Data Center Cooling Control Method Based on Solar-Powered ORC-VCR

Abstract

This invention discloses a data center cooling control method based on solar-powered ORC-VCR, comprising: defining the objective function and the decision variables corresponding to each objective function; obtaining the Pareto solution set of the cooling system after iteration based on an improved NSGA-II algorithm; then coupling the Pareto solution set with different decision methods to determine the Pareto optimal solution, thereby obtaining different Pareto optimal solution results for the cooling system; the decision methods include the entropy weight method, the superior and inferior solution ranking method based on relative entropy distance, and the preference multidimensional analysis linear programming technique; using Taylor diagrams to measure the root mean square error, correlation coefficient, and standard deviation of different Pareto optimal solutions of the cooling system, and identifying the unique Pareto optimal solution of the cooling system; when the cooling system deviates from the theoretical optimal operating condition, adjusting the equipment parameters of the cooling system to keep the cooling system in the optimal operating state. This invention can keep the refrigeration system in the optimal operating state, reduce the energy consumption and carbon emissions of the air conditioning system, and improve the system cooling efficiency.

Description

Technical Field

[0001] This invention relates to the field of cooling and heat dissipation technology, specifically to a data center cooling control method based on solar-powered ORC-VCR. Background Technology

[0002] Data centers are a crucial component of modern information technology infrastructure, supporting the operations of numerous industries, including the internet, finance, healthcare, and education. With the rapid development of new technologies such as cloud computing and big data, the scale and complexity of data centers are constantly increasing, placing ever higher demands on energy and the environment. Among these, data center cooling is particularly critical. The operation of data center servers and equipment generates a significant amount of heat; if this heat is not dissipated promptly, it can create localized hotspots, leading to decreased server performance, shortened equipment lifespan, and ultimately impacting the stability and security of the data center.

[0003] The current mainstream cooling method is to use precision air conditioning for room-level cooling to control the temperature of the data center within a suitable range.

[0004] However, existing data center cooling control systems have the following problems: First, the air conditioning system uses room-level cooling. Due to the low air density and thermal conductivity, in order to eliminate local hot spots and avoid server failures, more than 80% of the space inside the room is over-cooled. The cooling system has high energy consumption, increases carbon emissions, and also has an adverse impact on the environment. Second, room-level cooling lacks real-time adjustment capabilities. The load and usage of servers inside the data center are constantly changing, and traditional air conditioning systems cannot adjust according to real-time conditions. Energy saving still needs to be improved and enhanced. Third, traditional data center air conditioning systems suffer from problems such as inaccurate regulation, high energy consumption, and large carbon emissions. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes a data center cooling control method based on solar-powered ORC-VCR. While ensuring the normal operation of the data center, this method can precisely adjust the cooling equipment according to the server load and usage, keeping the cooling system in optimal operating condition, reducing energy consumption and carbon emissions of the air conditioning system, and improving system cooling efficiency.

[0006] To achieve the above objectives, the present invention provides a data center cooling control method based on solar-powered ORC-VCR, which is characterized by including the following steps: S1) Before operating the data center cooling system based on solar ORC-VCR, the objective function and the corresponding decision variables are defined. The objective function includes the energy efficiency ratio function and the carbon emission reduction function of the data center cooling system based on solar ORC-VCR. The decision variables include the flow rate, temperature, and pressure values ​​at each monitoring point. The mathematical formula of the objective function and the upper and lower limits of each decision variable are set. Based on the improved NSGA-II algorithm, the Pareto solution set of the data center cooling system based on solar ORC-VCR is obtained after iteration. Then, the Pareto solution set is coupled with different decision methods to determine the Pareto optimal solution and obtain different Pareto optimal solution results. The decision methods include the entropy weight method, the superior and inferior solution ranking method based on relative entropy distance, and the preference multidimensional analysis linear programming technique. S2) Taylor diagrams are used to measure the root mean square error, correlation coefficient and standard deviation of different Pareto optimal solutions, to identify the unique Pareto optimal solution, and thus determine the theoretical optimal operating conditions of the data center cooling system based on solar ORC-VCR. S3) During the operation of the data center cooling system based on solar ORC-VCR, when it deviates from the theoretical optimal operating conditions, the equipment parameters are adjusted based on the flow rate, temperature and pressure values ​​fed back from each monitoring point to keep the data center cooling system based on solar ORC-VCR in the optimal operating state. Further, in S1), the mathematical formula for the energy efficiency ratio function of the data center cooling system based on solar ORC-VCR is expressed as follows: In the formula, COP This indicates the energy efficiency ratio of a data center cooling system based on solar-powered ORC-VCR. P net This represents the net power generation of the ORC system. Q evaporator13 Indicates the cooling capacity of the VCR evaporator; The mathematical formula for the carbon emission reduction function of the data center cooling system based on solar ORC-VCR is expressed as follows: In the formula, C eq This indicates the carbon emission reduction of a data center cooling system based on solar-powered ORC-VCR. P net This represents the net power generation of the ORC system. 0.95 represents the amount of CO2 emissions reduced per unit of net electricity generated.

[0007] Furthermore, in S1), the calculation steps of the Pareto solution set coupled entropy weight method are as follows: a1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, P ij This represents the normalized cooling scheme matrix. p ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n, j represents the number of objective functions, j=1…m; a2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. P ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; a3) Obtain the solution matrix sorted in descending order, and search for the first individual in the sorted solution matrix as the Pareto optimal solution. The matrix representation of the descending order of values ​​is as follows: In the formula, W i This represents the matrix of schemes arranged in descending order. P ij This represents the normalized cooling scheme matrix. w j This represents the weight matrix of the objective function.

[0008] Furthermore, the calculation steps of the Pareto solution set coupling method based on relative entropy distance for ranking superior and inferior solutions are as follows: b1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, Q ij This represents the normalized cooling scheme matrix. q ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n, j represents the number of objective functions, j=1…m; b2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. Q ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; b3) Obtain the matrix of schemes arranged in descending order and calculate the relative entropy distance. d i + and d i - Through the maximum coefficient value Search for the Pareto optimal solution; The matrix representation of the descending order of values ​​is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. Q ij This represents the normalized cooling scheme matrix. w j The weight matrix represents the objective function; The formula for relative entropy distance is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. This indicates that for extremely large evaluation indicators, This indicates that for extremely small evaluation indicators, This indicates that for extremely large evaluation indicators, , indicating that the evaluation index is measured for extremely small sizes.

[0009] Furthermore, in S1), the computational steps of the Pareto unset coupling preference multidimensional analysis linear programming technique are as follows: c1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, Q ij This represents the normalized cooling scheme matrix. q ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n, j represents the number of objective functions, j=1…m; c2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. Q ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; c3) Obtain the scheme matrix sorted in descending order, and find the scheme with the minimum relative entropy distance.d i - Search for the Pareto optimal solution; The matrix representation of the descending order of values ​​is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. Q ij This represents the normalized cooling scheme matrix. w j The weight matrix represents the objective function; Minimum relative entropy distance d i - The formula is as follows In the formula, R i This represents the matrix of schemes arranged in descending order. This indicates that for extremely large evaluation indicators, , indicating that the evaluation index is measured for extremely small sizes.

[0010] Furthermore, in S2), the formula for calculating the root mean square error of the Pareto optimal solution is as follows: The formula for calculating the correlation coefficient of the Pareto optimal solution is as follows: The formula for calculating the standard deviation of the Pareto optimal solution is as follows: In the formula, R rmsd This represents the center root mean square error of the Pareto optimal solution. C coef The correlation coefficient represents the optimal solution of Pareto. S stdf This represents the standard deviation of the Pareto optimal solution. f Represents a normalized matrix. This represents the average value of the normalized matrix. r represents the ideal point matrix. This represents the average value of the ideal point matrix. i represents the number of decision-making methods being evaluated, i=1…n.

[0011] This invention also designs a data center cooling system based on solar ORC-VCR, wherein the system executes the control method described above, and its special feature is that the system includes a solar thermal collection system, an organic Rankine cycle power system, a vapor compression refrigeration cycle system, a gravity backplate heat pipe system, and a cooling system; The solar thermal system includes a first shut-off valve for connecting to the water inlet. The output end of the first shut-off valve is connected to the input end of an adjustable corner trough solar collector via a pipeline. The output end of the solar collector is connected to the input ends of a second and fourth shut-off valves via pipelines. The output end of the second shut-off valve is connected to an exhaust port / drain port via a pipeline. The output end of the fourth shut-off valve is connected to the input end of a fifth shut-off valve and a flow meter via pipelines. The output end of the fifth shut-off valve is connected to the input end of a hot water storage tank via a pipeline. The output end of the hot water storage tank is connected to the input end of a sixth shut-off valve via a pipeline. The output end of the sixth shut-off valve is connected to the input end of a third shut-off valve via a pipeline. The output end of the third shut-off valve is connected to the input end of the solar collector via a pipeline. The flow meter output pipe is connected to the heat source input of the ORC evaporator. The ORC evaporator is used to transfer heat. The ORC evaporator heat source output pipe is connected to the hot water pump input. The hot water pump output pipe is connected to the cold source input of the regenerator. The regenerator cold source output pipe is connected to the input of the third shut-off valve. The solar thermal collection system also includes a first temperature sensor and a first pressure sensor installed at the heat source input end of the ORC evaporator. The first temperature sensor is used to monitor the temperature of the hot water entering the ORC evaporator, and the first pressure sensor is used to monitor the pressure of the hot water entering the ORC evaporator. The organic Rankine cycle power system includes an ORC evaporator refrigerant input terminal connected by a pipeline, an ORC evaporator refrigerant output terminal connected by a pipeline to a turbine input terminal for expansion work, a turbine drive linkage to operate the vapor compression refrigeration cycle system, a turbine output terminal connected by a pipeline to a condenser first input terminal, a condenser first output terminal connected by a pipeline to an ORC water pump input terminal, and an ORC water pump output terminal connected by a pipeline to the ORC evaporator refrigerant input terminal. The organic Rankine cycle power system also includes a second pressure sensor located at the refrigerant input end of the ORC evaporator and a second temperature sensor located at the turbine input end. The second pressure sensor is used to monitor the pressure of the liquid refrigerant entering the ORC evaporator, and the second temperature sensor is used to monitor the temperature of the gaseous refrigerant entering the turbine. The vapor compression refrigeration cycle system includes a compressor driven by a connecting rod to compress refrigerant. The refrigerant output pipe of the compressor is connected to the second input pipe of the condenser. The second output pipe of the condenser is connected to the input pipe of an electronic expansion valve. The electronic expansion valve is used for throttling and pressure reduction. The output pipe of the electronic expansion valve is connected to the heating input pipe of a VCR evaporator. The VCR evaporator is used for evaporation and heat absorption. The heating output pipe of the VCR evaporator is connected to the refrigerant input pipe of the compressor. The vapor compression refrigeration cycle system also includes a third temperature sensor installed at the refrigerant input end of the compressor and a third pressure sensor installed at the refrigerant output end of the compressor. The third temperature sensor is used to monitor the temperature of the refrigerant entering the compressor, and the third pressure sensor is used to monitor the pressure of the refrigerant flowing out of the compressor. The gravity backplate heat pipe system includes a data center server. The server has corner louvers on its outer side and a cabinet on its inner side. The cabinet has a backplate heat pipe on its outer wall and a fan on its outer wall. The bottom inlet of the backplate heat pipe is connected to a circulating liquid pipe. Liquid refrigerant flows out of the circulating liquid pipe and into the backplate heat pipe. Air enters from the outer side of the server driven by the fan and is directionally ventilated by the corner louvers, carrying away the heat dissipated from the server. The dissipated heat exchanges heat with the liquid refrigerant flowing into the backplate heat pipe, causing the liquid refrigerant to evaporate into gaseous refrigerant. The gaseous refrigerant flows into the circulating gas pipe through the top outlet of the backplate heat pipe. After exchanging heat with the cold source in the vapor compression refrigeration cycle system, the gaseous refrigerant in the circulating gas pipe condenses into liquid refrigerant. The liquid refrigerant flows into the circulating liquid pipe under the effect of the height difference and continues to absorb heat, realizing a heat absorption-heat release cycle. The gravity backplate heat pipe system also includes a fourth temperature sensor located at the bottom input end of the backplate heat pipe and a fifth temperature sensor located at the top output end of the backplate heat pipe. The fourth temperature sensor is used to monitor the temperature of the refrigerant entering the backplate heat pipe, and the fifth temperature sensor is used to monitor the temperature of the refrigerant flowing out of the backplate heat pipe. The cooling system includes a seventh shut-off valve for connecting to the cooling water inlet. The output pipe of the seventh shut-off valve is connected to the input of the cooling water tank. The cooling water in the cooling water tank exchanges heat with the condenser. The output pipe of the cooling water tank is connected to the input of the cooling water pump. The output pipe of the cooling water pump is connected to the heat source input of the regenerator. The heat source output of the regenerator serves as the cooling water outlet.

[0012] Furthermore, when performing step S1), the decision variables include the rotation angle of the adjustable angle trough solar collector, the flow rate of the flow meter, the temperature value of the first temperature sensor, the pressure value of the first pressure sensor, the temperature value of the second temperature sensor, the pressure value of the second pressure sensor, the temperature value of the third temperature sensor, the pressure value of the third pressure sensor, the temperature value of the fourth temperature sensor, and the temperature value of the fifth temperature sensor.

[0013] Furthermore, when performing step S3), the equipment parameters that need to be adjusted include the speed of the hot water pump, the speed of the ORC water pump, the speed of the fan, and the speed of the cooling water pump.

[0014] The advantages of this invention are: 1. The traditional NSGA-II algorithm has shortcomings. After obtaining the Pareto solution set of the system at the end of the iteration, it needs to further combine decision methods to determine the Pareto optimal solution. However, different decision methods often determine the optimal solution inconsistently and with large differences, which complicates the optimization process and easily leads to incorrect optimal solutions. The improved NSGA-II algorithm in this invention couples different decision methods to determine the optimal solution after the iteration, obtaining different optimal solution results. Then, a Taylor diagram is used to measure the root mean square error, correlation coefficient, and standard deviation of different optimal solution results, thereby identifying the unique Pareto optimal solution. 2. In decision-making methods, traditional Top-Order Solution Ranking (TOPSIS) and traditional linear programming techniques (LINMAP) determine Pareto optimal solutions by searching for the shortest geometric distance between the Pareto optimization boundary and the ideal point, and the longest geometric distance between the boundary and the non-ideal point. However, the distances between the Pareto optimization boundary and each ideal point and each non-ideal point may be equal, making it impossible to select a Pareto optimal solution. The Top-Order Solution Ranking (TOPSIS) and Preference Multidimensional Analysis Linear Programming Technique (LINMAP) in this invention introduce relative entropy (REntropy) to estimate the difference in probability distributions, replacing the traditional Euclidean distance. The Top-Order Solution Ranking (TOPSIS) based on relative entropy distance... and The relative entropy distance is determined by the maximum coefficient value. Searching for Pareto optimal solutions; preference for multidimensional analysis linear programming techniques (LINMAP) with the minimum relative entropy distance. Search for the Pareto optimal solution; 3. The cooling system of this invention features vertically arranged backplate heat pipes on the side of the server rack. The heat-absorbing end of the backplate heat pipes is in close contact with the server rack. The heat generated by the server is concentrated around the backplate heat pipes by the fan. The liquid refrigerant entering the backplate heat pipes absorbs the heat from the server and evaporates into gaseous refrigerant. The gaseous refrigerant exchanges heat with the cold source in the VCR system and condenses into liquid, promoting the vapor compression refrigeration cycle of the VCR system. Simultaneously, due to the height difference, the condensed liquid flows back to the server rack to continue absorbing heat. This cooling method has the following advantages: First, it eliminates the need for a water pump, ensuring high reliability and avoiding the risk of coolant leakage compared to indirect liquid cooling. Second, it meets the daytime and nighttime cooling needs of the data center room. Third, by using the rack backplate heat pipe air intake method, it directly reduces the internal high temperature of the server, precisely avoiding localized hot spots and server thermal failures, and preventing overcooling inside the data center room. Fourth, it allows for adjustments to the solar thermal collection system and ORC-VCR system based on the server load and usage, achieving high efficiency and energy saving. 4. The cooling system in this invention utilizes solar energy to achieve the conversion process of light-heat-electricity-cooling, saving energy while adapting to the needs of different times and environments. It can automatically switch the corresponding working mode and precisely adjust the cooling equipment according to the server load and usage, keeping the cooling system in the best operating state, reducing the energy consumption and carbon emissions of the air conditioning system, and improving the system's cooling efficiency. This invention provides a data center cooling control method based on solar-powered ORC-VCR, which can directly reduce the internal high temperature of servers and avoid excessive cooling inside the data center room. At the same time, it can adjust the solar thermal system and ORC-VCR system according to the server load and usage, so that the cooling system is kept in the optimal operating state, reducing the energy consumption and carbon emissions of the air conditioning system and improving the system cooling efficiency. Attached Figure Description

[0015] Figure 1 This is a flowchart of the improved NSGA-II algorithm in this invention; Figure 2 The Pareto optimal solution is determined using traditional Euclidean distance; Figure 3 Select the Pareto optimal solution distribution for different decision-making methods; Figure 4 Taylor diagrams are used to measure decision points; Figure 5 This is a block diagram of the overall structure of the data center cooling system based on solar-powered ORC-VCR in this invention. In the diagram: 1. Solar collector; 2. Regenerator; 3. ORC evaporator; 4. Flow meter; 5. Hot water pump; 6. Hot water storage tank; 7. Turbine; 8. ORC water pump; 9. Condenser; 10. Compressor; 11. Connecting rod; 12. Electronic expansion valve; 13. VCR evaporator; 14. Backplate heat pipe; 15. Server; 16. Cabinet; 17. Corner louvers; 18. Fan; 19. Cooling water tank; 20. Cooling water pump; 21. First shut-off valve; 22. Second shut-off valve; 23. Third shut-off valve; 24. Fourth shut-off valve; 25. Fifth shut-off valve; 26. Sixth shut-off valve; 27. Seventh shut-off valve; 28. First temperature sensor; 29. ​​First pressure sensor; 30. Second temperature sensor; 31. Third temperature sensor; 32. Second pressure sensor; 33. Third pressure sensor; 34. Fourth temperature sensor; 35. Fifth temperature sensor. Detailed Implementation

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] In the description of this invention, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention.

[0018] This invention relates to a data center cooling control method based on solar-powered ORC-VCR, which is implemented through a data center cooling system based on solar-powered ORC-VCR.

[0019] The data center cooling system based on solar-powered ORC-VCR includes a solar thermal system, an organic Rankine cycle power system, a vapor compression refrigeration cycle system, a gravity backplate heat pipe system, and a cooling system. The organic Rankine cycle (ORC) uses a low-boiling-point organic compound as the working fluid, while the vapor compression refrigeration cycle (VCR), also known as a mechanical compression refrigeration cycle, is simply a compression cycle. Figure 5 The diagram shown is an overall structural block diagram of the data center cooling system based on solar-powered ORC-VCR in this invention.

[0020] The solar thermal system includes a first shut-off valve 21 for connecting to the water inlet. The output of the first shut-off valve 21 is connected to the input of an adjustable angle trough solar collector 1 via a pipeline. The output of the solar collector 1 is connected to the input of a second shut-off valve 22 and a fourth shut-off valve 24 via pipelines. The output of the second shut-off valve 22 is connected to an exhaust port / drain port via a pipeline. The output of the fourth shut-off valve 24 is connected to the input of a fifth shut-off valve 25 and a flow meter 4 via pipelines. The output of the fifth shut-off valve 25 is connected to the input of a hot water storage tank 6 via a pipeline. The output of the hot water storage tank 6 is connected to the input of a sixth shut-off valve 26 via a pipeline. The output of the sixth shut-off valve 26 is connected to the input of a third shut-off valve 23 via a pipeline. The output of the third shut-off valve 23 is connected to the input of the solar collector 1 via a pipeline.

[0021] The output pipe of the flow meter 4 is connected to the heat source input of the ORC evaporator 3. The ORC evaporator 3 is used to transfer heat. The output pipe of the ORC evaporator 3 is connected to the input of the hot water pump 5. The output pipe of the hot water pump 5 is connected to the cold source input of the regenerator 2. The output pipe of the cold source of the regenerator 2 is connected to the input of the third shut-off valve 23.

[0022] The solar thermal collector system also includes a first temperature sensor 28 and a first pressure sensor 29 installed at the heat source input end of the ORC evaporator 3. The first temperature sensor 28 is used to monitor the temperature of the hot water entering the ORC evaporator 3, and the first pressure sensor 29 is used to monitor the pressure of the hot water entering the ORC evaporator 3.

[0023] The organic Rankine cycle power system includes a refrigerant input terminal of an ORC evaporator 3 connected by a pipeline. The refrigerant output terminal of the ORC evaporator 3 is connected by a pipeline to the input terminal of a turbine 7 used for expansion work. The turbine 7 drives a connecting rod 11 to operate the vapor compression refrigeration cycle system. The output terminal of the turbine 7 is connected by a pipeline to the first input terminal of a condenser 9. The first output terminal of the condenser 9 is connected by a pipeline to the input terminal of an ORC water pump 8. The output terminal of the ORC water pump 8 is connected by a pipeline to the refrigerant input terminal of the ORC evaporator 3.

[0024] The organic Rankine cycle power system also includes a second pressure sensor 32 located at the refrigerant input end of the ORC evaporator 3 and a second temperature sensor 30 located at the input end of the turbine 7. The second pressure sensor 32 is used to monitor the pressure of the liquid refrigerant entering the ORC evaporator 3, and the second temperature sensor 30 is used to monitor the temperature of the gaseous refrigerant entering the turbine 7.

[0025] The vapor compression refrigeration cycle system includes a compressor 10 driven by a connecting rod 11 to compress refrigerant. The refrigerant output pipe of the compressor 10 is connected to the second input pipe of the condenser 9. The second output pipe of the condenser 9 is connected to the input pipe of the electronic expansion valve 12. The electronic expansion valve 12 is used for throttling and pressure reduction. The output pipe of the electronic expansion valve 12 is connected to the heating input pipe of the VCR evaporator 13. The VCR evaporator 13 is used for evaporation and heat absorption. The heating output pipe of the VCR evaporator 13 is connected to the refrigerant input pipe of the compressor 10.

[0026] The vapor compression refrigeration cycle system further includes a third temperature sensor 31 disposed at the refrigerant input end of the compressor 10 and a third pressure sensor 33 disposed at the refrigerant output end of the compressor 10. The third temperature sensor 31 is used to monitor the temperature of the refrigerant entering the compressor 10, and the third pressure sensor 33 is used to monitor the pressure of the refrigerant flowing out of the compressor 10.

[0027] The gravity backplate heat pipe system includes a server 15 in a data center. The outer side of the server 15 is equipped with corner louvers 17, and the inner side of the server 15 is equipped with a cabinet 16. A backplate heat pipe 14 is installed on the outer wall of the cabinet 16, and a fan 18 is installed on the outer wall of the backplate heat pipe 14. The bottom input end of the backplate heat pipe 14 is connected to a circulating liquid pipe. Liquid refrigerant flows out of the circulating liquid pipe and enters the backplate heat pipe 14. Air, driven by the fan 18, enters from the outer side of the server 15 and is directionally ventilated by the corner louvers 17, carrying away the heat emitted from the server 15. The emitted heat exchanges heat with the liquid refrigerant flowing into the backplate heat pipe 14, causing the liquid refrigerant to evaporate into gaseous refrigerant. The gaseous refrigerant flows into a circulating gas pipe through the top output end of the backplate heat pipe 14. After exchanging heat with the cold source in the vapor compression refrigeration cycle system, the gaseous refrigerant in the circulating gas pipe condenses into liquid refrigerant. The liquid refrigerant flows into the circulating liquid pipe under the influence of the height difference and continues to absorb heat, realizing a heat absorption-heat release cycle.

[0028] The gravity backplate heat pipe system also includes a fourth temperature sensor 34 located at the bottom input end of the backplate heat pipe 14 and a fifth temperature sensor 35 located at the top output end of the backplate heat pipe 14. The fourth temperature sensor 34 is used to monitor the temperature of the refrigerant entering the backplate heat pipe 14, and the fifth temperature sensor 35 is used to monitor the temperature of the refrigerant flowing out of the backplate heat pipe 14.

[0029] The difference between the bottom height of the VCR evaporator 13 and the top height of the server 15 is greater than 1m.

[0030] The cooling system includes a seventh shut-off valve 27 for connecting to the cooling water inlet. The output pipe of the seventh shut-off valve 27 is connected to the input of the cooling water tank 19. The cooling water in the cooling water tank 19 exchanges heat with the condenser 9. The output pipe of the cooling water tank 19 is connected to the input of the cooling water pump 20. The output pipe of the cooling water pump 20 is connected to the heat source input of the regenerator 2. The heat source output of the regenerator 2 serves as the cooling water outlet.

[0031] The cooling system removes refrigerant heat from the ORC-VCR system in condenser 9 and preheats the solar-heated water in regenerator 2. This system utilizes solar energy to provide cooling capacity around the clock, significantly reducing data center air conditioning energy consumption.

[0032] The data center cooling system based on solar-powered ORC-VCR can switch between daytime and nighttime operating modes, indirectly converting low-temperature hot water generated by solar energy into cold energy, realizing the conversion process of light-heat-electricity-cooling. It also utilizes the heat generated by the servers to absorb heat and evaporate the cold source in the VCR system, promoting the vapor compression refrigeration cycle of the VCR system. This not only meets the cooling needs of the data center room during the day and night, but also directly reduces the internal high temperature of the servers, avoiding overcooling of the data center room. At the same time, the solar thermal collector system and ORC-VCR system can be adjusted according to the server load and usage to achieve high efficiency and energy saving.

[0033] This invention relates to a data center cooling control method based on solar-powered ORC-VCR, comprising the following steps: S1) Before operating the data center cooling system based on solar ORC-VCR, the objective function and the corresponding decision variables are defined. The objective function includes the energy efficiency ratio function and the carbon emission reduction function of the data center cooling system based on solar ORC-VCR. The decision variables include the flow rate, temperature, and pressure values ​​of each monitoring point. The mathematical formula of the objective function and the upper and lower limits of each decision variable are set. Based on the improved NSGA-II algorithm, the Pareto solution set of the data center cooling system based on solar ORC-VCR is obtained after iteration. Then, the Pareto solution set is coupled with different decision methods to determine the Pareto optimal solution and obtain different Pareto optimal solution results. The decision methods include the entropy weight method, the superior and inferior solution ranking method based on relative entropy distance, and the preference multidimensional analysis linear programming technique.

[0034] Specifically, in S1), the mathematical formula for the energy efficiency ratio function of the data center cooling system based on solar ORC-VCR is expressed as follows: In the formula, COP This indicates the energy efficiency ratio of a data center cooling system based on solar-powered ORC-VCR. P net This represents the net power generation of the ORC system. Q evaporator13 Indicates the cooling capacity of the VCR evaporator; The mathematical formula for the carbon emission reduction function of the data center cooling system based on solar ORC-VCR is expressed as follows: In the formula, C eq This indicates the carbon emission reduction of a data center cooling system based on solar-powered ORC-VCR. P net This represents the net power generation of the ORC system. 0.95 represents the amount of CO2 emissions reduced per unit of net electricity generated.

[0035] In this embodiment, the decision variables include the rotation angle δ of the adjustable angle trough solar collector 1, the flow rate Q of the flow meter 4, the temperature value T1 of the first temperature sensor 28, the pressure value P1 of the first pressure sensor 29, the temperature value T2 of the second temperature sensor 30, the pressure value P2 of the second pressure sensor 32, the temperature value T3 of the third temperature sensor 31, the pressure value P3 of the third pressure sensor 33, the temperature value T4 of the fourth temperature sensor 34, and the temperature value T5 of the fifth temperature sensor 35.

[0036] S2) Taylor diagrams are used to measure the root mean square error, correlation coefficient and standard deviation of different Pareto optimal solutions, to identify the unique Pareto optimal solution, and thus determine the theoretical optimal operating conditions of the data center cooling system based on solar ORC-VCR.

[0037] The traditional NSGA-II algorithm has shortcomings. After obtaining the Pareto solution set of the system at the end of the iteration, it needs to further combine decision methods to determine the Pareto optimal solution. However, different decision methods often determine the optimal solution inconsistently and with large differences, which complicates the optimization process and makes it easy to obtain the wrong optimal solution.

[0038] This invention employs an improved NSGA-II algorithm to construct a system operation model and clarify the optimal combination of decision variables when the system achieves the target performance. Specifically, it consists of two steps: First, before system operation, a target function representing performance and its related decision variables are selected. Then, the mathematical formula for the target function is set, and the upper and lower limits of each decision variable are input. The population size is set to Pop, and the number of evolutions is set to Gen. The improvement of the NSGA-II algorithm lies in coupling different decision methods after iteration to determine the optimal solution, obtaining different optimal solution results. Second, a Taylor diagram is used to measure the root mean square error, correlation coefficient, and standard deviation of different optimal solution results, thereby clarifying the unique Pareto optimal solution. Figure 1 The diagram shows a flowchart of the improved NSGA-II algorithm in this invention.

[0039] Table 1 below shows the parameter optimization settings for the improved NSGA-II algorithm in this invention.

[0040] Table 1 Parameter Optimization Settings In decision-making methods, traditional Top-First-Solve-Solve (TOPSIS) and traditional linear programming techniques (LINMAP) determine Pareto optimal solutions by searching for the shortest geometric distance between the Pareto optimization boundary and the ideal point, and the longest geometric distance between the boundary and the non-ideal point. However, the distances between the Pareto optimization boundary and each ideal point and each non-ideal point may be equal, making it impossible to select a Pareto optimal solution. Figure 2 As shown, the traditional Euclidean distance is used to determine the Pareto optimal solution. Therefore, a relative entropy is introduced to estimate the difference in probability distributions, replacing the traditional Euclidean distance. .

[0041] Among them, the Top-Ranking Solution Method based on Relative Entropy Distance (TOPSIS) and the Preference Multidimensional Analysis Linear Programming Technique (LINMAP) have improved upon the original methods by introducing relative entropy, used to estimate the differences in probability distributions, to replace the traditional Euclidean distance. Specifically, the Top-Ranking Solution Method based on Relative Entropy Distance uses... and The relative entropy distance is determined by the maximum coefficient value. Searching for Pareto optimal solutions; favoring multidimensional analysis and linear programming techniques to minimize relative entropy distance. Search for the Pareto optimal solution.

[0042] Taking the above conditions as an example, the main calculation steps of this method are explained in detail: Specifically, the calculation steps of the Pareto solution set coupled entropy weight method are as follows: a1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, P ij This represents the normalized cooling scheme matrix. p ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n. The number of cooling schemes in this system is the population size in the NSGA-II algorithm. j represents the number of objective functions, j=1…m; a2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. P ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; a3) Obtain the solution matrix sorted in descending order, and search for the first individual in the sorted solution matrix as the Pareto optimal solution. The matrix representation of the descending order of values ​​is as follows: In the formula, W i This represents the matrix of schemes arranged in descending order. P ij This represents the normalized cooling scheme matrix. w j This represents the weight matrix of the objective function.

[0043] Specifically, the calculation steps of the Pareto solution set coupling method based on relative entropy distance for ranking solutions are as follows: b1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, Q ij This represents the normalized cooling scheme matrix. q ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n, j represents the number of objective functions, j=1…m; b2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. Q ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; b3) Obtain the matrix of schemes arranged in descending order and calculate the relative entropy distance. d i + and d i - Through the maximum coefficient value Search for the Pareto optimal solution; The matrix representation of the descending order of values ​​is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. Q ij This represents the normalized cooling scheme matrix. w j The weight matrix represents the objective function; The formula for relative entropy distance is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. This indicates that for extremely large evaluation indicators, This indicates that for extremely small evaluation indicators, This indicates that for extremely large evaluation indicators, , indicating that the evaluation index is measured for extremely small sizes.

[0044] Specifically, the computational steps of the Pareto unset coupling preference multidimensional analysis linear programming technique are as follows: c1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, Q ij This represents the normalized cooling scheme matrix. q ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n, j represents the number of objective functions, j=1…m; c2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. Q ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; c3) Obtain the scheme matrix sorted in descending order, and find the scheme with the minimum relative entropy distance. d i - Search for the Pareto optimal solution; The matrix representation of the descending order of values ​​is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. Q ij This represents the normalized cooling scheme matrix. w j The weight matrix represents the objective function; Minimum relative entropy distance d i - The formula is as follows In the formula, R i This represents the matrix of schemes arranged in descending order. This indicates that for extremely large evaluation indicators, , indicating that the evaluation index is measured for extremely small sizes.

[0045] Figure 3 As shown, the Pareto optimal solution distribution is selected for different decision-making methods. From Figure 3 As can be seen, the three decision-making methods—Shannon Entropy, the improved TOPSIS, and the improved LINMPA—determine different Pareto optimal solutions, requiring further Taylor diagram analysis to determine the unique Pareto optimal solution.

[0046] Root mean square error of different Pareto optimal solutions for the cooling system ( ), correlation coefficient ( ) and standard deviation ( The numbers ) represent the differences, similarities, and magnitudes of change between the simulation results and the ideal solution in the Taylor diagram, respectively.

[0047] Specifically, the formula for calculating the root mean square error of the Pareto optimal solution is as follows: The formula for calculating the correlation coefficient of the Pareto optimal solution is as follows: The formula for calculating the standard deviation of the Pareto optimal solution is as follows: In the formula, R rmsd This represents the center root mean square error of the Pareto optimal solution. C coefThe correlation coefficient represents the optimal solution of Pareto. S stdf This represents the standard deviation of the Pareto optimal solution. f Represents a normalized matrix. This represents the average value of the normalized matrix. r represents the ideal point matrix. This represents the average value of the ideal point matrix. i represents the number of decision-making methods being evaluated, i = 1…n. In this implementation, n is taken as 3.

[0048] like Figure 4 The diagram shown is a schematic of using a Taylor diagram to measure decision points in this embodiment.

[0049] S3) During the operation of the data center cooling system based on solar ORC-VCR, when it deviates from the theoretical optimal operating conditions, the equipment parameters are adjusted according to the flow rate, temperature and pressure values ​​fed back from each monitoring point to keep the cooling system in the best operating state.

[0050] Table 2 below shows the Taylor diagram decision point results in this embodiment.

[0051] Table 2. Taylor Chart Decision Point Results like Figure 4 As shown in Table 2, when the correlation coefficients and standard deviations of each decision point are close, the working condition corresponding to the decision point with the smallest central root mean square error is determined by the Taylor diagram to be the theoretically optimal condition. Based on this condition, the equipment is adjusted to obtain the optimal operating state of the system.

[0052] In this embodiment, the equipment parameters that need to be adjusted include the speed of the hot water pump 5, the speed of the ORC water pump 8, the speed of the fan 18, and the speed of the cooling water pump 20.

[0053] This invention provides a data center cooling control method based on solar-powered ORC-VCR, which can directly reduce the internal high temperature of servers and avoid excessive cooling inside the data center room. At the same time, it can adjust the solar thermal system and ORC-VCR system according to the server load and usage, so that the cooling system is kept in the optimal operating state, reducing the energy consumption and carbon emissions of the air conditioning system and improving the system cooling efficiency.

[0054] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A data center cooling control method based on solar-powered ORC-VCR, characterized in that, Includes the following steps: S1) Before operating the data center cooling system based on solar ORC-VCR, the objective function and the corresponding decision variables are defined. The objective function includes the energy efficiency ratio function and the carbon emission reduction function of the data center cooling system based on solar ORC-VCR. The decision variables include the flow rate, temperature, and pressure values ​​at each monitoring point. The mathematical formula of the objective function and the upper and lower limits of each decision variable are set. Based on the improved NSGA-II algorithm, the Pareto solution set of the data center cooling system based on solar ORC-VCR is obtained after iteration. Then, the Pareto solution set is coupled with different decision methods to determine the Pareto optimal solution and obtain different Pareto optimal solution results. The decision methods include the entropy weight method, the superior and inferior solution ranking method based on relative entropy distance, and the preference multidimensional analysis linear programming technique. The solar-based ORC-VCR data center cooling system includes a solar thermal collection system, an organic Rankine cycle power system, a vapor compression refrigeration cycle system, a gravity backplate heat pipe system, and a cooling system. The solar thermal system includes a first shut-off valve (21) for connecting to the water inlet. The output end of the first shut-off valve (21) is connected to the input end of an adjustable corner trough solar collector (1) via a pipeline. The output end of the solar collector (1) is connected to the input end of a second shut-off valve (22) and the input end of a fourth shut-off valve (24) via pipelines. The output end of the second shut-off valve (22) is connected to an exhaust port / drain port via a pipeline. The output end of the fourth shut-off valve (24) is connected to the input end of a fifth shut-off valve (25) and the input end of a flow meter (4) via pipelines. The output end of the fifth shut-off valve (25) is connected to the input end of a hot water storage tank (6) via a pipeline. The output end of the hot water storage tank (6) is connected to the input end of a sixth shut-off valve (26) via a pipeline. The output end of the sixth shut-off valve (26) is connected to the input end of a third shut-off valve (23) via a pipeline. The output end of the third shut-off valve (23) is connected to the input end of the solar collector (1) via a pipeline. The flow meter (4) output pipe is connected to the heat source input of the ORC evaporator (3). The ORC evaporator (3) is used to transfer heat. The ORC evaporator (3) heat source output pipe is connected to the input of the hot water pump (5). The hot water pump (5) output pipe is connected to the cold source input of the regenerator (2). The regenerator (2) cold source output pipe is connected to the input of the third shut-off valve (23). The solar thermal system also includes a first temperature sensor (28) and a first pressure sensor (29) installed at the heat source input end of the ORC evaporator (3). The first temperature sensor (28) is used to monitor the temperature of the hot water entering the ORC evaporator (3), and the first pressure sensor (29) is used to monitor the pressure of the hot water entering the ORC evaporator (3). The organic Rankine cycle power system includes a refrigerant input terminal of an ORC evaporator (3) connected by a pipeline, a refrigerant output terminal of the ORC evaporator (3) connected by a pipeline to the input terminal of a turbine (7) used for expansion work, the turbine (7) driving a connecting rod (11) to make the vapor compression refrigeration cycle system work, a pipeline connecting the output terminal of the turbine (7) connected by a pipeline to the first input terminal of a condenser (9), a pipeline connecting the first output terminal of the condenser (9) connected by a pipeline to the input terminal of an ORC water pump (8), and a pipeline connecting the output terminal of the ORC water pump (8) connected by a pipeline to the refrigerant input terminal of the ORC evaporator (3). The organic Rankine cycle power system also includes a second pressure sensor (32) installed at the refrigerant input end of the ORC evaporator (3) and a second temperature sensor (30) installed at the input end of the turbine (7). The second pressure sensor (32) is used to monitor the pressure of the liquid refrigerant entering the ORC evaporator (3), and the second temperature sensor (30) is used to monitor the temperature of the gaseous refrigerant entering the turbine (7). The vapor compression refrigeration cycle system includes a compressor (10) driven by a connecting rod (11) to compress refrigerant. The refrigerant output end of the compressor (10) is connected to the second input end of a condenser (9). The second output end of the condenser (9) is connected to the input end of an electronic expansion valve (12). The electronic expansion valve (12) is used for throttling and pressure reduction. The output end of the electronic expansion valve (12) is connected to the heating input end of a VCR evaporator (13). The VCR evaporator (13) is used for evaporation and heat absorption. The heating output end of the VCR evaporator (13) is connected to the refrigerant input end of the compressor (10). The vapor compression refrigeration cycle system also includes a third temperature sensor (31) installed at the refrigerant input end of the compressor (10) and a third pressure sensor (33) installed at the refrigerant output end of the compressor (10). The third temperature sensor (31) is used to monitor the temperature of the refrigerant entering the compressor (10), and the third pressure sensor (33) is used to monitor the pressure of the refrigerant flowing out of the compressor (10). The gravity backplate heat pipe system includes a server (15) in a data center. The outer side of the server (15) is provided with corner louvers (17), and the inner side of the server (15) is provided with a cabinet (16). The outer wall of the cabinet (16) is provided with a backplate heat pipe (14), and the outer wall of the backplate heat pipe (14) is provided with a fan (18). The bottom inlet of the backplate heat pipe (14) is connected to a circulating liquid pipe. Liquid refrigerant flows out of the circulating liquid pipe and enters the backplate heat pipe (14). Air enters from the outer side of the server (15) under the drive of the fan (18). The heat emitted from the server (15) is carried away by the directional airflow through the corner louvers (17). The emitted heat exchanges heat with the liquid refrigerant flowing into the back plate heat pipe (14), causing the liquid refrigerant to evaporate into gaseous refrigerant. The gaseous refrigerant flows into the circulation pipe through the top output end of the back plate heat pipe (14). After exchanging heat with the cold source in the vapor compression refrigeration cycle system, the gaseous refrigerant in the circulation pipe condenses into liquid refrigerant. The liquid refrigerant flows into the circulation liquid pipe under the action of the height difference, and continues to absorb heat to realize the heat absorption-heat release cycle. The gravity backplate heat pipe system also includes a fourth temperature sensor (34) located at the bottom input end of the backplate heat pipe (14) and a fifth temperature sensor (35) located at the top output end of the backplate heat pipe (14). The fourth temperature sensor (34) is used to monitor the temperature of the refrigerant entering the backplate heat pipe (14), and the fifth temperature sensor (35) is used to monitor the temperature of the refrigerant flowing out of the backplate heat pipe (14). The cooling system includes a seventh shut-off valve (27) for connecting to the cooling water inlet. The output pipe of the seventh shut-off valve (27) is connected to the input of the cooling water tank (19). The cooling water in the cooling water tank (19) exchanges heat with the condenser (9). The output pipe of the cooling water tank (19) is connected to the input of the cooling water pump (20). The output pipe of the cooling water pump (20) is connected to the heat source input of the regenerator (2). The heat source output of the regenerator (2) serves as the cooling water outlet. S2) Taylor diagrams are used to measure the root mean square error, correlation coefficient and standard deviation of different Pareto optimal solutions, to identify the unique Pareto optimal solution, and thus determine the theoretical optimal operating conditions of the data center cooling system based on solar ORC-VCR. S3) During the operation of the data center cooling system based on solar ORC-VCR, when it deviates from the theoretical optimal operating conditions, the equipment parameters are adjusted based on the flow rate, temperature and pressure values ​​fed back from each monitoring point to keep the data center cooling system based on solar ORC-VCR in the optimal operating state. The equipment parameters include the rotation speed of the hot water pump (5), the rotation speed of the ORC water pump (8), the rotation speed of the fan (18), and the rotation speed of the cooling water pump (20).

2. The data center cooling control method based on solar ORC-VCR according to claim 1, characterized in that: In S1), the mathematical formula for the energy efficiency ratio function of the data center cooling system based on solar ORC-VCR is expressed as follows: In the formula, COP This indicates the energy efficiency ratio of a data center cooling system based on solar-powered ORC-VCR. P net This represents the net power generation of the ORC system. Q evaporator13 Indicates the cooling capacity of the VCR evaporator; The mathematical formula for the carbon emission reduction function of the data center cooling system based on solar ORC-VCR is expressed as follows: In the formula, C eq This indicates the carbon emission reduction of a data center cooling system based on solar-powered ORC-VCR. P net This represents the net power generation of the ORC system. 0.95 represents the amount of CO2 emissions reduced per unit of net electricity generated.

3. The data center cooling control method based on solar ORC-VCR according to claim 2, characterized in that: In S1), the calculation steps of the Pareto solution set coupled entropy weight method are as follows: a1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, P ij This represents the normalized cooling scheme matrix. p ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n, j represents the number of objective functions, j=1…m; a2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. P ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; a3) Obtain the solution matrix sorted in descending order, and search for the first individual in the sorted solution matrix as the Pareto optimal solution. The matrix representation of the descending order of values ​​is as follows: In the formula, W i This represents the matrix of schemes arranged in descending order. P ij This represents the normalized cooling scheme matrix. w j This represents the weight matrix of the objective function.

4. The data center cooling control method based on solar ORC-VCR according to claim 2, characterized in that: In S1), the calculation steps of the Pareto solution set coupling method based on relative entropy distance for ranking superior and inferior solutions are as follows: b1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, Q ij This represents the normalized cooling scheme matrix. q ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n, j represents the number of objective functions, j=1…m; b2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. Q ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; b3) Obtain the matrix of schemes arranged in descending order and calculate the relative entropy distance. d i + and d i - Through the maximum coefficient value Search for the Pareto optimal solution; The matrix representation of the descending order of values ​​is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. Q ij This represents the normalized cooling scheme matrix. w j The weight matrix represents the objective function; The formula for relative entropy distance is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. This indicates that for extremely large evaluation indicators, This indicates that for extremely small evaluation indicators, This indicates that for extremely large evaluation indicators, , indicating that the evaluation index is measured for extremely small sizes.

5. The data center cooling control method based on solar ORC-VCR according to claim 2, characterized in that: In S1), the calculation steps of the Pareto unset coupling preference multidimensional analysis linear programming technique are as follows: c1) Normalize the cooling scheme matrix of the data center cooling system based on solar ORC-VCR. The normalization formula is as follows: In the formula, Q ij This represents the normalized cooling scheme matrix. q ij This represents the calculated value of the objective function. i represents the number of cooling schemes, i=1…n, j represents the number of objective functions, j=1…m; c2) Calculate the weight matrix of the data center cooling system based on solar ORC-VCR. The formula for the weight matrix is ​​as follows: In the formula, w j The weight matrix represents the objective function. h j The information entropy index represents the objective function. The formula for the information entropy index is as follows: In the formula, h j The information entropy index represents the objective function. Q ij This represents the normalized cooling scheme matrix. i represents the number of cooling schemes, i=1…n; c3) Obtain the scheme matrix sorted in descending order, and find the scheme with the minimum relative entropy distance. d i - Search for the Pareto optimal solution; The matrix representation of the descending order of values ​​is as follows: In the formula, R i This represents the matrix of schemes arranged in descending order. Q ij This represents the normalized cooling scheme matrix. w j The weight matrix represents the objective function; Minimum relative entropy distance d i - The formula is as follows In the formula, R i This represents the matrix of schemes arranged in descending order. This indicates that for extremely large evaluation indicators, , indicating that the evaluation index is measured for extremely small sizes.

6. The data center cooling control method based on solar ORC-VCR according to claim 1, characterized in that: In S2), the formula for calculating the root mean square error of the Pareto optimal solution is as follows: The formula for calculating the correlation coefficient of the Pareto optimal solution is as follows: The formula for calculating the standard deviation of the Pareto optimal solution is as follows: In the formula, R rmsd This represents the center root mean square error of the Pareto optimal solution. C coef The correlation coefficient represents the optimal solution of Pareto. S stdf This represents the standard deviation of the Pareto optimal solution. f Represents a normalized matrix. This represents the average value of the normalized matrix. r represents the ideal point matrix. This represents the average value of the ideal point matrix. i represents the number of decision-making methods being evaluated, i=1…n.

7. The data center cooling control method based on solar ORC-VCR according to claim 1, characterized in that: When performing step S1), the decision variables include the rotation angle of the adjustable angle trough solar collector (1), the flow rate of the flow meter (4), the temperature value of the first temperature sensor (28), the pressure value of the first pressure sensor (29), the temperature value of the second temperature sensor (30), the pressure value of the second pressure sensor (32), the temperature value of the third temperature sensor (31), the pressure value of the third pressure sensor (33), the temperature value of the fourth temperature sensor (34), and the temperature value of the fifth temperature sensor (35).

Images