step 7, if the cooling demand will increase and the electricity price rises, the cooling capacity is increased in advance by utilizing the thermal inertia of the building and the appropriate indoor temperature range, and the indoor temperature is adjusted to a lower level to ensure that when the heat dissipation demand increases, the indoor temperature is still controlled within an appropriate range; in this step, a portion of the cooling electricity load is transferred by utilizing the peak-valley electricity price difference, thereby reducing costs.

4. The data center power-heat dispatching method combined with building virtual energy storage as claimed in claim 1, characterized in that: In step 8, if the cooling demand increases and the electricity price decreases, the cooling capacity is reduced in advance, but still within the indoor temperature range required by the data center; the cooling capacity is increased during the period when the electricity price is lower to meet the heat dissipation demand; if the cooling demand decreases, the cooling capacity is reduced regardless of whether the electricity price increases or decreases.

5. The data center power-heat dispatching method combined with building virtual energy storage as claimed in claim 1, characterized in that: In step 10, the situation of each scheduling is recorded, including indoor temperature, cooling capacity, power consumption, and indoor temperature response speed. Combined with big data analysis or machine learning methods, the lowest operating cost is used as the objective function, and the building thermal inertia, indoor temperature range, and heat dissipation are used as constraints to construct an optimized scheduling method for power-heat synergy.

6. A data center power-heat dispatching system combined with building virtual energy storage that implements the data center power-heat dispatching method combined with building virtual energy storage as described in any one of claims 1 to 5, characterized in that: The data center power-heat dispatching system combined with building virtual energy storage includes: The power consumption prediction module is used to obtain the historical power load data of the data center and predict the power consumption in the future based on the historical data; A transfer module is used to obtain the time-shiftable load and the non-time-shiftable load from the historical total power load of the data center, and redistribute the time-shiftable load to the lowest electricity price moment within the time range of the time-shiftable load; Modeling module, used to analyze and model the correlation between historical power load, cooling demand and indoor temperature of data centers under different meteorological conditions; The cooling demand forecasting module is used to forecast the cooling demand in the future based on the obtained data and to determine whether the cooling demand will increase; The electricity price judgment module is used to judge whether the electricity price will rise in the future; increase the cooling capacity of the data center in advance; reduce the cooling capacity of the data center in advance; The monitoring module is used to monitor whether the air temperature in the data center is within the appropriate range; if not, the cooling capacity is increased or decreased again according to the actual situation; The data collection module is used to collect data from the above steps to more accurately predict the cooling demand and improve the scheduling strategy in the next scheduling.

7. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the data center power-heat scheduling method combined with building virtual energy storage as described in any one of claims 1-5.

8. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the data center power-heat scheduling method combined with building virtual energy storage as described in any one of claims 1 to 5.

9. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the data center power-heat dispatching system combined with building virtual energy storage as described in claim 7.

Images