Data center system and management method of data center system

Abstract

The invention relates to the technical field of data center management, in particular to a data center system and a management method of the data center system. By integrating various renewable energy sources and energy storage equipment, complementation and optimal utilization of energy sources are achieved, energy waste is reduced, the energy utilization efficiency is improved, the renewable energy sources and efficient energy conversion equipment are adopted, dependence on traditional fossil energy is reduced, and therefore carbon emission is reduced, low-carbon transformation of the data center can be achieved easily, and the energy utilization efficiency of the data center is improved. The system capacity configuration is solved by utilizing an improved multi-target locust optimization algorithm, so that a system configuration scheme with optimal comprehensive performance in the aspects of energy, environment and economy can be found, and the reliability and the stability of the system are improved; the data center system has the remarkable advantages of improving energy efficiency, reducing carbon emission, optimizing system configuration, enhancing reliability, saving cost and the like, and sustainable development of the data center industry is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of data center management, and in particular to a data center system and a management method for the data center system. Background Art

[0002] With the rapid development of information technology, data centers, as an important infrastructure of the digital society, are expanding in size. However, the energy consumption and carbon emissions of data centers are becoming increasingly serious. Traditional data centers usually rely on local power grids or specialized fossil power generation facilities to supply electricity, which not only leads to a large amount of energy consumption, but also generates a large amount of carbon emissions, causing great pressure on the environment.

[0003] Under the background of the "carbon peak and carbon neutrality" strategy and energy transformation, it is imperative to find a low-cost and sustainable data center energy supply solution. The multi-energy complementary distributed energy system (DMES) introduces renewable energy and energy storage technology, which can meet the cooling, heating and electricity load requirements of data centers with a higher energy utilization rate, providing an important solution for the low-carbon transformation of data center energy supply systems.

[0004] However, DMES involves multiple energy sources, multiple devices, and multiple energy conversion processes, and its design and operation optimization issues are relatively complex. At present, although research on data center DMES optimization has made some progress, there are still some problems that need to be solved. For example, in DMES considering renewable energy coupling, the system optimization model needs to further consider key factors such as the proportion of renewable energy and the reliability of data center energy supply; in addition, there is a relative lack of research on how to effectively deal with the problem of large cooling load and electricity load demand in data centers. Therefore, a data center system and a management method for data center system are proposed to address the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide a data center system and a management method of the data center system to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A data center system, comprising:

[0008] Infrastructure: The infrastructure includes energy supply system and energy transmission system; the energy supply system includes environmental energy module, conventional energy module, energy conversion equipment and energy storage equipment;

[0009] Data collection end: The data collection end includes an energy data collection module and an equipment operation data module arranged corresponding to the energy supply system and the energy transmission system, and obtains energy data and equipment operation data;

[0010] Processing center: The processing center includes data sorting module, data analysis module and result display module;

[0011] The data sorting module receives energy data and equipment operation data from the data collection end for data sorting and storage;

[0012] The data analysis module establishes a communication connection with the data sorting module, analyzes the performance and efficiency of the equipment according to the equipment operation data, and evaluates the operation status of the equipment; analyzes the energy consumption according to the energy data, and calculates indicators such as life cycle cost, energy consumption and carbon emissions; combines the operation strategy and uses the improved multi-objective locust optimization algorithm to solve the system capacity configuration; predicts the electrical load, cooling load and heating load according to the historical load data and environmental data;

[0013] The result display module displays the processing results of the data analysis module to the user in a visual manner.

[0014] Preferably, the environmentally friendly energy module includes a wind power system and a photovoltaic system. The wind power system includes a wind turbine installed at a suitable location around the data center, and the photovoltaic system includes a photovoltaic panel array installed on the top of the data center building or in the surrounding open space; the conventional energy module includes an internal combustion engine and a gas boiler; the energy conversion equipment includes waste heat recovery equipment, ground source heat pumps and absorption chillers arranged on the internal combustion engine and the gas boiler; the energy storage equipment includes batteries, energy storage equipment and ice storage tanks.

[0015] Preferably, the energy transmission system includes circuit transmission, cold transmission and heat transmission. Circuit transmission transmits the electricity generated by wind power systems, photovoltaic systems, and internal combustion engines to the electrical equipment in the data center through transmission lines, and is connected to the power grid at the same time; cold transmission transmits the cold generated by absorption chillers, ground source heat pumps, and ice storage systems to the data center through pipelines; heat transmission transmits the heat generated by waste heat recovery systems, gas boilers, and ground source heat pumps to the office and living areas through pipelines.

[0016] Preferably, the energy data acquisition module installs power sensors at the output ends of wind power systems, photovoltaic systems, internal combustion engines, and energy storage devices to collect power generation and power consumption data in real time; installs cooling sensors at the output ends of absorption refrigerators, ground source heat pumps, and ice storage systems to collect cooling capacity data; installs heat sensors at the output ends of waste heat recovery systems, gas boilers, and ground source heat pumps to collect heating data; installs solar radiation intensity sensors, wind speed sensors, and dry-bulb temperature sensors around the data center to collect solar radiation intensity, wind speed, and dry-bulb temperature.

[0017] Preferably, the equipment operation data module installs sensors on each device to collect the equipment's operating status, partial load rate, efficiency, and record the start and stop time and operating time of each device.

[0018] Preferably, the data sorting module receives energy data and equipment operation data from the data acquisition terminal, cleans, screens and integrates the received energy data and equipment operation data, removes abnormal data and duplicate data, and stores the sorted data in a database.

[0019] Preferably, the data analysis module analyzes the performance and efficiency of the equipment and evaluates the operating status of the equipment based on the equipment operating data; analyzes the energy consumption based on the energy data and calculates indicators such as life cycle cost, energy consumption and carbon emissions; combines the operating strategy and uses the improved multi-objective locust optimization algorithm to solve the system capacity configuration and optimize the system in terms of energy, environment and economy; uses data analysis and machine learning algorithms to establish a load forecasting model based on historical load data and environmental data to predict the changing trends of the data center electrical load, cooling load and heating load within a set time in the future.

[0020] A management method for a data center system includes a management method using the data center system.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] Improve energy efficiency: By integrating multiple renewable energy sources and energy storage equipment, we can achieve complementary and optimal use of energy, reduce energy waste, and improve energy efficiency;

[0023] Reduce carbon emissions: Use renewable energy and efficient energy conversion equipment to reduce dependence on traditional fossil energy, thereby reducing carbon emissions, helping to achieve low-carbon transformation of data centers and complying with environmental protection requirements;

[0024] Optimize system configuration: Use the improved multi-objective locust optimization algorithm to solve the system capacity configuration, find the system configuration solution with the best comprehensive performance in terms of energy, environment and economy, and improve the reliability and stability of the system;

[0025] Enhanced energy supply reliability: The addition of energy storage equipment and its connection to the power grid can improve the reliability of energy supply, reduce dependence on the power grid, and ensure that data centers can still operate stably under extreme conditions;

[0026] Realize refined management: Through the data collection terminal and processing center, it is possible to monitor and analyze energy data and equipment operation data in real time, realize refined management of data center energy, discover problems in time and make optimization adjustments;

[0027] Cost savings: By optimizing system configuration and operation strategies, life cycle costs, including initial investment, operation costs, and maintenance costs, are reduced, while the cost of purchasing electricity from the power grid is reduced, bringing economic benefits to data center operations;

[0028] Improve overall performance: Combining operation strategies and data analysis can optimize the system in terms of energy, environment, economy, etc., improve the overall performance of the data center, and make it more competitive. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The present invention is a schematic diagram of the overall structure of a data center system and a management method of the data center system. DETAILED DESCRIPTION

[0030] 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 accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0031] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0032] Example:

[0033] See also Figure 1 , this embodiment provides a technical solution:

[0034] A data center system, comprising:

[0035] Infrastructure: The infrastructure includes energy supply system and energy transmission system; the energy supply system includes environmental energy module, conventional energy module, energy conversion equipment and energy storage equipment;

[0036] The energy transmission system includes circuit transmission, cold transmission and heat transmission. Circuit transmission transmits the electricity generated by wind power system, photovoltaic system and internal combustion engine to the power equipment in the data center through transmission lines, and is connected to the power grid at the same time. Cold transmission transmits the cold generated by absorption chiller, ground source heat pump and ice storage system to the data center through pipelines. Heat transmission transmits the heat generated by waste heat recovery system, gas boiler and ground source heat pump to the office and living area through pipelines.

[0037] Environmentally friendly energy modules include wind power systems and photovoltaic systems. The wind power system includes wind turbines installed at appropriate locations around the data center, and the photovoltaic system includes photovoltaic panel arrays installed on the top of the data center building or in the surrounding open space. Conventional energy modules include internal combustion engines and gas boilers. Energy conversion equipment includes waste heat recovery equipment, ground source heat pumps and absorption chillers installed on internal combustion engines and gas boilers. Energy storage equipment includes batteries, energy storage equipment and ice storage tanks.

[0038] Data collection end: The data collection end includes an energy data collection module and an equipment operation data module arranged corresponding to the energy supply system and the energy transmission system, and obtains energy data and equipment operation data;

[0039] Among them, the energy data acquisition module installs power sensors at the output ends of wind power systems, photovoltaic systems, internal combustion engines, and energy storage equipment to collect power generation and power consumption data in real time; installs cooling sensors at the output ends of absorption chillers, ground-source heat pumps, and ice storage systems to collect cooling capacity data; installs heat sensors at the output ends of waste heat recovery systems, gas boilers, and ground-source heat pumps to collect heating data; installs solar radiation intensity sensors, wind speed sensors, and dry-bulb temperature sensors around the data center to collect solar radiation intensity, wind speed, and dry-bulb temperature;

[0040] The equipment operation data module installs sensors on each device to collect the equipment's operating status, partial load rate, efficiency, and record the start and stop time and operating time of each device;

[0041] Processing center: The processing center includes data sorting module, data analysis module and result display module;

[0042] The data sorting module receives energy data and equipment operation data from the data collection end for data sorting and storage;

[0043] The data analysis module establishes a communication connection with the data sorting module, analyzes the performance and efficiency of the equipment according to the equipment operation data, and evaluates the operation status of the equipment; analyzes the energy consumption according to the energy data, and calculates indicators such as life cycle cost, energy consumption and carbon emissions; combines the operation strategy and uses the improved multi-objective locust optimization algorithm to solve the system capacity configuration; predicts the electrical load, cooling load and heating load according to the historical load data and environmental data;

[0044] The result display module displays the processing results of the data analysis module to the user in a visual way. The specific steps include:

[0045] Data acquisition: The result display module obtains the processed result data from the data analysis module, including equipment performance data, energy consumption data, life cycle cost, energy consumption, carbon emissions, power purchase rate of the power grid, heat waste rate and other indicator data, as well as the forecast data of electric load, cooling load and heating load;

[0046] Select visualization tools: Choose appropriate visualization tools based on the characteristics of the data and presentation requirements, such as data visualization software, chart libraries, or visualization libraries in programming languages;

[0047] Design visualization layout: Determine the layout and style of visualization, including chart types (such as bar charts, line charts, pie charts, scatter plots, etc.), color matching, axis settings, titles and annotations, etc., to ensure a clear and beautiful display;

[0048] Draw charts: Use the selected visualization tools to convert the data into corresponding chart forms; for example, use a bar chart to show the performance comparison of different devices, a line chart to show the trend of energy consumption, and a pie chart to show the proportion of energy sources;

[0049] Integration and display: Integrate the drawn charts to form a complete visual interface; multiple charts and data can be displayed to users in the form of dashboards, reports or interactive interfaces;

[0050] Add interactive features: To enhance user experience and understanding, add interactive features such as hovering the mouse to display detailed data, zooming and panning charts, filtering and sorting data, etc., so that users can explore and analyze data more deeply;

[0051] Testing and optimization: Before displaying, test the visualization results to ensure the accuracy and readability of the charts; based on the test results, optimize and adjust the visualization to meet user needs;

[0052] Display to users: The final visualization results can be displayed to users through display screens, web interfaces or mobile applications, allowing users to intuitively view and understand the results of data analysis.

[0053] In this embodiment, the data sorting module receives energy data and equipment operation data from the data acquisition terminal, cleans, filters and integrates the received energy data and equipment operation data, removes abnormal data and duplicate data, and stores the sorted data in the database.

[0054] In this embodiment, the data analysis module analyzes the performance and efficiency of the equipment and evaluates the operating status of the equipment based on the equipment operation data; analyzes the energy consumption based on the energy data, and calculates indicators such as life cycle cost, energy consumption and carbon emissions; combines the operation strategy and uses the improved multi-objective locust optimization algorithm to solve the system capacity configuration and optimize the system in terms of energy, environment and economy; uses data analysis and machine learning algorithms to establish a load forecasting model based on historical load data and environmental data to predict the changing trends of the data center electrical load, cooling load and heating load within a set time in the future.

[0055] In this embodiment,

[0056] Furthermore, the specific steps of solving the system capacity configuration using the improved multi-objective locust optimization algorithm are as follows:

[0057] Problem definition and objective function setting:

[0058] Identify the decision variables for system capacity configuration, such as the capacity of wind power systems, photovoltaic systems, internal combustion engines, waste heat recovery equipment, ground source heat pumps, absorption chillers, batteries, thermal storage equipment, and ice storage tanks;

[0059] Set objective functions, including life cycle cost (LCC), life cycle carbon emissions (LCCE), life cycle energy consumption (LCEC), grid purchase rate (Rgrid) and heat waste rate (Rrest); these objective functions will comprehensively consider the energy, environmental and economic performance of the system;

[0060] Initialize the population:

[0061] Set IMOGOA parameters such as population size (N), number of iterations (tmax), and mutation rate;

[0062] The initial population is randomly generated, and each individual represents a system capacity configuration scheme;

[0063] Calculate fitness:

[0064] For each individual (system capacity configuration scheme), calculate its corresponding objective function value, i.e., fitness, according to the equipment model and system operation constraints;

[0065] Find non-dominated Pareto optimal solutions:

[0066] In the current population, find non-dominated individuals, that is, solutions that are not inferior to other individuals in all objective functions; these non-dominated solutions constitute the Pareto frontier;

[0067] Population variation:

[0068] Introduce population mutation mechanism to increase the algorithm's search capability and avoid falling into local optimality; specific mutation operations can be designed according to actual conditions, such as randomly perturbing certain decision variables of individuals;

[0069] Updated locust positions:

[0070] According to the mathematical model of IMOGOA, the position of each locust in the search space is updated; the position update of the locust is affected by the population force, gravity and wind force, while considering the linear reduction parameters and the optimal position;

[0071] Calculate the fitness of the new position:

[0072] For the updated locust position (i.e., the new system capacity configuration plan), recalculate its fitness, i.e., the objective function value;

[0073] Find nondominated solutions and update the bucket:

[0074] In the new population, find the non-dominated Pareto optimal solution again and update it to the storage set; the storage set is used to save the optimal solution found during the algorithm iteration process;

[0075] Determine whether the maximum number of iterations has been reached:

[0076] If the maximum number of iterations has not been reached, return to step 5 to continue iterating; otherwise, proceed to the next step;

[0077] Output the optimal solution set:

[0078] After the algorithm is finished, the optimal solution set in the storage set is output, that is, a set of system capacity configuration plans with good comprehensive performance in terms of energy, environment and economy.

[0079] Furthermore, the specific steps of using data analysis and machine learning algorithms to establish a load forecasting model based on historical load data and environmental data are as follows:

[0080] Data collection and compilation:

[0081] Collect historical load data, including data on electrical load, cooling load and heating load, as well as corresponding time information;

[0082] Collect relevant environmental data, such as solar radiation intensity, wind speed, dry bulb temperature, humidity, etc., and record the time information of these environmental data;

[0083] Clean and preprocess the data, remove abnormal data and missing values, and standardize or normalize the data;

[0084] Feature Engineering:

[0085] Analyze the characteristics of historical load data and environmental data, and select important characteristics related to load;

[0086] It is possible to consider extracting features from time information, such as hours, days, weeks, months, seasons, etc., to capture the periodic changes in load;

[0087] Perform appropriate transformation and combination of environmental data, such as calculating average temperature, temperature difference, etc., to enhance the expressiveness of features;

[0088] Choosing the right machine learning algorithm:

[0089] According to the characteristics of the problem and the nature of the data, choose the appropriate machine learning algorithm, such as linear regression, decision tree, random forest, support vector machine, neural network, etc.;

[0090] You can consider using multiple algorithms for comparison and testing, and select the algorithm with the best performance;

[0091] Train the model:

[0092] Divide the sorted historical data into training set and test set;

[0093] Use the training set data to train the selected machine learning algorithm and adjust the algorithm's parameters to optimize the model's performance;

[0094] Techniques such as cross-validation can be used to evaluate the stability and generalization ability of the model;

[0095] Model evaluation and optimization:

[0096] Use the test set data to evaluate the trained model and calculate evaluation indicators such as root mean square error (RMSE), mean absolute error (MAE), accuracy, etc.

[0097] According to the evaluation results, analyze the deficiencies of the model and optimize the model, such as adjusting features, selecting a more appropriate algorithm, increasing the amount of data, etc.;

[0098] Establishing load forecasting model:

[0099] Based on the optimized model, a load forecasting model is established. This model can predict the changing trend of the data center's electrical load, cooling load, and heating load within a set period of time in the future according to the input environmental data and time information.

[0100] Model deployment and application:

[0101] Deploy the established load forecasting model to the actual system to achieve real-time load forecasting;

[0102] Based on the load forecast results, decision support is provided for the energy management and system operation of the data center, such as optimizing equipment operation strategies, adjusting energy supply plans, etc.

[0103] A data center system management method comprises the following steps:

[0104] Data collection:

[0105] Energy data is collected in real time through the energy data collection module, including installing power sensors at the output end of wind power system, photovoltaic system, internal combustion engine, and energy storage equipment to collect power generation and power consumption data; installing cooling sensors at the output end of absorption chiller, ground source heat pump, and ice storage system to collect cooling capacity data; installing heat sensors at the output end of waste heat recovery system, gas boiler, and ground source heat pump to collect heating data; installing solar radiation intensity sensors, wind speed sensors, and dry bulb temperature sensors around the data center to collect solar radiation intensity, wind speed, and dry bulb temperature;

[0106] At the same time, sensors are installed on each device through the equipment operation data module to collect the equipment's operating status, partial load rate, efficiency and other operating parameters, and record the start and stop time and operating time of each device;

[0107] Data organization and storage:

[0108] The data sorting module receives energy data and equipment operation data from the data collection end, cleans, filters and integrates them, and removes abnormal and duplicate data;

[0109] Store the sorted data in the database for subsequent data analysis and processing;

[0110] Data analysis and decision making:

[0111] The data analysis module analyzes the performance and efficiency of the equipment and evaluates the operating status of the equipment based on the equipment operation data;

[0112] Based on energy data, analyze energy consumption and calculate indicators such as life cycle cost, energy consumption and carbon emissions;

[0113] Combined with the operation strategy, the improved multi-objective locust optimization algorithm is used to solve the system capacity configuration and optimize the system in terms of energy, environment and economy;

[0114] Using data analysis and machine learning algorithms, a load forecasting model is established based on historical load data and environmental data to predict the changing trends of the data center's electrical load, cooling load, and heating load within a set period of time in the future;

[0115] Results presentation and decision support:

[0116] The result display module displays the processing results of the data analysis module to the user in a visual way, including equipment performance reports, energy consumption trend charts, optimization plan suggestions, etc.

[0117] Users make decisions based on the displayed results, such as adjusting equipment operation strategies, optimizing system capacity configuration, and formulating energy procurement plans, to achieve efficient operation and sustainable development of the data center.

[0118] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A data center system, characterized in that: include: Infrastructure: The infrastructure includes an energy supply system and an energy transmission system; the energy supply system includes an environmentally friendly energy module, a conventional energy module, energy conversion equipment and energy storage equipment; Data collection end: The data collection end includes an energy data collection module and an equipment operation data module arranged corresponding to the energy supply system and the energy transmission system, and obtains energy data and equipment operation data; Processing center: The processing center includes a data sorting module, a data analysis module and a result display module; The data sorting module receives energy data and equipment operation data from the data collection end to sort and store the data; The data analysis module establishes a communication connection with the data sorting module, analyzes the performance and efficiency of the equipment according to the equipment operation data, and evaluates the operation status of the equipment; analyzes the energy consumption according to the energy data, and calculates indicators such as life cycle cost, energy consumption and carbon emissions; combines the operation strategy and uses the improved multi-objective locust optimization algorithm to solve the system capacity configuration; predicts the electric load, cooling load and heating load according to the historical load data and environmental data; The result display module displays the processing results of the data analysis module to the user in a visual manner.

2. The data center system and the data center system management method according to claim 1, characterized in that: The environmentally friendly energy module includes a wind power system and a photovoltaic system. The wind power system includes a wind turbine installed at a suitable location around the data center, and the photovoltaic system includes a photovoltaic panel array installed on the top of the data center building or in the surrounding open space; the conventional energy module includes an internal combustion engine and a gas boiler; the energy conversion equipment includes waste heat recovery equipment, ground source heat pumps and absorption refrigerators arranged on the internal combustion engine and the gas boiler; the energy storage equipment includes batteries, energy storage equipment and ice storage tanks.

3. The data center system and the data center system management method according to claim 1, characterized in that: The energy transmission system includes circuit transmission, cold transmission and heat transmission. Circuit transmission transmits the electricity generated by wind power systems, photovoltaic systems and internal combustion engines to the power equipment in the data center through transmission lines, and is connected to the power grid at the same time; cold transmission transmits the cold generated by absorption refrigerators, ground source heat pumps and ice storage systems to the data room through pipelines; heat transmission transmits the heat generated by waste heat recovery systems, gas boilers and ground source heat pumps to the office and living areas through pipelines.

4. The data center system and the data center system management method according to claim 1, characterized in that: The energy data acquisition module installs power sensors at the output ends of wind power systems, photovoltaic systems, internal combustion engines, and energy storage devices to collect power generation and power consumption data in real time; installs cooling sensors at the output ends of absorption refrigerators, ground-source heat pumps, and ice storage systems to collect cooling capacity data; installs heat sensors at the output ends of waste heat recovery systems, gas boilers, and ground-source heat pumps to collect heating data; installs solar radiation intensity sensors, wind speed sensors, and dry-bulb temperature sensors around the data center to collect solar radiation intensity, wind speed, and dry-bulb temperature.

5. The data center system and the data center system management method according to claim 1, characterized in that: The equipment operation data module installs sensors on each device to collect the equipment's operating status, partial load rate, efficiency, and record the start and stop time and operating time of each device.

6. The data center system and the data center system management method according to claim 1, characterized in that: The data sorting module receives energy data and equipment operation data from the data acquisition terminal, cleans, screens and integrates the received energy data and equipment operation data, removes abnormal data and duplicate data, and stores the sorted data in a database.

7. The data center system and the data center system management method according to claim 1, characterized in that: The data analysis module analyzes the performance and efficiency of the equipment and evaluates the operating status of the equipment based on the equipment operation data; analyzes the energy consumption based on the energy data and calculates indicators such as life cycle cost, energy consumption and carbon emissions; combines the operation strategy and uses the improved multi-objective locust optimization algorithm to solve the system capacity configuration and optimize the system in terms of energy, environment and economy; uses data analysis and machine learning algorithms to establish a load forecasting model based on historical load data and environmental data to predict the changing trends of the data center's electrical load, cooling load and heating load within a set time in the future.

8. A data center system management method, characterized in that: The invention comprises a management method using any one of the data center systems described in claims 1-7.

Images