Data center cooling control method using hot channel and related device

Abstract

The invention discloses a data center cooling control method using a hot channel and a related device, and relates to the technical field of data processing, and the method comprises the steps: constructing a digital twin model of a data center to analyze the heat production of the data center; determining airflow organization simulation data of the data center based on the digital twin model to perform airflow organization influence analysis; performing heat dissipation energy consumption control analysis based on the airflow organization influence analysis data; a cooling strategy is constructed based on heat production, airflow organization influence analysis data and heat dissipation energy consumption control analysis data in combination with cold and hot channel load operation analysis; performing cooling analogue simulation of the data center based on the cooling strategy, and optimizing the cooling strategy based on a cooling analogue simulation result to obtain an optimized cooling strategy; and constructing a self-adaptive control system, and controlling the hot channel and the cold channel to cool the data center based on the optimized cooling strategy in combination with the self-adaptive control system. According to the invention, the cooling treatment of the data center can achieve a more ideal effect.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a data center cooling control method and related device using a hot channel. Background Art

[0002] With the rapid development of the Internet and computer technologies, various services provided based on the Internet have become increasingly numerous. As the carrier of Internet services, the construction requirements of data centers are also constantly changing and increasing. During the operation of a data center, a large amount of heat is generated. If its cooling management is improper, it will cause the equipment to overheat, resulting in a decline in equipment performance and a shortening of service life, thus affecting the stability and security of the data center. Therefore, the cooling management of data centers is very important. Heat generation analysis is an extremely important step in data center cooling management. Currently, the heat generation analysis of data centers is usually achieved by relevant personnel through data comparison, but this method is too dependent on the professional qualities of relevant personnel and it is difficult to ensure the accuracy of heat generation analysis. At the same time, most current cooling strategies for data centers lack consideration of heat dissipation energy consumption control, resulting in the inability of the cooling process of data centers to take into account energy-saving control. And how to achieve adaptive control adjustment in the data center cooling process is also a problem that enterprises need to consider. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a data center cooling control method and related device using a hot channel, which can effectively regulate the cooling process of the data center and make the cooling process of the data center reach a more ideal effect.

[0004] To solve the above technical problems, the present invention provides a data center cooling control method using a hot channel, the method comprising: Constructing a digital twin model of the data center and analyzing the heat generation amount of the data center based on the digital twin model; Determining the airflow organization simulation data of the data center based on the digital twin model, and performing an airflow organization impact analysis based on the airflow organization simulation data to obtain airflow organization impact analysis data; Performing a heat dissipation energy consumption control analysis based on the airflow organization impact analysis data to obtain heat dissipation energy consumption control analysis data; Constructing a cooling strategy based on the heat generation amount, the airflow organization impact analysis data, and the heat dissipation energy consumption control analysis data in combination with the hot and cold channel load operation analysis; Performing a cooling simulation of the data center based on the cooling strategy to obtain a cooling simulation result, and optimizing the cooling strategy based on the cooling simulation result to obtain an optimized cooling strategy; Build an adaptive control system, and based on the optimized cooling strategy, combine the adaptive control system to control the hot channel and the cold channel for cooling the data center.

[0005] Optionally, building a digital twin model of the data center and analyzing the heat generation of the data center based on the digital twin model includes: Build a three-dimensional model of the data center based on the building data and equipment parameters of the data center; Perform mapping analysis on the three-dimensional model and the object entities of the data center to obtain the mapping point offset value, and correct the three-dimensional model based on the mapping point offset value to obtain the corrected three-dimensional model; Analyze the equipment loss of the data center, and build a digital twin model based on the equipment loss and the corrected three-dimensional model; Perform a heat source distribution simulation based on the digital twin model to obtain the corresponding temperature field, and analyze the heat generation of the data center using the heat generation analysis model based on the corresponding temperature field and the temperature of the dissipation points.

[0006] Optionally, determining the airflow organization simulation data of the data center based on the digital twin model and performing an airflow organization impact analysis based on the airflow organization simulation data to obtain the airflow organization impact analysis data includes: Perform an airflow organization simulation based on the digital twin model to obtain the airflow organization simulation data, and evaluate the airflow organization simulation data to obtain the airflow organization evaluation data; Perform an airflow organization impact analysis based on the airflow organization evaluation data to obtain the airflow organization impact analysis data.

[0007] Optionally, performing a heat dissipation energy consumption control analysis based on the airflow organization impact analysis data to obtain the heat dissipation energy consumption control analysis data includes: Build a cooling energy consumption control model using the airflow organization impact analysis data based on the power of the refrigeration system in the hot and cold channels and the air velocity of the data center; Build a constraint function based on the total energy consumption target, and perform a heat dissipation energy consumption control analysis based on the constraint function and the cooling energy consumption control model combined with the particle swarm algorithm to obtain the heat dissipation energy consumption control analysis data.

[0008] Optionally, building a cooling strategy based on the heat generation, the airflow organization impact analysis data, and the heat dissipation energy consumption control analysis data combined with the load operation analysis of the hot and cold channels includes: Build a load operation analysis model based on the historical hot and cold channel temperature difference data; Perform temperature prediction of the hot and cold channels based on the heat generation, the airflow organization impact analysis data, and the heat dissipation energy consumption control analysis data to obtain the temperature prediction data; Based on the load operation analysis model, temperature prediction data is used to analyze the load operation status of the hot and cold channels, and load operation status data is obtained; Based on the heat generation amount, air flow organization influence analysis data, heat dissipation energy consumption control analysis data, and load operation status data, the control parameters of the hot and cold channels are determined; Set thermal safety parameters, and based on the control parameters, thermal safety parameters, and heat dissipation energy consumption control analysis data, use Bayesian optimizer data to construct a cooling strategy.

[0009] Optionally, performing a cooling simulation of the data center based on the cooling strategy to obtain a cooling simulation result, and optimizing the cooling strategy based on the cooling simulation result to obtain an optimized cooling strategy, including: Input the cooling strategy into the simulation software to perform a cooling simulation of the data center to obtain a cooling simulation result; Extract feedback data based on the cooling simulation result, and optimize the cooling strategy based on the feedback data to obtain an optimized cooling strategy.

[0010] Optionally, constructing an adaptive control system, and based on the optimized cooling strategy, combining the adaptive control system to control the hot and cold channels for cooling treatment of the data center, including: Establish an object model of the data center, and construct an internal model controller based on the object model; Based on the object model, use a fractional-order complex network system to construct a state error system; Based on the equivalent model generated by the object model, perform parameter tuning on the conventional PID controller to obtain a conventional PID controller with tuned parameters; Construct an adaptive control system based on the internal model controller, the state error system, and the conventional PID controller with tuned parameters; During the process of controlling the hot and cold channels for cooling treatment of the data center based on the optimized cooling strategy, detect the deviation value between the real-time temperature of the data center and the preset target temperature, and based on the adaptive control system, generate a regulation parameter using the deviation value, and adjust the operation parameters of the hot and cold channels in real time based on the regulation parameter.

[0011] In addition, the present invention also provides a data center cooling control device using a hot channel, and the device includes: Heat generation analysis module: used to construct a digital twin model of the data center, and analyze the heat generation amount of the data center based on the digital twin model; Air flow organization influence analysis module: used to determine the air flow organization simulation data of the data center based on the digital twin model, and perform air flow organization influence analysis based on the air flow organization simulation data to obtain air flow organization influence analysis data; Heat dissipation energy consumption control analysis module: used to perform heat dissipation energy consumption control analysis based on the airflow organization impact analysis data to obtain heat dissipation energy consumption control analysis data; Cooling strategy module: used to construct a cooling strategy based on the heat generation amount, airflow organization impact analysis data, and heat dissipation energy consumption control analysis data in combination with the hot and cold channel load operation analysis; Strategy optimization module: used to perform cooling simulation of the data center based on the cooling strategy to obtain cooling simulation results, and optimize the cooling strategy based on the cooling simulation results to obtain an optimized cooling strategy; Cooling processing module: used to construct an adaptive control system, and control the hot channel and cold channel for cooling processing of the data center based on the optimized cooling strategy in combination with the adaptive control system.

[0012] In addition, the present invention also provides an electronic device, the electronic device includes a processor and a memory, the memory is used to store instructions, and the processor is used to call the instructions in the memory so that the electronic device executes the above data center cooling control method using the hot channel.

[0013] In addition, the present invention also provides a computer-readable storage medium, the computer-readable storage medium stores computer instructions, and when the computer instructions run on an electronic device, the electronic device is enabled to execute the above data center cooling control method using the hot channel.

[0014] In the embodiment of the present invention, constructing a digital twin model of the data center to analyze the heat generation amount of the data center can effectively improve the accuracy and efficiency of the heat generation analysis of the data center. Determining the airflow organization simulation data of the data center based on the digital twin model, performing airflow organization impact analysis based on the airflow organization simulation data, and performing heat dissipation energy consumption control analysis based on the airflow organization impact analysis data make the heat dissipation energy consumption control analysis more refined and can improve the reliability of the heat dissipation energy consumption control analysis. Constructing a cooling strategy based on the heat generation amount, airflow organization impact analysis data, and heat dissipation energy consumption control analysis data in combination with the hot and cold channel load operation analysis enables the constructed cooling strategy to take into account the energy-saving control of the hot and cold channels and their refrigeration systems while achieving effective cooling of the data center, and can effectively save energy. Performing cooling simulation of the data center based on the cooling strategy, optimizing the cooling strategy based on the cooling simulation results to obtain an optimized cooling strategy can make the optimized cooling strategy more in line with the actual situation of the data center. Constructing an adaptive control system and controlling the hot channel and cold channel for cooling processing of the data center based on the optimized cooling strategy in combination with the adaptive control system can effectively regulate the cooling process of the data center and avoid a large deviation between the actual cooling effect and the expected effect. Description of the Drawings

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0016] Figure 1 is a schematic flowchart of a data center cooling control method using a hot aisle in an embodiment of the present invention; Figure 2 is a schematic structural composition diagram of a data center cooling control device using a hot aisle in an embodiment of the present invention; Figure 3 is a schematic structural composition diagram of an electronic device in an embodiment of the present invention. Detailed implementation manners

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0018] Embodiment 1

[0019] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of a data center cooling control method using a hot aisle in an embodiment of the present invention, and the method includes: S11: Construct a digital twin model of the data center, and analyze the heat generation amount of the data center based on the digital twin model; In the specific implementation process of the present invention, the construction of the digital twin model of the data center and the analysis of the heat generation amount of the data center based on the digital twin model include: constructing a three-dimensional model of the data center based on the building data and equipment parameters of the data center; performing mapping analysis on the three-dimensional model and the object entities of the data center to obtain the mapping point offset value, and correcting the three-dimensional model based on the mapping point offset value to obtain a corrected three-dimensional model; analyzing the equipment losses of the data center, and constructing a digital twin model based on the equipment losses and the corrected three-dimensional model; performing a heat source distribution simulation on the digital twin model to obtain the corresponding temperature field, and analyzing the heat generation amount of the data center by using a heat generation analysis model in combination with the temperature at the dissipation points based on the corresponding temperature field.

[0020] Specifically, a three-dimensional model of the data center is constructed based on the building data and equipment parameters of the data center. The building data includes the structure, equipment layout and configuration of the data center, and the equipment parameters include the size, structure, operating status, temperature, etc. of each device in the data center. The building data and equipment parameters of the data center are input into three-dimensional model generation software to obtain a three-dimensional model of the data center. Perform mapping analysis on the three-dimensional model and the object entities of the data center, collect the point cloud data of each target object in the data center, and the target objects include each device in the data center. Discrete point removal is performed on the point cloud data. Discrete point removal can be achieved by calculating the Euclidean distance between the point cloud data and the three-dimensional model of the data center, and taking the point cloud data points with Euclidean distance exceeding the distance threshold as discrete points and removing them to obtain standardized point cloud data. Based on the standardized point cloud, the coordinate points of the target object entity are obtained, and the offset value is calculated according to the coordinate points of the target object entity and the corresponding mapping points in the three-dimensional model of the data center. The mapping point is the coordinate point of the target object entity and the corresponding spatial point in the three-dimensional model of the data center, that is, the mapping point offset value is obtained, and the three-dimensional model is corrected based on the mapping point offset value, that is, the spatial point coordinates corresponding to the data center are adjusted according to the mapping point offset value to obtain a corrected three-dimensional model. Analyze the equipment loss of the data center, obtain the operation and maintenance data of each device in the data center, and construct an equipment degradation analysis model. The equipment degradation analysis model can be a deep convolutional neural network model. According to the equipment degradation analysis model, use the operation and maintenance data to analyze the degradation status of the equipment, and determine the corresponding equipment loss according to the degradation status of the equipment. The operation and maintenance data includes the number of equipment repairs, repair fault types, etc., and a digital twin model is constructed based on the equipment loss and the corrected three-dimensional model. The equipment loss and the corrected three-dimensional model are input into the object software of the digital twin body, and the relevant data of the equipment loss and the corrected three-dimensional model can be interacted with the data twin body object in the object software to form a digital twin model. Based on the digital twin model, a heat source distribution simulation is carried out. Each device twin model in the digital twin model is used as a heat source, and fluid dynamics simulation analysis is performed according to the distribution of the heat source. The fluid dynamics simulation analysis can be used for the simulation analysis of heat conduction and heat transfer to obtain the corresponding temperature field.Analyze the heat production of the data center by using the heat production analysis model based on the corresponding temperature field and combining the temperature analysis of the dissipation points. The dissipation points refer to the key positions where the air flow in the data center flows, such as the areas where the server temperature changes significantly, the positions of the exhaust fans, etc. Obtain the equipment power data in the data center, analyze the heat production waveform data of the equipment according to the equipment power data. The heat production waveform data is the heat change data during the heat production process of the equipment. Construct a heat production prediction model according to the heat production waveform data. The heat production prediction model is a convergence model obtained by inputting the heat production waveform data as a sample data set into a deep neural network for training. Analyze the heat production of the dissipation points according to the heat production prediction model, and analyze the dissipated heat according to the heat production and the preset heat transfer coefficient. The dissipated heat is the heat dissipation formed after the equipment produces heat and contacts the surrounding environment. For example, the equipment produces heat and contacts the air for heat dissipation. Use the exponential smoothing algorithm to perform exponential smoothing on the heat production and the dissipated heat to obtain the heat production and the dissipated heat at the next moment. The exponential smoothing algorithm is a data analysis algorithm for short-term prediction. Analyze the temperature change data of the dissipation points according to the heat production and the heat production at the next moment and the dissipated heat and the dissipated heat at the next moment. Input the corresponding temperature field and the temperature change data of the dissipation points into the heat production analysis model for heat production analysis. The heat production analysis model can adopt a deep learning model to obtain the heat production of the data center. The heat production of the data center obtained in this way is closer to the actual situation and can more comprehensively reflect the temperature conditions inside the data center.

[0021] S12: Determine the air flow organization simulation data of the data center based on the digital twin model, and perform air flow organization impact analysis based on the air flow organization simulation data to obtain air flow organization impact analysis data; In the specific implementation process of the present invention, the determining the air flow organization simulation data of the data center based on the digital twin model, and performing air flow organization impact analysis based on the air flow organization simulation data to obtain air flow organization impact analysis data includes: performing air flow organization simulation based on the digital twin model to obtain air flow organization simulation data, and evaluating the air flow organization simulation data to obtain air flow organization evaluation data; performing air flow organization impact analysis based on the air flow organization evaluation data to obtain air flow organization impact analysis data.

[0022] Specifically, based on the digital twin model, air distribution simulation is carried out. In the simulation software according to the digital twin model, there are computational fluid dynamics (CFD) simulation tools that can simulate air distribution to obtain air distribution simulation data, such as wind speed distribution, gas flow conditions, temperature field distribution, changes, etc. Then, the air distribution simulation data is evaluated. The air distribution evaluation indicators can adopt the inlet and outlet air temperature difference indicator, heating index, and heat mixing index. The heating index is used to characterize the mixing degree of the supply air in the cold aisle with the hot air before entering. The heat mixing index is used to characterize the worst air mixing situation in the racks of the data center cabinets. The average temperature difference indicator of the inlet and outlet air of the rack is used to evaluate the possibility of local hot spots, that is, according to the air distribution evaluation index values obtained from the analysis of the air distribution simulation data. The closer the heating index is to 0, the less the mixing of cold and hot air, and the higher the supply air efficiency. The smaller the heat mixing index, the smaller the degree of mixing of cold and hot air, and the more stable the thermal environment of the data center. The smaller the inlet and outlet air temperature difference indicator, the higher the air distribution evaluation. Thus, air distribution evaluation data is obtained. Based on the air distribution evaluation data, an air distribution impact analysis is carried out. By analyzing the air distribution evaluation data, the factors affecting air distribution are determined. By matching the respective evaluation index values in the air distribution evaluation data with the factors affecting air distribution, the factors affecting air distribution include the enclosure degree of the cold and hot aisles, wind speed, wind direction, supply air temperature, etc. The matching data of the air distribution evaluation data and the corresponding factors affecting air distribution is generated as air distribution impact analysis data.

[0023] S13: Based on the air distribution impact analysis data, a heat dissipation energy consumption control analysis is carried out to obtain heat dissipation energy consumption control analysis data; In the specific implementation process of the present invention, the heat dissipation energy consumption control analysis based on the air distribution impact analysis data to obtain heat dissipation energy consumption control analysis data includes: constructing a cooling energy consumption control model based on the power of the refrigeration system in the cold and hot aisles and the air velocity in the data center using the air distribution impact analysis data; constructing a constraint function based on the total energy consumption target, and combining the constraint function and the cooling energy consumption control model with the particle swarm optimization algorithm to carry out heat dissipation energy consumption control analysis to obtain heat dissipation energy consumption control analysis data.

[0024] Specifically, based on the power of the refrigeration system in the hot and cold channels and the air flow rate in the data center, a cooling energy consumption control model is constructed using the data obtained from the analysis of the influence of air flow organization. According to the power of the refrigeration system in the hot and cold channels, the energy consumption corresponding to the power of the refrigeration system is determined. The air flow rate data, the total power, and the energy consumption of the refrigeration system in the data center are obtained under different set temperatures and set wind speeds of the refrigeration system. The above data, combined with the data obtained from the analysis of the influence of air flow organization and the different heat generation values in the data center, are used as training data. The extreme learning machine model is selected as the initial cooling energy consumption control model, and the extreme learning machine model is trained using the training data. The trained extreme learning machine model is the cooling energy consumption control model. A constraint function is constructed based on the total energy consumption target, and a total energy consumption target threshold is set. The constraint function is set according to the total energy consumption target threshold, which means that the total energy consumption of the refrigeration system cannot exceed the total energy consumption target threshold. Based on the constraint function and the cooling energy consumption control model, the particle swarm optimization algorithm is used to analyze the cooling energy consumption control. Using the particle swarm optimization algorithm, the operating parameters of the refrigeration system are analyzed using the cooling energy consumption control model under the constraint function, and the set temperature and set wind speed of the refrigeration system with the lowest total power corresponding to different heat generation amounts in the data center are obtained, that is, the cooling energy consumption control analysis data is obtained.

[0025] S14: Construct a cooling strategy based on the heat generation amount, the data obtained from the analysis of the influence of air flow organization, and the cooling energy consumption control analysis data in combination with the load operation analysis of the hot and cold channels; In the specific implementation process of the present invention, the construction of the cooling strategy based on the heat generation amount, the data obtained from the analysis of the influence of air flow organization, and the cooling energy consumption control analysis data in combination with the load operation analysis of the hot and cold channels includes: constructing a load operation analysis model based on the historical hot and cold channel temperature difference data; predicting the temperature of the hot and cold channels based on the heat generation amount, the data obtained from the analysis of the influence of air flow organization, and the cooling energy consumption control analysis data to obtain temperature prediction data; analyzing the load operation state of the hot and cold channels using the load operation analysis model with the temperature prediction data to obtain load operation state data; determining the control parameters of the hot and cold channels based on the heat generation amount, the data obtained from the analysis of the influence of air flow organization, the cooling energy consumption control analysis data, and the load operation state data; setting the thermal safety parameters, and constructing a cooling strategy using the control parameters, the thermal safety parameters, and the cooling energy consumption control analysis data with the Bayesian optimizer data.

[0026] Specifically, a load operation analysis model is constructed based on historical cold and hot aisle temperature difference data. The historical cold and hot aisle temperature difference data is obtained, and the historical cold and hot aisle temperature difference data is used as a data set for model training to obtain the load operation analysis model. Temperature prediction of the cold and hot aisles is performed based on heat production, airflow organization impact analysis data, and heat dissipation energy consumption control analysis data. The operating parameters of the refrigeration system in the cold and hot aisles are determined from the heat production, airflow organization impact analysis data, and heat dissipation energy consumption control analysis data. The operating parameters include operating wind speed, power, refrigeration temperature, etc. The temperature prediction model is used to predict the temperature of the cold and hot aisles during operation based on the operating parameters of the refrigeration system, and the predicted cold aisle temperature value and the predicted hot aisle temperature value are obtained, that is, temperature prediction data is obtained. Based on the load operation analysis model, the load operation status of the cold and hot aisles is analyzed using the temperature prediction data. The temperature difference value between the predicted cold aisle temperature value and the predicted hot aisle temperature value is calculated based on the temperature prediction data, and the load operation status of the air conditioning system in the cold aisle and the heat dissipation equipment in the hot aisle is analyzed using the load operation analysis model based on the temperature difference value. The heat dissipation equipment can use an exhaust component, that is, load operation status data is obtained. The control parameters of the cold and hot aisles are determined based on the heat production, airflow organization impact analysis data, heat dissipation energy consumption control analysis data, and load operation status data, that is, the operating parameters of the air conditioner in the cold aisle and the heat dissipation equipment in the hot aisle are determined using the operating parameter sequence analysis model based on the heat production, airflow organization impact analysis data, heat dissipation energy consumption control analysis data, and load operation status data, such as the refrigeration temperature of the air conditioner in the cold aisle and the fan speed of the heat dissipation equipment in the hot aisle. A thermal safety parameter is set, and the thermal safety parameter is the thermal safety temperature target, that is, the temperature safety threshold. A cooling strategy is constructed using the Bayesian optimizer based on the control parameters, thermal safety parameters, and heat dissipation energy consumption control analysis data. A Bayesian optimizer is constructed. The Bayesian optimizer includes a kernel function based on Gaussian regression and an acquisition function. The acquisition function uses the upper confidence bound algorithm. The control parameters are error-optimized using the Bayesian optimizer in combination with the thermal safety parameters and heat dissipation energy consumption control analysis data. Since there may be certain modeling errors in the analysis model when determining the control parameters through the analysis model, the analysis model has been trained with relevant data sets to reduce the generalization error. However, the generalization error is an average measure. Therefore, the control parameters obtained from the analysis model may still have relatively large errors. Therefore, after the Bayesian optimizer realizes this error, the control parameters are error-optimized to find more optimal control parameters under temperature and energy consumption constraints. A cooling strategy is constructed using the control parameters after error optimization to obtain a more reliable cooling strategy.

[0027] S15: Perform a cooling simulation of the data center based on the cooling strategy to obtain a cooling simulation result, and optimize the cooling strategy based on the cooling simulation result to obtain an optimized cooling strategy; In the specific implementation process of the present invention, the cooling simulation of the data center is carried out based on the cooling strategy to obtain the cooling simulation result, and the cooling strategy is optimized based on the cooling simulation result to obtain the optimized cooling strategy, including: inputting the cooling strategy into the simulation software to carry out the cooling simulation of the data center to obtain the cooling simulation result; extracting feedback data based on the cooling simulation result, and optimizing the cooling strategy based on the feedback data to obtain the optimized cooling strategy.

[0028] Specifically, input the cooling strategy into the simulation software to carry out the cooling simulation of the data center to obtain the cooling simulation result. Extract feedback data based on the cooling simulation result, determine the cooling effect of the data center according to the cooling simulation result, conduct feedback analysis on the cooling effect, compare the cooling effect with the expected effect to obtain the feedback data, and optimize the cooling strategy based on the feedback data, that is, adjust the device control parameters in the hot and cold channels of the cooling strategy according to the feedback data to obtain the optimized cooling strategy.

[0029] S16: Construct an adaptive control system, and based on the optimized cooling strategy, combine the adaptive control system to control the hot channel and the cold channel for the cooling process of the data center.

[0030] In the specific implementation process of the present invention, the construction of the adaptive control system, and based on the optimized cooling strategy, combine the adaptive control system to control the hot channel and the cold channel for the cooling process of the data center, including: establishing an object model of the data center, constructing an internal model controller based on the object model; constructing a state error system based on the object model using a fractional-order complex network system; carrying out parameter tuning on a conventional PID controller based on the equivalent model generated by the object model to obtain a conventional PID controller with tuned parameters; constructing an adaptive control system based on the internal model controller, the state error system, and the conventional PID controller with tuned parameters; during the process of controlling the hot channel and the cold channel for the cooling process of the data center based on the optimized cooling strategy, detecting the deviation value between the real-time temperature of the data center and the preset target temperature, generating an adjustment parameter based on the deviation value using the adaptive control system, and adjusting the operating parameters of the hot channel and the cold channel in real time based on the adjustment parameter.

[0031] Specifically, an object model of the data center is established. The object model is an abstraction of the data center entity. Taking the data center as the controlled object, a preset-level abstraction is made through the controlled object and the corresponding function, and then the corresponding object model is formed. Based on the object model, an internal model controller is constructed. The object model is divided into more simplified units through the preset level. Each unit has a corresponding set of basis functions. The solution vector is calculated according to the basis functions. The relative order of the object model can be obtained through the geometric combination interpolation of the solution vector and the basis functions in all units. According to the relative order, a low-pass filter function is used to construct the internal model controller. The low-pass filter function is used to ensure the stability and robustness of the internal model controller, and adding the low-pass filter function can make its parameters become adjustable parameters of the internal model controller. The internal model controller can parallel the controlled object to the corresponding loop, making the output result closer to the ideal output value. Based on the object model, a state error system is constructed using a fractional-order complex network system. A differential equation corresponding to the object model is established. The fractional-order derivative is taken with a fractional-order order. The fractional-order derivative is a fractional-order derivative in the sense of Caputo. According to the fractional-order order, a fractional-order complex network system is constructed using the corresponding differential equation. The fractional-order complex network system increases the infinite memory and hereditary properties through the fractional-order derivative, which is more in line with the actual network. Moreover, it increases the degrees of freedom through the fractional-order derivative, making the network system more practical in the actual application field. Considering the existence of uncertainties and unpredictable nonlinear states, a non-periodic orbit and an equivalent perturbation for the parameters corresponding to the nonlinear state are introduced to construct a state equation. According to the state equation, a state error system between the object model and the fractional-order complex network system is constructed.The parameters of a conventional Proportion Integration Differentiation (PID) controller are tuned based on an equivalent model generated from an object model. The conventional PID controller is constructed with the proportional coefficient, integral coefficient, and differential coefficient obtained from a database. By fitting the object model and using the frequency domain analysis method to obtain the parameters of the low-order equivalent system of the object model, the characteristics in the main frequency band are highlighted. Through model fitting with the low-order equivalent system parameters, an equivalent model is obtained. The equivalent model is divided into more simplified units through preset levels, and each unit has a corresponding set of basis functions. The solution vector is calculated according to the basis functions, and the two-order derivatives at zero complex variable are calculated according to the solution vector to obtain the first-order derivative and the second-order derivative. The external disturbance parameter is calculated through a preset filtering function. The target complex variable is obtained based on the first-order derivative and the second-order derivative. A first-order inertial transfer function is constructed based on the target complex variable and the external disturbance parameter. The definite integral of the first-order inertial transfer function is calculated based on the parameter splitting decomposition theorem and transformed into two simple integrals to obtain the target parameters. The parameters of the conventional PID controller are tuned through the target parameters, that is, the control unit parameters are changed through the target parameters, such as the proportional coefficient and integral coefficient of the controller, so that the obtained controller can output more accurate results, and thus the conventional PID controller with tuned parameters is obtained. An adaptive control system is constructed based on an internal model controller, a state error system, and the conventional PID controller with tuned parameters. An extended state observer is introduced. The extended state observer regards the uncertain terms or disturbances in the system as one of the states in the system, establishes a new state space, and observes this new state space. The system equivalent disturbance of the state error system is observed through the extended state observer to obtain the extended state vector. The extended state feedback compensation is obtained through the extended state vector using the preset bandwidth parameter. The conventional PID controller with tuned parameters is adjusted according to the extended state feedback compensation, that is, the parameters of the conventional PID controller are optimized. The adjustment of the extended state feedback on the PID controller is mainly reflected in optimizing the parameters of the PID controller by estimating the internal state and external disturbance of the system, improving the robustness and control accuracy of the system. Then, the internal model controller is fused with the optimized conventional PID controller, that is, the control parameters between the two are fused to form an adaptive control system. During the process of cooling the data center based on an optimized cooling strategy to control the hot channel and the cold channel, the deviation value between the real-time temperature of the data center and the preset target temperature is detected, and the adjustment parameter is generated based on the deviation value using the adaptive control system. The operating parameters of the hot channel and the cold channel are adjusted in real time based on the adjustment parameter, that is, the control parameters of the heat dissipation equipment in the hot channel and the control parameters of the air conditioning system in the cold channel are adjusted to achieve better cooling performance.

[0032] In the embodiments of the present invention, constructing a digital twin model of a data center to analyze the heat generation of the data center can effectively improve the accuracy and efficiency of heat generation analysis in the data center. Determining the airflow organization simulation data of the data center based on the digital twin model, and conducting an analysis of the influence of the airflow organization based on the airflow organization simulation data, and conducting an analysis of heat dissipation energy consumption control based on the data of the influence analysis of the airflow organization makes the analysis of heat dissipation energy consumption control more refined and can improve the reliability of the analysis of heat dissipation energy consumption control. Constructing a cooling strategy by combining the heat generation amount, the data of the influence analysis of the airflow organization, and the data of the heat dissipation energy consumption control analysis with the load operation analysis of the hot and cold channels enables the constructed cooling strategy to take into account the energy-saving control of the hot and cold channels and their refrigeration systems while achieving effective cooling of the data center, and can effectively save energy. Conducting a cooling simulation of the data center based on the cooling strategy, and optimizing the cooling strategy based on the results of the cooling simulation to obtain an optimized cooling strategy can make the optimized cooling strategy more in line with the actual situation of the data center. Constructing an adaptive control system, and controlling the hot and cold channels for cooling the data center based on the optimized cooling strategy combined with the adaptive control system can effectively regulate the cooling process of the data center and avoid a large deviation between the actual cooling effect and the expected effect.

[0033] Embodiment 2

[0034] Please refer to Figure 2 , Figure 2 which is a schematic structural diagram of a data center cooling control device using a hot channel in the embodiments of the present invention. The device includes: Heat generation analysis module 21: used to construct a digital twin model of the data center and analyze the heat generation amount of the data center based on the digital twin model; Airflow organization influence analysis module 22: used to determine the airflow organization simulation data of the data center based on the digital twin model, and conduct an analysis of the influence of the airflow organization based on the airflow organization simulation data to obtain data of the influence analysis of the airflow organization; Heat dissipation energy consumption control analysis module 23: used to conduct an analysis of heat dissipation energy consumption control based on the data of the influence analysis of the airflow organization to obtain data of the heat dissipation energy consumption control analysis; Cooling strategy module 24: used to construct a cooling strategy by combining the heat generation amount, the data of the influence analysis of the airflow organization, and the data of the heat dissipation energy consumption control analysis with the load operation analysis of the hot and cold channels; Strategy optimization module 25: used to conduct a cooling simulation of the data center based on the cooling strategy to obtain the results of the cooling simulation, and optimize the cooling strategy based on the results of the cooling simulation to obtain an optimized cooling strategy; Cooling processing module 26: used to construct an adaptive control system, and control the hot and cold channels for cooling the data center based on the optimized cooling strategy combined with the adaptive control system.

[0035] In the specific implementation process of the present invention, for the specific implementation manner of the device item, reference may be made to the implementation manner of the above method item, which will not be elaborated herein.

[0036] In an embodiment of the present invention, constructing a digital twin model of a data center to analyze the heat generation of the data center can effectively improve the accuracy and efficiency of heat generation analysis of the data center. Determining the airflow organization simulation data of the data center based on the digital twin model, and conducting an analysis of the influence of the airflow organization based on the airflow organization simulation data, and conducting an analysis of heat dissipation energy consumption control based on the data of the influence analysis of the airflow organization, making the analysis of heat dissipation energy consumption control more refined, and improving the reliability of the analysis of heat dissipation energy consumption control. Constructing a cooling strategy by combining the heat generation amount, the data of the influence analysis of the airflow organization, and the data of the heat dissipation energy consumption control analysis with the load operation analysis of the hot and cold channels, so that the constructed cooling strategy can not only achieve effective cooling of the data center, but also take into account the energy-saving control of the hot and cold channels and their refrigeration systems, and can effectively save energy. Conducting a cooling simulation of the data center based on the cooling strategy, and optimizing the cooling strategy based on the results of the cooling simulation to obtain an optimized cooling strategy, which can make the optimized cooling strategy more in line with the actual situation of the data center. Constructing an adaptive control system, and controlling the hot and cold channels for cooling the data center based on the optimized cooling strategy combined with the adaptive control system, which can effectively regulate the cooling process of the data center and avoid a large deviation between the actual cooling effect and the expected effect.

[0037] A computer-readable storage medium provided by an embodiment of the present invention, on which a computer program is stored, and when the program is executed by a processor, it implements the method for controlling the cooling of a data center using a hot channel in any one of the above embodiments. Wherein, the computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or optical card. That is, the storage device includes any medium that can store or transmit information in a readable form by a device (such as a computer, mobile phone), and can be a read-only memory, a magnetic disk or an optical disk, etc.

[0038] Embodiment III

[0039] Please refer to Figure 3 , Figure 3It is a schematic diagram of the structural composition of the electronic device in the embodiment of the present invention.

[0040] The embodiment of the present invention also provides an electronic device, as Figure 3 shown, the electronic device includes a memory 31, a processor 33, and a computer program 32 stored in the memory 31 and executable on the processor 33. Those skilled in the art can understand that Figure 3 the illustrated electronic device does not constitute a limitation on all devices, and may include more or fewer components than shown, or combine certain components. The memory 31 can be used to store the computer program 32 and each functional module. The processor 33 runs the computer program 32 stored in the memory 31, thereby performing various functional applications and data processing of the device. The memory can be an internal memory or an external memory, or include both an internal memory and an external memory. The internal memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, or a random access memory. The external memory can include a hard disk, a floppy disk, a ZIP disk, a USB flash drive, a magnetic tape, etc. The processor 33 can be a central processing unit (CPU), or can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, a single-chip microcomputer, or the processor 33 can also be any conventional processor, etc. The processors and memories disclosed in the present invention include but are not limited to these types of processors and memories. The processors and memories disclosed in the present invention are only examples and not limitations.

[0041] As an embodiment, the electronic device includes: one or more processors 33, a memory 31, one or more computer programs 32, wherein the one or more computer programs 32 are stored in the memory 31 and configured to be executed by the one or more processors 33, and the one or more computer programs 32 are configured to execute the data center cooling control method using a hot channel in any of the above embodiments. For the specific implementation process, please refer to the above embodiments and will not be elaborated here.

[0042] In the embodiments of the present invention, constructing a digital twin model of a data center to analyze the heat generation of the data center can effectively improve the accuracy and efficiency of heat generation analysis of the data center. Determining the airflow organization simulation data of the data center based on the digital twin model, and conducting an analysis of the influence of airflow organization based on the airflow organization simulation data, and conducting an analysis of heat dissipation energy consumption control based on the data of the influence analysis of airflow organization, making the analysis of heat dissipation energy consumption control more refined and capable of improving the reliability of the analysis of heat dissipation energy consumption control. Constructing a cooling strategy by combining the heat generation, the data of the influence analysis of airflow organization, and the data of the analysis of heat dissipation energy consumption control with the analysis of the operation of the hot and cold aisle loads, so that the constructed cooling strategy not only realizes the effective cooling of the data center but also takes into account the energy-saving control of the hot and cold aisles and their refrigeration systems, and can effectively save energy. Conducting a cooling simulation of the data center based on the cooling strategy, and optimizing the cooling strategy based on the results of the cooling simulation to obtain an optimized cooling strategy, which can make the optimized cooling strategy more in line with the actual situation of the data center. Constructing an adaptive control system, and controlling the hot and cold aisles for the cooling process of the data center based on the optimized cooling strategy combined with the adaptive control system, which can effectively regulate the cooling process of the data center and avoid a large deviation between the actual cooling effect and the expected effect.

[0043] In addition, the above has introduced in detail a method and related device for cooling control of a data center using a hot aisle provided by the embodiments of the present invention. Specific examples should have been used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A data center cooling control method using a hot channel, characterized in that The method includes: Constructing a digital twin model of the data center and analyzing the heat generation amount of the data center based on the digital twin model; Determining the airflow organization simulation data of the data center based on the digital twin model, and performing an analysis of the influence of the airflow organization based on the airflow organization simulation data to obtain the airflow organization influence analysis data; Performing a heat dissipation energy consumption control analysis based on the airflow organization influence analysis data to obtain the heat dissipation energy consumption control analysis data; Constructing a cooling strategy based on the heat generation amount, the airflow organization influence analysis data, and the heat dissipation energy consumption control analysis data in combination with the load operation analysis of the hot and cold channels; Performing a cooling simulation of the data center based on the cooling strategy to obtain the cooling simulation result, and optimizing the cooling strategy based on the cooling simulation result to obtain the optimized cooling strategy; Constructing an adaptive control system, and controlling the hot and cold channels for the cooling process of the data center based on the optimized cooling strategy in combination with the adaptive control system.

2. The data center cooling control method using a hot channel according to claim 1, wherein The constructing a digital twin model of the data center and analyzing the heat generation amount of the data center based on the digital twin model includes: Constructing a three-dimensional model of the data center based on the building data and equipment parameters of the data center; Performing a mapping analysis on the three-dimensional model and the object entities of the data center to obtain the mapping point offset value, and correcting the three-dimensional model based on the mapping point offset value to obtain the corrected three-dimensional model; Analyzing the equipment loss of the data center, and constructing a digital twin model based on the equipment loss and the corrected three-dimensional model; Performing a heat source distribution simulation on the digital twin model to obtain the corresponding temperature field, and analyzing the heat generation amount of the data center by using the heat generation analysis model based on the corresponding temperature field in combination with the temperature of the dissipation points.

3. The data center cooling control method using a hot channel according to claim 1, wherein The determining the airflow organization simulation data of the data center based on the digital twin model, and performing an analysis of the influence of the airflow organization based on the airflow organization simulation data to obtain the airflow organization influence analysis data includes: Performing an airflow organization simulation on the digital twin model to obtain the airflow organization simulation data, and evaluating the airflow organization simulation data to obtain the airflow organization evaluation data; Performing an analysis of the influence of the airflow organization based on the airflow organization evaluation data to obtain the airflow organization influence analysis data.

4. The data center cooling control method using a hot channel according to claim 1, characterized in that, The performing a heat dissipation energy consumption control analysis based on the airflow organization influence analysis data to obtain the heat dissipation energy consumption control analysis data includes: Constructing a cooling energy consumption control model by using the airflow organization influence analysis data based on the power of the refrigeration system in the hot and cold channels and the air velocity of the data center; Constructing a constraint function based on the total energy consumption target, and performing a heat dissipation energy consumption control analysis based on the constraint function and the cooling energy consumption control model in combination with the particle swarm optimization algorithm to obtain the heat dissipation energy consumption control analysis data.

5. The data center cooling control method using a hot channel according to claim 1, characterized in that The constructing a cooling strategy based on the heat generation amount, the airflow organization influence analysis data, and the heat dissipation energy consumption control analysis data in combination with the load operation analysis of the hot and cold channels includes: Constructing a load operation analysis model based on the historical hot and cold channel temperature difference data; Performing a temperature prediction of the hot and cold channels based on the heat generation amount, the airflow organization influence analysis data, and the heat dissipation energy consumption control analysis data to obtain the temperature prediction data; Performing a load operation state analysis of the hot and cold channels by using the temperature prediction data based on the load operation analysis model to obtain the load operation state data; Determine the control parameters of the hot and cold channels based on the heat generation amount, air flow organization influence analysis data, heat dissipation energy consumption control analysis data, and load operation status data; Set the thermal safety parameters, and construct a cooling strategy based on the control parameters, thermal safety parameters, and heat dissipation energy consumption control analysis data using a Bayesian optimizer.

6. The data center cooling control method using a hot channel according to claim 1, wherein Perform a cooling simulation of the data center based on the cooling strategy to obtain a cooling simulation result, and optimize the cooling strategy based on the cooling simulation result to obtain an optimized cooling strategy, including: Input the cooling strategy into the simulation software to perform a cooling simulation of the data center and obtain a cooling simulation result; Extract feedback data based on the cooling simulation result, and optimize the cooling strategy based on the feedback data to obtain an optimized cooling strategy.

7. The data center cooling control method using a hot channel according to claim 1, characterized in that, Construct an adaptive control system, and control the hot and cold channels for the cooling process of the data center based on the optimized cooling strategy combined with the adaptive control system, including: Establish an object model of the data center, and construct an internal model controller based on the object model; Construct a state error system based on the object model using a fractional-order complex network system; Tune the parameters of the conventional PID controller based on the equivalent model generated by the object model to obtain a conventional PID controller with tuned parameters; Construct an adaptive control system based on the internal model controller, the state error system, and the conventional PID controller with tuned parameters; 8. A data center cooling control device using a hot channel, characterized in that, During the process of controlling the hot and cold channels for the cooling process of the data center based on the optimized cooling strategy, detect the deviation value between the real-time temperature of the data center and the preset target temperature, generate an adjustment parameter based on the deviation value using the adaptive control system, and adjust the operation parameters of the hot and cold channels in real time based on the adjustment parameter. The device includes: Heat generation analysis module: used to construct a digital twin model of the data center and analyze the heat generation amount of the data center based on the digital twin model; Air flow organization influence analysis module: used to determine the air flow organization simulation data of the data center based on the digital twin model, and perform an air flow organization influence analysis based on the air flow organization simulation data to obtain air flow organization influence analysis data; Heat dissipation energy consumption control analysis module: used to perform a heat dissipation energy consumption control analysis based on the air flow organization influence analysis data to obtain heat dissipation energy consumption control analysis data; Cooling strategy module: used to construct a cooling strategy based on the heat generation amount, air flow organization influence analysis data, and heat dissipation energy consumption control analysis data in combination with the load operation analysis of the hot and cold channels; Strategy optimization module: used to perform a cooling simulation of the data center based on the cooling strategy to obtain a cooling simulation result, and optimize the cooling strategy based on the cooling simulation result to obtain an optimized cooling strategy; 9. An electronic device, the electronic device comprising a processor and a memory, characterized in that, Cooling processing module: used to construct an adaptive control system, and control the hot and cold channels for the cooling process of the data center based on the optimized cooling strategy combined with the adaptive control system. The memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the data center cooling control method using the hot channel as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when run on an electronic device, cause the electronic device to execute the data center cooling control method using a hot channel as described in any one of claims 1 to 7.

Images