A data processing method and system for a carbon-energy prediction hybrid model in a data center

Abstract

The present invention relates to the technical field of energy and carbon data processing, and specifically discloses a data processing method and system for an energy and carbon prediction hybrid model in a data center. The method includes: collecting the basic load data, environmental condition data, and cooling equipment data of the data center; processing the basic load data of the data center to obtain a cooling demand evaluation index, matching to obtain an initial cooling medium flow margin, and obtaining each target area; processing the environmental condition data to obtain a data center environmental stability evaluation index, and comprehensively analyzing to obtain a cooling system efficiency evaluation value; correcting the prediction result of the energy and carbon prediction hybrid model. By providing a data processing method and system for an energy and carbon prediction hybrid model in a data center, the present invention can accurately understand the actual performance of the cooling system, so as to make precise matching according to actual needs, which helps to avoid over-design or under-design, and thus achieve the goal of energy conservation and emission reduction.

Description

Technical Field

[0001] The present invention relates to the technical field of energy and carbon data processing, and specifically to a data processing method and system for an energy and carbon prediction hybrid model in a data center. Background Art

[0002] Currently, as the infrastructure of modern information technology, the energy consumption and carbon emissions of data centers are increasing day by day. In order to reduce energy consumption and carbon emissions, it is particularly important to predict the energy and carbon of data centers. With the continuous progress of technology and the continuous deepening of applications, the hybrid model data processing method will play an increasingly important role in the field of energy and carbon prediction in data centers.

[0003] For example, the invention patent with the publication number CN114116183B is a method and system for data center service load scheduling based on deep reinforcement learning. The method includes: Step S1: Reading the characteristic information of heterogeneous service loads; Step S2: Real-time monitoring of the available computing resources of servers; Step S3: Making a scheduling decision based on the characteristic information of the load, the available resources of the server, and the real-time electricity price; Step S4: Modeling the scheduling process using a Markov decision process, aiming at minimizing the electricity cost of the data center, and training the scheduler using the Monte Carlo policy gradient algorithm.

[0004] For example, the invention patent with the publication number CN114860704A is a method and device for energy consumption monitoring and carbon emission accounting for a data center, including: Preprocessing the collected real-time energy consumption data to obtain an energy consumption report of each target device at equal time granularity; Calculating the total carbon emissions of each target device according to the energy consumption report, and calculating the carbon emission efficiency of the IT devices in the target device according to the total carbon emissions, generating a carbon emission report for each target device according to the total carbon emissions and the carbon emission efficiency, and inputting the energy consumption report and the carbon emission report into a deep neural network model for prediction to obtain a prediction result.

[0005] However, in the process of implementing the technical solutions of the present invention in the embodiments of the present application, it is found that the above technologies have at least the following technical problems: Currently, the data processing methods of the energy and carbon prediction hybrid model in data centers pay more attention to the selection and optimization of models, but in some cases, the integrity and accuracy of cooling system data may be ignored, resulting in deviations in prediction results, and it is difficult to guarantee the accuracy and effectiveness of data center energy and carbon emission management. Summary of the Invention

[0006] In view of the deficiencies of the prior art, the present invention provides a data processing method and system for an energy and carbon prediction hybrid model in a data center, which can effectively solve the problems involved in the above background art.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions: In the first aspect of the present invention, a data processing method for an energy-carbon prediction hybrid model of a data center is provided, including: collecting basic load data, environmental condition data, and cooling equipment data of the data center.

[0008] Process the basic load data of the data center to obtain a cooling demand assessment index, match the initial cooling medium flow margin according to the cooling demand assessment index and perform regional division to obtain each target area.

[0009] Process the environmental condition data to obtain a data center environmental stability assessment index, process the cooling equipment data to obtain a cooling equipment reliability assessment value, combine the cooling demand assessment index, comprehensively analyze to obtain a cooling system efficiency assessment value, and match the cooling medium flow margin demand value according to the cooling system efficiency assessment value.

[0010] Correct the prediction result of the energy-carbon prediction hybrid model according to the initial cooling medium flow margin and the cooling medium flow margin demand value.

[0011] As a further method, the process of processing the basic load data of the data center to obtain a cooling demand assessment index is as follows: The basic load data includes the floor area of the data center, IT equipment density, historical average operating temperature of IT equipment, and historical average CPU utilization rate.

[0012] Extract the critical data center floor area, critical IT equipment density, critical IT equipment operating temperature, and critical CPU utilization rate from the cooling system database, and comprehensively analyze to obtain a cooling demand assessment index.

[0013] As a further method, the specific analysis process of matching the initial cooling medium flow margin according to the cooling demand assessment index and performing regional division is as follows: Input the cooling demand assessment index into the cooling system database to match the corresponding initial cooling medium flow margin, input the data center floor area into the cooling system database to match the corresponding regional division size, and perform regional division on the data center according to the regional division size to obtain each target area.

[0014] As a further method, the process of processing the environmental condition data to obtain a data center environmental stability assessment index is as follows: The environmental condition data includes outdoor temperature, machine room dew point temperature, relative humidity, and dustiness.

[0015] Extract the critical outdoor temperature, reference standard machine room dew point temperature, allowable deviation machine room dew point temperature, reference standard relative humidity, allowable deviation relative humidity, and critical dustiness from the cooling system database, and comprehensively analyze to obtain a data center environmental stability assessment index.

[0016] As a further method, the reliability evaluation value of the cooling equipment is obtained by processing the cooling equipment data. The specific processing process is as follows: the cooling equipment data includes the cooling capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient.

[0017] Extract the reference standard cooling capacity of the cooling equipment, the allowable deviation of the cooling capacity of the cooling equipment, the reference standard flow pressure of the cooling medium, the allowable deviation of the flow pressure of the cooling medium, the reference standard number of redundant cooling equipment, the allowable deviation of the number of redundant cooling equipment, and the critical pipeline resistance coefficient from the cooling system database, and comprehensively analyze to obtain the reliability evaluation value of the cooling equipment.

[0018] As a further method, the efficiency evaluation value of the cooling system is obtained through comprehensive analysis. The specific analysis process is as follows: monitor the temperature of each target area to obtain the average monitored temperature of each target area, extract the area temperature threshold from the cooling system database, compare the average monitored temperature of each target area with the area temperature threshold, count the target areas where the average monitored temperature is greater than or equal to the area temperature threshold, mark them as each temperature anomaly area, and obtain the number of temperature anomaly areas.

[0019] Take the difference between the average monitored temperature of each temperature anomaly area and the area temperature threshold to obtain the temperature anomaly amplitude of each temperature anomaly area.

[0020] Based on the cooling demand evaluation index, the data center environment stability evaluation index, the reliability evaluation value of the cooling equipment, the temperature anomaly amplitude of each temperature anomaly area, and the number of temperature anomaly areas, comprehensively analyze to obtain the efficiency evaluation value of the cooling system.

[0021] As a further method, the required value of the cooling medium flow margin is obtained by matching according to the efficiency evaluation value of the cooling system. The specific process is as follows: input the efficiency evaluation value of the cooling system into the cooling system database to match the required value of the cooling medium flow margin corresponding to each interval of the efficiency evaluation value of the cooling system. The efficiency evaluation value of the cooling system is used to quantify the efficiency of the cooling system.

[0022] As a further method, the prediction result of the energy-carbon prediction hybrid model is corrected according to the initial cooling medium flow margin and the required value of the cooling medium flow margin. The specific process is as follows: compare the initial cooling medium flow margin with the required value of the cooling medium flow margin. If the initial cooling medium flow margin is greater than or equal to the required value of the cooling medium flow margin, no additional operation is performed. If the initial cooling medium flow margin is less than the required value of the cooling medium flow margin, the initial cooling medium flow margin is increased until it reaches the required value of the cooling medium flow margin.

[0023] The difference between the initial cooling medium flow margin and the cooling medium flow margin demand value is obtained to get the cooling medium flow margin difference. The cooling medium flow margin difference threshold is extracted from the cooling system database, and the cooling medium flow margin difference is compared with the cooling medium flow margin difference threshold. If the cooling medium flow margin difference is greater than or equal to the cooling medium flow margin difference threshold, the cooling medium flow margin difference is input into the cooling system database to match and obtain the corresponding energy-carbon prediction correction value, and the result of the energy-carbon prediction hybrid model is corrected according to the energy-carbon prediction correction value. If the cooling medium flow margin difference is less than the cooling medium flow margin difference threshold, no additional operation is performed.

[0024] As a further method, the cooling system efficiency evaluation value has the following specific numerical expression:

[0025]

[0026] Wherein, represents the cooling system efficiency evaluation value, represents the cooling demand evaluation index, represents the data center environment stability evaluation index, represents the cooling equipment reliability evaluation value, represents the number of temperature anomaly regions, represents the number of critical temperature anomaly regions, represents the temperature anomaly amplitude of the xth temperature anomaly region, represents the critical temperature anomaly amplitude, represents the cooling system efficiency evaluation influence factor corresponding to the set cooling demand evaluation index, represents the cooling system efficiency evaluation influence factor corresponding to the set data center environment stability evaluation index, represents the cooling system efficiency evaluation influence factor corresponding to the set cooling equipment reliability evaluation value, represents the cooling system efficiency evaluation influence factor corresponding to the set number of temperature anomaly regions, represents the cooling system efficiency evaluation influence factor corresponding to the set temperature anomaly amplitude. x represents the number of each temperature anomaly region, x = 1, 2, 3,..., y, and y represents the total number of temperature anomaly regions.

[0027] The second aspect of the present invention provides a data processing system for an energy-carbon prediction hybrid model of a data center, including: a data acquisition module for acquiring the basic load data, environmental condition data, and cooling equipment data of the data center.

[0028] The area division module is used to process the basic load data of the data center to obtain a cooling demand assessment index, match the initial cooling medium flow margin according to the cooling demand assessment index, and perform area division to obtain each target area.

[0029] The cooling system efficiency analysis module is used to process the environmental condition data to obtain a data center environmental stability assessment index, process the cooling equipment data to obtain a cooling equipment reliability assessment value, combine the cooling demand assessment index, comprehensively analyze to obtain a cooling system efficiency assessment value, and match the cooling medium flow margin demand value according to the cooling system efficiency assessment value.

[0030] The energy-carbon prediction hybrid model correction module is used to correct the prediction result of the energy-carbon prediction hybrid model according to the initial cooling medium flow margin and the cooling medium flow margin demand value.

[0031] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0032] (1) By providing a data processing method and system for an energy-carbon prediction hybrid model of a data center, the present invention can accurately understand the actual performance of the cooling system, so as to make precise matching according to actual needs, which helps to avoid over-design or under-design, ensure that the cooling system can meet the cooling needs of equipment or the environment without causing energy waste, and thus achieve the goal of energy conservation and emission reduction.

[0033] (2) By evaluating the cooling demand assessment index, the present invention can accurately understand the cooling demand of the data center, so as to take targeted cooling measures, avoid over-cooling or under-cooling, improve energy efficiency, reduce waste of cooling energy, further improve energy efficiency, reasonably allocate cooling resources, ensure that they operate in an efficient working state, and avoid resource waste.

[0034] (3) By evaluating the data center environmental stability assessment index, the present invention can reflect the overall condition of the data center environment, help the operation and maintenance team reasonably allocate resources, help maintain the best performance state of the equipment, improve the efficiency and accuracy of data processing and analysis, and can also optimize the energy use strategy to reduce the energy consumption of the data center. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The present invention is further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the following drawings without creative efforts.

[0036] Figure 1 It is a schematic flow chart of the method of the present invention.

[0037] Figure 2 Schematic diagram of the connection of system modules of the present invention.

[0038] Figure 3 Schematic diagram of the functional relationship between the cooling system efficiency evaluation value and the cooling demand evaluation index of the present invention. Detailed implementation manners

[0039] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0040] Referring to Figure 1 As shown, a data processing method for an energy-carbon prediction hybrid model of a data center provided by the first aspect of the present invention includes: collecting basic load data, environmental condition data, and cooling equipment data of the data center.

[0041] Processing the basic load data of the data center to obtain a cooling demand evaluation index, matching the initial cooling medium flow margin according to the cooling demand evaluation index and performing regional division to obtain each target region.

[0042] Processing the environmental condition data to obtain a data center environmental stability evaluation index, processing the cooling equipment data to obtain a cooling equipment reliability evaluation value, and comprehensively analyzing in combination with the cooling demand evaluation index to obtain a cooling system efficiency evaluation value, and matching the cooling medium flow margin demand value according to the cooling system efficiency evaluation value.

[0043] Correcting the prediction result of the energy-carbon prediction hybrid model according to the initial cooling medium flow margin and the cooling medium flow margin demand value.

[0044] Specifically, processing the basic load data of the data center to obtain a cooling demand evaluation index, and the specific processing process is as follows: the basic load data includes the floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate.

[0045] Extracting the critical floor area of the data center, the critical IT equipment density, the critical IT equipment operating temperature, and the critical CPU utilization rate from the cooling system database, and comprehensively analyzing to obtain a cooling demand evaluation index.

[0046] In a specific embodiment, the floor area of the data center refers to the ground area occupied by the data center building, which can be directly obtained by referring to the architectural design drawings of the data center; the IT equipment density refers to the number of IT equipment that can be accommodated per square meter, which can be calculated by counting the number of equipment in the data center and then dividing it by the floor area of the data center; the historical average operating temperature of IT equipment refers to the average operating temperature of IT equipment in the data center during the monitoring period, which can be measured by temperature sensors; the historical average CPU utilization rate refers to the average value of the CPU utilization rate of servers of IT equipment in the data center during the monitoring period, which can be obtained by using a performance monitoring tool to get the CPU utilization rate and then calculating it.

[0047] Further, the cooling demand assessment index, with the specific numerical expression as:

[0048]

[0049] Wherein, represents the cooling demand assessment index, represents the floor area of the data center, represents the critical floor area of the data center, represents the IT equipment density, represents the critical IT equipment density, represents the historical average operating temperature of IT equipment, represents the critical operating temperature of IT equipment, represents the historical average CPU utilization rate, represents the critical CPU utilization rate, represents the cooling demand assessment impact factor corresponding to the set floor area of the data center, represents the cooling demand assessment impact factor corresponding to the set IT equipment density, represents the cooling demand assessment impact factor corresponding to the set operating temperature of IT equipment, represents the cooling demand assessment impact factor corresponding to the set CPU utilization rate.

[0050] The algorithm of this embodiment combines the floor area of the data center, the IT equipment density, the historical average operating temperature of IT equipment, and the historical average CPU utilization rate, and comprehensively analyzes to obtain the cooling demand assessment index. The size of the floor area directly affects the deployment space of IT equipment, and thus affects the equipment density; the deployment of IT equipment with a higher density will cause the internal temperature of the data center to rise. If the cooling system cannot meet the heat dissipation requirements, it may affect the performance and reliability of the equipment; when the CPU utilization rate is higher, more heat will be generated, resulting in an increase in the operating temperature. If the heat dissipation system cannot discharge the heat in time, it may lead to a decrease in CPU performance. Through comprehensive analysis, a more comprehensive cooling demand assessment index can be obtained.

[0051] It should be noted that in this embodiment, four key factors are considered, namely the floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate. This helps to more accurately evaluate the workload and performance requirements of the equipment, thereby enabling more reasonable resource allocation and scheduling. It can optimize the cooling system of the data center, reduce energy consumption and carbon emissions, achieve green operation, promptly detect potential overheating problems, take preventive measures to avoid equipment failures and outages, and ensure that the data center can still operate stably in the face of various challenges, guaranteeing the continuity and availability of the business. By standardizing the floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate, it is ensured that they are compared on the same scale, improving the fairness and comparability of the evaluation. By weighting the impacts of the floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate, their relative importance in the evaluation index is reflected. The weights of different factors can be adjusted according to different requirements, making the formula highly adaptable. It is not difficult to see that the larger the floor area of the data center, or the IT equipment density, or the historical average operating temperature of the IT equipment, or the historical average CPU utilization rate, the larger the cooling demand evaluation index. By evaluating the cooling demand evaluation index, the cooling demand of the data center can be accurately understood, thereby taking targeted cooling measures to avoid overcooling or undercooling situations, improve energy efficiency, help optimize the airflow organization, reduce waste of cooling energy, further improve energy efficiency, reasonably allocate cooling resources, ensure that they operate in a highly efficient state, avoid resource waste, further enhance equipment reliability, and ensure the stability and reliability of the cooling system of the data center.

[0052] In a specific embodiment, the value ranges of the cooling demand assessment impact factors corresponding to the floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate are between 0 and 1, which represent the numerical values of the influence degrees of the floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate on the cooling demand assessment index. Each cooling demand assessment impact factor can be obtained from the cooling system database. By adjusting the values of the impact factors, the influence degrees of different factors on the final cooling demand assessment index can be flexibly adjusted. The acquisition method can be a pre-set mapping relationship. For example, a mapping set is formed by the floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate and the corresponding weight factors of the pre-set floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate in the cooling system database. The real-time floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate are brought into the mapping set to obtain the corresponding weight factors of the floor area of the data center, the IT equipment density, the historical average operating temperature of the IT equipment, and the historical average CPU utilization rate. The mapping relationship therein can be a one-to-one or many-to-one relationship.

[0053] Furthermore, an initial cooling medium flow margin is matched according to the cooling demand assessment index and regional division is carried out. The specific analysis process is as follows: The cooling demand assessment index is input into the cooling system database to match the corresponding initial cooling medium flow margin. The floor area of the data center is input into the cooling system database to match the corresponding regional division size. The data center is divided into regions according to the regional division size to obtain each target region.

[0054] Specifically, the environmental condition data is processed to obtain the data center environmental stability assessment index. The specific processing process is as follows: The environmental condition data includes outdoor temperature, computer room dew point temperature, relative humidity, and dustiness.

[0055] The reference standard outdoor temperature, the allowable deviation of the outdoor temperature, the reference standard computer room dew point temperature, the allowable deviation of the computer room dew point temperature, the reference standard relative humidity, the allowable deviation of the relative humidity, and the critical dustiness are extracted from the cooling system database, and the data center environmental stability assessment index is obtained through comprehensive analysis.

[0056] In a specific embodiment, the outdoor temperature refers to the ambient temperature outside the data center building and can be measured by a thermometer; the dew point temperature in the computer room refers to the temperature at which the air in the computer room cools to saturation under the conditions that the water vapor content and air pressure remain unchanged and can be directly measured by a dew point thermometer; the relative humidity refers to the percentage of the water vapor pressure in the air to the saturated water vapor pressure and is an important indicator for evaluating the air humidity condition and can be measured by a hygrometer; the dustiness refers to the concentration of suspended particulate matter in the air and is an important indicator for evaluating the cleanliness of the data center and can be measured by a particle counter.

[0057] Furthermore, the evaluation index of the environmental stability of the data center, the specific numerical expression is:

[0058]

[0059] Wherein, represents the evaluation index of the environmental stability of the data center, e represents the natural constant, represents the outdoor temperature, represents the reference standard outdoor temperature, represents the allowable deviation of the outdoor temperature, represents the dew point temperature in the computer room, represents the reference standard dew point temperature in the computer room, represents the allowable deviation of the dew point temperature in the computer room, represents the relative humidity, represents the reference standard relative humidity, represents the allowable deviation of the relative humidity, represents the dustiness, represents the critical dustiness, represents the influence factor of the evaluation of the environmental stability of the data center corresponding to the set outdoor temperature, represents the influence factor of the evaluation of the environmental stability of the data center corresponding to the set dew point temperature in the computer room, represents the influence factor of the evaluation of the environmental stability of the data center corresponding to the set relative humidity, represents the influence factor of the evaluation of the environmental stability of the data center corresponding to the set dustiness.

[0060] The algorithm of this embodiment combines the outdoor temperature, the dew point temperature of the computer room, the relative humidity, and the dustiness, and comprehensively analyzes to obtain the data center environment stability evaluation index. The change of the outdoor temperature will directly affect the temperature of the computer room in the data center. The higher the outdoor high temperature, the more likely it is to cause the temperature of the computer room to rise, increasing the burden on the cooling system; both the dew point temperature and the relative humidity are methods to represent the water content in the air, and there is a close relationship between them. In the computer room environment, a reasonable dew point temperature can prevent condensation inside the equipment, avoiding affecting the normal operation of the equipment. At the same time, an appropriate relative humidity also helps to maintain the stability of the equipment and extend its service life; the relative humidity in the air will affect the suspension state of dust particles. When the relative humidity is higher, the water vapor in the air is easily condensed on the dust particles, increasing their weight, and thus it is easier to deposit on the ground or equipment, helping to reduce the dust concentration in the air; the presence of dust particles will also affect the relative humidity in the computer room. Dust particles can adsorb the water vapor in the air, thus changing the local humidity condition. Through comprehensive analysis, a more comprehensive data center environment stability evaluation index can be obtained.

[0061] It should be explained that in this embodiment, four key factors are considered, namely the outdoor temperature, the dew point temperature of the computer room, the relative humidity, and the dustiness, which can prevent equipment failures caused by too high or too low temperatures, prevent condensation in the computer room, avoid equipment from getting damp and corroded, thereby improving the stability and reliability of the equipment, and can also keep the internal electronic components of the equipment in a normal working state, improve the heat dissipation efficiency of the equipment, help reduce the cooling energy consumption of the data center, and further reduce the equipment failure rate caused by environmental factors. By standardizing the outdoor temperature, the dew point temperature of the computer room, the relative humidity, and the dustiness, it is ensured that they are compared on the same magnitude, improving the fairness and comparability of the evaluation. By weighting the influences of the outdoor temperature, the dew point temperature of the computer room, the relative humidity, and the dustiness, their relative importance in the evaluation index is reflected, and the weights of different factors can be adjusted according to different needs, making the formula have good adaptability. It is not difficult to see that the smaller the deviation of the outdoor temperature or the dew point temperature of the computer room or the relative humidity deviation or the dustiness, the larger the data center environment stability evaluation index. By evaluating the data center environment stability evaluation index, it helps the data center operation and maintenance team to quickly take measures to avoid the problem from deteriorating further, thereby improving the operation efficiency of the data center, can reflect the overall condition of the data center environment, help the operation and maintenance team to reasonably allocate resources, can timely detect and correct the environmental factors that are not conducive to the stable operation of the equipment, help to maintain the best performance state of the equipment, improve the efficiency and accuracy of data processing and analysis, and can also optimize the energy use strategy, such as adjusting the operating parameters of the cooling system, which can reduce the energy consumption of the data center.

[0062] In a specific embodiment, the value ranges of the influencing factors for the evaluation of the data center environmental stability corresponding to the outdoor temperature, the computer room dew point temperature, the relative humidity, and the dustiness are between 0 and 1, which represent the numerical values of the influence degrees of the outdoor temperature, the computer room dew point temperature, the relative humidity, and the dustiness on the data center environmental stability evaluation index. Each influencing factor for the data center environmental stability evaluation can be obtained from the cooling system database. By adjusting the values of the influencing factors, the influence degrees of different factors on the final data center environmental stability evaluation index can be flexibly adjusted. The corresponding relationship can be a pre-set mapping relationship. For example, the outdoor temperature, the computer room dew point temperature, the relative humidity, and the dustiness form a mapping set with the weight factors corresponding to the preset outdoor temperature, computer room dew point temperature, relative humidity, and dustiness in the cooling system database. The real-time outdoor temperature, computer room dew point temperature, relative humidity, and dustiness are brought into the mapping set to obtain the weight factors corresponding to the outdoor temperature, computer room dew point temperature, relative humidity, and dustiness. The mapping relationship therein can be a one-to-one or many-to-one relationship.

[0063] Specifically, the reliability evaluation value of the cooling equipment is obtained by processing the cooling equipment data. The specific processing process is as follows: The cooling equipment data includes the cooling capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipe resistance coefficient.

[0064] The reference standard cooling capacity of the cooling equipment, the allowable deviation of the cooling capacity of the cooling equipment, the reference standard flow pressure of the cooling medium, the allowable deviation of the flow pressure of the cooling medium, the reference standard number of redundant cooling equipment, the allowable deviation of the number of redundant cooling equipment, and the critical pipe resistance coefficient are extracted from the cooling system database, and the reliability evaluation value of the cooling equipment is obtained through comprehensive analysis.

[0065] In a specific embodiment, the cooling capacity refers to the heat removed from the data center environment by the cooling equipment per minute, which can be calculated by a heat load calculation tool; the flow pressure of the cooling medium refers to the pressure borne by the cooling medium when flowing in the cooling system, which can be measured by a pressure sensor; the number of redundant cooling equipment refers to the number of additional cooling equipment installed to ensure the normal operation of the data center in case of a cooling equipment failure, which can be obtained by statistics; the pipe resistance coefficient refers to the ratio of the resistance suffered by the cooling medium when flowing in the pipe to the flow velocity, which can be obtained by a pipe design tool.

[0066] Further, the reliability evaluation value of the cooling equipment, the specific numerical expression is:

[0067]

[0068] Wherein, represents the reliability evaluation value of the cooling equipment, e represents the natural constant, represents the cooling capacity of the cooling equipment, Indicates the refrigerating capacity of the reference standard cooling equipment, Indicates the allowable deviation of the refrigerating capacity of the cooling equipment, Indicates the flow pressure of the cooling medium, Indicates the flow pressure of the reference standard cooling medium, Indicates the allowable deviation of the flow pressure of the cooling medium, Indicates the number of redundant cooling equipment, Indicates the number of redundant cooling equipment of the reference standard, Indicates the allowable deviation of the number of redundant cooling equipment, Indicates the pipe resistance coefficient, Indicates the critical pipe resistance coefficient, Indicates the influencing factor of the reliability evaluation of the cooling equipment corresponding to the set refrigerating capacity of the cooling equipment, Indicates the influencing factor of the reliability evaluation of the cooling equipment corresponding to the set flow pressure of the cooling medium, Indicates the influencing factor of the reliability evaluation of the cooling equipment corresponding to the set number of redundant cooling equipment, Indicates the influencing factor of the reliability evaluation of the cooling equipment corresponding to the set pipe resistance coefficient.

[0069] The algorithm of this embodiment combines the refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipe resistance coefficient, and comprehensively analyzes to obtain the reliability evaluation value of the cooling equipment. The refrigerating capacity of the cooling equipment is usually related to the flow pressure of its cooling medium. In most cases, as the flow pressure of the cooling medium increases, the refrigerating capacity of the cooling equipment will also increase accordingly; when the refrigeration demand of the data center increases, it may be necessary to increase the number of redundant cooling equipment to ensure sufficient refrigerating capacity. At the same time, the existence of redundant cooling equipment can also improve the reliability and stability of the data center; the pipe resistance coefficient is one of the important factors affecting the flow pressure of the cooling medium. When the pipe resistance coefficient is larger, the cooling medium will be subject to greater resistance when flowing in the pipe, resulting in an increase in the flow pressure. Through comprehensive analysis, a more comprehensive reliability evaluation value of the cooling equipment can be obtained.

[0070] It should be noted that in this embodiment, four key factors are considered, namely the refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient. This helps to rationally allocate the refrigerating capacity of the cooling equipment, ensuring that the data center can maintain a suitable temperature while avoiding energy waste caused by excessive cooling. It can ensure the efficient operation of the cooling system, reduce the increase in energy consumption and equipment wear caused by too high or too low pressure, and can also optimize the pipeline design, reduce the pipeline resistance coefficient, and reduce the flow resistance of the cooling medium in the pipeline, further improving the cooling efficiency. By standardizing the refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient, it is ensured that they are compared on the same order of magnitude, improving the fairness and comparability of the evaluation. By weighting the influences of the refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient, the relative importance of them in the evaluation index is reflected. The weights of different factors can be adjusted according to different needs, making the formula highly adaptable. It is not difficult to see that when the deviation of the refrigerating capacity of the cooling equipment, the deviation of the flow pressure of the cooling medium, the deviation of the number of redundant cooling equipment, or the pipeline resistance coefficient is smaller, the reliability evaluation value of the cooling equipment is larger. By evaluating the reliability evaluation value of the cooling equipment, it can quickly take over the work when the main cooling equipment fails, ensuring the continuous cooling of the data center, improving the stability and reliability of the system, helping to achieve comprehensive monitoring and dynamic adjustment of the cooling system, promptly discovering and solving problems, preventing system crashes caused by abnormalities in a single factor, being able to utilize energy more efficiently, and reducing unnecessary energy consumption and carbon emissions.

[0071] In a specific embodiment, the value ranges of the influencing factors of the reliability evaluation of the cooling equipment corresponding to the refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient are between 0 and 1, which represent the numerical values of the influence degrees of the refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient on the reliability evaluation value of the cooling equipment. Each influencing factor of the reliability evaluation of the cooling equipment can be obtained from the cooling system database. By adjusting the values of the influencing factors, the influence degrees of different factors on the final reliability evaluation value of the cooling equipment can be flexibly adjusted. The acquisition method can be a pre-set mapping relationship. For example, the refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient form a mapping set with the weight factors corresponding to the pre-set refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient in the cooling system database. Substituting the real-time refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient into the mapping set to obtain the weight factors corresponding to the refrigerating capacity of the cooling equipment, the flow pressure of the cooling medium, the number of redundant cooling equipment, and the pipeline resistance coefficient. The mapping relationship therein can be a one-to-one or many-to-one relationship.

[0072] Specifically, the cooling system efficiency evaluation value is obtained through comprehensive analysis. The specific analysis process is as follows: Monitor the temperatures of each target area to obtain the average monitored temperature of each target area. Extract the area temperature threshold from the cooling system database. Compare the average monitored temperature of each target area with the area temperature threshold, count the target areas where the average monitored temperature is greater than or equal to the area temperature threshold, mark them as temperature anomaly areas, and obtain the number of temperature anomaly areas.

[0073] Subtract the area temperature threshold from the average monitored temperature of each temperature anomaly area and take the absolute value to obtain the temperature anomaly amplitude of each temperature anomaly area.

[0074] Based on the cooling demand evaluation index, the data center environment stability evaluation index, the cooling equipment reliability evaluation value, the temperature anomaly amplitude of each temperature anomaly area, and the number of temperature anomaly areas, the cooling system efficiency evaluation value is obtained through comprehensive analysis.

[0075] Furthermore, the cooling system efficiency evaluation value has the following specific numerical expression:

[0076]

[0077] Where, represents the cooling system efficiency evaluation value, represents the cooling demand evaluation index, represents the data center environment stability evaluation index, represents the cooling equipment reliability evaluation value, represents the number of temperature anomaly areas, represents the number of critical temperature anomaly areas, represents the temperature anomaly amplitude of the x-th temperature anomaly area, represents the critical temperature anomaly amplitude, represents the cooling system efficiency evaluation impact factor corresponding to the set cooling demand evaluation index, represents the cooling system efficiency evaluation impact factor corresponding to the set data center environment stability evaluation index, represents the cooling system efficiency evaluation impact factor corresponding to the set cooling equipment reliability evaluation value, represents the cooling system efficiency evaluation impact factor corresponding to the set number of temperature anomaly areas, represents the cooling system efficiency evaluation impact factor corresponding to the set temperature anomaly amplitude, x represents the number of each temperature anomaly area, x = 1, 2, 3,..., y, and y represents the total number of temperature anomaly areas.

[0078] As Figure 3 shown, in a specific embodiment, = = = 0.4, = = 0.3, = 0.6, = 1 piece, = 5 pieces, = 4 °C, = 10 °C, y = 1, when = 0.1, the functional relationship between the cooling system efficiency evaluation value and the cooling demand evaluation index is as shown by curve a; when = 0.5, the functional relationship between the cooling system efficiency evaluation value and the cooling demand evaluation index is as shown by curve b; when = 1, the functional relationship between the cooling system efficiency evaluation value and the cooling demand evaluation index is as shown by curve c.

[0079] The algorithm of this embodiment combines the cooling demand evaluation index, the data center environment stability evaluation index, the cooling equipment reliability evaluation value, the temperature anomaly amplitude of each temperature anomaly area, and the number of temperature anomaly areas, and comprehensively analyzes to obtain the cooling system efficiency evaluation value. The cooling demand evaluation index reflects the load condition of the cooling equipment. If the cooling demand is greater, the cooling equipment may be in a high-load operation state for a long time, which will increase the wear and failure risk of the equipment and reduce the reliability evaluation value of the cooling equipment; the data center environment stability evaluation index is also related to the number of temperature anomaly areas and the temperature anomaly amplitude. If there are more temperature anomaly areas in the data center, or the temperature anomaly amplitude is greater, then the environmental stability of the data center will also be correspondingly reduced; the reliability of the cooling equipment also directly affects the temperature control of the data center. If the cooling equipment fails or its performance deteriorates, it may lead to temperature anomaly areas in the data center; the existence of temperature anomaly areas may also reflect the performance problems of the cooling equipment. If the cooling equipment cannot effectively control the temperature of the data center, temperature anomaly areas may occur. Comprehensive analysis can obtain a more comprehensive cooling system efficiency evaluation value.

[0080] It should be noted that in this embodiment, five key factors are considered, namely the cooling demand assessment index, the data center environment stability assessment index, the cooling equipment reliability assessment value, the temperature anomaly amplitude of each temperature anomaly area, and the number of temperature anomaly areas, to ensure that while the data center maintains a suitable temperature, overcooling or undercooling is avoided, thereby improving the operating efficiency of the cooling system, enabling timely detection and resolution of temperature control problems, reducing the impact of temperature anomalies on the operation of the data center, maintaining the stability of the data center environment, and ensuring that the data center environment is always in the best state, thereby improving data processing capabilities, helping to optimize the resource allocation and layout of the data center, and improving the utilization efficiency and overall performance of resources. By standardizing the temperature anomaly amplitude and the number of temperature anomaly areas of each temperature anomaly area, it is ensured that they are compared on the same scale, improving the fairness and comparability of the evaluation. By weighting the impacts of the cooling demand assessment index, the data center environment stability assessment index, the cooling equipment reliability assessment value, the temperature anomaly amplitude of each temperature anomaly area, and the number of temperature anomaly areas, their relative importance in the evaluation index is reflected, and the weights of different factors can be adjusted according to different requirements, making the formula highly adaptable. It is not difficult to see that when the cooling demand assessment index is larger, or the data center environment stability assessment index is larger, or the cooling equipment reliability assessment value is larger, or the temperature anomaly amplitude of each temperature anomaly area is smaller, or the number of temperature anomaly areas is smaller, the cooling system efficiency assessment value is larger. By evaluating the cooling system efficiency assessment value, the actual performance of the cooling system can be accurately understood, so as to make a precise match according to actual needs, helping to avoid overdesign or underdesign, ensuring that the cooling system can meet the cooling needs of equipment or the environment without causing energy waste, and also helping to identify the energy efficiency bottlenecks in the system, and then taking targeted optimization measures, which can improve the energy efficiency ratio of the system, identify the energy-saving potential in the system, and take corresponding improvement measures, thereby achieving the goal of energy conservation and emission reduction, and further significantly reducing energy consumption and carbon emissions.

[0081] In a specific embodiment, the value range of the cooling demand assessment index, the data center environment stability assessment index, the cooling equipment reliability assessment value, the temperature anomaly amplitude, and the number of temperature anomaly regions corresponding to the cooling system efficiency assessment influencing factor is between 0 and 1, which represents the numerical value of the influence degree of the cooling demand assessment index, the data center environment stability assessment index, the cooling equipment reliability assessment value, the temperature anomaly amplitude, and the number of temperature anomaly regions on the cooling system efficiency assessment value. Each cooling system efficiency assessment influencing factor can be obtained from the cooling system database. By adjusting the value of the influencing factor, the influence degree of different factors on the final cooling system efficiency assessment value can be flexibly adjusted. The acquisition method can be a pre-set mapping relationship. For example, the cooling demand assessment index, the data center environment stability assessment index, the cooling equipment reliability assessment value, the temperature anomaly amplitude, and the number of temperature anomaly regions form a mapping set with the weight factors corresponding to the pre-set cooling demand assessment index, the data center environment stability assessment index, the cooling equipment reliability assessment value, the temperature anomaly amplitude, and the number of temperature anomaly regions in the cooling system database. The real-time cooling demand assessment index, the data center environment stability assessment index, the cooling equipment reliability assessment value, the temperature anomaly amplitude, and the number of temperature anomaly regions are brought into the mapping set to obtain the weight factors corresponding to the cooling demand assessment index, the data center environment stability assessment index, the cooling equipment reliability assessment value, the temperature anomaly amplitude, and the number of temperature anomaly regions. The mapping relationship therein can be a one-to-one or many-to-one relationship.

[0082] Further, the cooling medium flow margin demand value is matched according to the cooling system efficiency assessment value. The specific process is as follows: The cooling system efficiency assessment value is input into the cooling system database to match the cooling medium flow margin demand value corresponding to each cooling system efficiency assessment value interval. The cooling system efficiency assessment value is used to quantify the efficiency of the cooling system.

[0083] Specifically, the prediction result of the energy-carbon prediction hybrid model is corrected according to the initial cooling medium flow margin and the cooling medium flow margin demand value. The specific process is as follows: The initial cooling medium flow margin and the cooling medium flow margin demand value are compared. If the initial cooling medium flow margin is greater than or equal to the cooling medium flow margin demand value, no additional operation is performed. If the initial cooling medium flow margin is less than the cooling medium flow margin demand value, the initial cooling medium flow margin is increased until it reaches the cooling medium flow margin demand.

[0084] The initial cooling medium flow margin and the cooling medium flow margin requirement value are subtracted to obtain the cooling medium flow margin difference, the cooling medium flow margin difference threshold is extracted from the cooling system database, the cooling medium flow margin difference is compared with the cooling medium flow margin difference threshold, if the cooling medium flow margin difference is greater than or equal to the cooling medium flow margin difference threshold, the cooling medium flow margin difference is input into the cooling system database to match the corresponding energy-carbon prediction correction value, the energy-carbon prediction hybrid model result is corrected according to the energy-carbon prediction correction value, that is, the energy-carbon prediction correction value and the energy-carbon prediction hybrid model result are added to obtain an updated energy-carbon prediction hybrid model result, if the cooling medium flow margin difference is less than the cooling medium flow margin difference threshold, no additional operation is performed.

[0085] Furthermore, the hybrid model for energy-carbon prediction specifically includes: firstly, medium- and long-term trend prediction is performed using the ARIMA time series model to characterize the periodic laws and development trends of energy consumption and carbon emissions, then short-term fine prediction is performed using the DNN+LSTM machine learning model to mine the nonlinear relationship between basic load data, environmental condition data, cooling equipment data and energy consumption and carbon emissions, and finally, multi-model fusion is performed, that is, the medium- and long-term prediction results of the time series model are superimposed and combined with the short-term prediction results of the machine learning model to obtain the final energy-carbon prediction value. The input of the DNN+LSTM machine learning model is the basic load data, environmental condition data and cooling equipment data, and the output is the energy consumption prediction value and the carbon emission prediction value. The model mainly includes the Embedding layer, the DNN layer and the LSTM layer. The Embedding layer is used to map the basic load data, environmental condition data and cooling equipment data into dense vectors, the DNN layer is used to extract high-order nonlinear combinations of feature vectors, and the LSTM layer is used to learn the long- and short-term dependencies of the time series. The model training process is to first build a multi-dimensional monitoring data set of basic load data, environmental condition data and cooling equipment data. Each record corresponds to a sampling time point, which is sliced ​​according to a sliding window (such as 24 hours) to generate time series samples, embedding discrete features, normalizing continuous features, and then splicing them. A DNN+LSTM model is built using frameworks such as Pytorch and Tensorflow, the RMSE loss function is defined, and the Adam optimizer is used for training until convergence. The performance is evaluated on the test set for online prediction.

[0086] Reference Figure 2 As shown, the second aspect of the present invention provides a data processing system for a data center energy-carbon prediction hybrid model, including: a data acquisition module for collecting basic load data, environmental condition data and cooling equipment data of the data center.

[0087] The region division module is used to process the basic load data of the data center to obtain a cooling demand evaluation index, match the initial cooling medium flow margin according to the cooling demand evaluation index and conduct region division to obtain each target region.

[0088] The cooling system efficiency analysis module is used to process the environmental condition data to obtain a data center environmental stability evaluation index, process the cooling equipment data to obtain a cooling equipment reliability evaluation value, combine with the cooling demand evaluation index, comprehensively analyze to obtain a cooling system efficiency evaluation value, and match the cooling medium flow margin demand value according to the cooling system efficiency evaluation value.

[0089] The energy-carbon prediction hybrid model correction module is used to correct the prediction result of the energy-carbon prediction hybrid model according to the initial cooling medium flow margin and the cooling medium flow margin demand value.

[0090] The cooling system database is used to store cooling system-related data, including: critical data center floor area, critical IT equipment density, critical IT equipment operating temperature, critical CPU utilization rate, critical outdoor temperature, reference standard computer room dew point temperature, allowable deviation of computer room dew point temperature, reference standard relative humidity, allowable deviation of relative humidity, critical dustiness, reference standard cooling equipment refrigerating capacity, allowable deviation of cooling equipment refrigerating capacity, reference standard cooling medium flow pressure, allowable deviation of cooling medium flow pressure, reference standard redundant cooling equipment quantity, allowable deviation of redundant cooling equipment quantity, critical pipeline resistance coefficient, cooling system efficiency evaluation influencing factor corresponding to the set cooling demand evaluation index, cooling system efficiency evaluation influencing factor corresponding to the set data center environmental stability evaluation index, cooling system efficiency evaluation influencing factor corresponding to the set cooling equipment reliability evaluation value, cooling system efficiency evaluation influencing factor corresponding to the set number of temperature anomaly regions, cooling system efficiency evaluation influencing factor corresponding to the set temperature anomaly amplitude, and cooling medium flow margin difference threshold and other indicators.

[0091] The above content is only an example and explanation of the structure of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they shall fall within the protection scope of the present invention.

Claims

1. A data processing method for a hybrid model of data center energy and carbon prediction, characterized in that: include: Collect basic load data, environmental condition data and cooling equipment data of the data center; The basic load data of the data center is processed to obtain a cooling demand assessment index, and the initial cooling medium flow margin is obtained according to the cooling demand assessment index and divided into regions to obtain each target region; The environmental condition data is processed to obtain the data center environmental stability evaluation index, and the cooling equipment data is processed to obtain the cooling equipment reliability evaluation value. Combined with the cooling demand evaluation index, a comprehensive analysis is performed to obtain the cooling system efficiency evaluation value. According to the cooling system efficiency evaluation value, the cooling medium flow margin demand value is obtained. The prediction results of the energy-carbon prediction hybrid model are corrected according to the initial cooling medium flow margin and the cooling medium flow margin requirement value; The prediction result of the energy-carbon prediction hybrid model is corrected according to the initial cooling medium flow margin and the cooling medium flow margin requirement value. The specific process is as follows: The initial cooling medium flow margin is compared with the cooling medium flow margin requirement value. If the initial cooling medium flow margin is greater than or equal to the cooling medium flow margin requirement value, no additional operation is performed. If the initial cooling medium flow margin is less than the cooling medium flow margin requirement value, the initial cooling medium flow margin is increased until the cooling medium flow margin requirement value is reached. The initial cooling medium flow margin and the cooling medium flow margin requirement value are subtracted to obtain the cooling medium flow margin difference, the cooling medium flow margin difference threshold is extracted from the cooling system database, the cooling medium flow margin difference is compared with the cooling medium flow margin difference threshold, if the cooling medium flow margin difference is greater than or equal to the cooling medium flow margin difference threshold, the cooling medium flow margin difference is input into the cooling system database to match the corresponding energy-carbon prediction correction value, the energy-carbon prediction hybrid model result is corrected according to the energy-carbon prediction correction value, if the cooling medium flow margin difference is less than the cooling medium flow margin difference threshold, no additional operation is performed.

2. The data processing method of a data center energy carbon prediction hybrid model according to claim 1 is characterized by: The basic load data of the data center is processed to obtain the cooling demand evaluation index, and the specific processing process is as follows: The basic load data includes data center floor space, IT equipment density, historical average operating temperature of IT equipment, and historical average CPU utilization; The critical data center floor space, critical IT equipment density, critical IT equipment operating temperature and critical CPU utilization are extracted from the cooling system database, and a cooling demand assessment index is obtained through comprehensive analysis.

3. The data processing method of a data center energy carbon prediction hybrid model according to claim 2 is characterized by: The initial cooling medium flow margin is obtained according to the cooling demand assessment index matching and regional division is performed. The specific analysis process is as follows: The cooling demand assessment index is input into the cooling system database to match the corresponding initial cooling medium flow margin, the data center footprint is input into the cooling system database to match the corresponding regional division size, and the data center is divided into regions according to the regional division size to obtain each target regions.

4. The data processing method of a data center energy carbon prediction hybrid model according to claim 1 is characterized by: The environmental condition data is processed to obtain the data center environmental stability evaluation index, and the specific processing process is: The environmental condition data includes outdoor temperature, dew point temperature of the equipment room, relative humidity and dust level; The reference standard outdoor temperature, allowable deviation outdoor temperature, reference standard computer room dew point temperature, allowable deviation computer room dew point temperature, reference standard relative humidity, allowable deviation relative humidity and critical dust degree are extracted from the cooling system database, and a comprehensive analysis is performed to obtain the data center environmental stability assessment index.

5. The data processing method of a data center energy carbon prediction hybrid model according to claim 1 is characterized by: The cooling equipment data is processed to obtain the cooling equipment reliability evaluation value, and the specific processing process is as follows: The cooling equipment data includes cooling capacity of the cooling equipment, flow pressure of the cooling medium, number of redundant cooling equipment and pipeline resistance coefficient; The reference standard cooling equipment cooling capacity, allowable deviation cooling equipment cooling capacity, reference standard cooling medium flow pressure, allowable deviation cooling medium flow pressure, reference standard redundant cooling equipment quantity, allowable deviation redundant cooling equipment quantity and critical pipe resistance coefficient are extracted from the cooling system database, and the reliability assessment value of the cooling equipment is obtained through comprehensive analysis.

6. The data processing method of a data center energy carbon prediction hybrid model according to claim 5 is characterized by: The comprehensive analysis obtains the cooling system efficiency evaluation value, and the specific analysis process is as follows: Monitor the temperature of each target area to obtain the average monitored temperature of each target area, extract the regional temperature threshold from the cooling system database, compare the average monitored temperature of each target area with the regional temperature threshold, count the target areas whose average monitored temperature is greater than or equal to the regional temperature threshold, mark them as temperature abnormal areas, and obtain the number of temperature abnormal areas; The average monitored temperature of each temperature abnormal area is subtracted from the regional temperature threshold and the absolute value is taken to obtain the temperature abnormal amplitude of each temperature abnormal area; Based on the cooling demand assessment index, data center environmental stability assessment index, cooling equipment reliability assessment value, temperature anomaly amplitude of each temperature anomaly area and the number of temperature anomaly areas, a comprehensive analysis is conducted to obtain the cooling system efficiency assessment value.

7. The data processing method of a hybrid model for predicting energy and carbon emissions in a data center according to claim 6, characterized in that: The cooling medium flow margin requirement value is obtained by matching the cooling system efficiency evaluation value, and the specific process is as follows: The cooling system efficiency evaluation value is input into the cooling system database to match the cooling medium flow margin requirement value corresponding to each cooling system efficiency evaluation value interval. The cooling system efficiency evaluation value is used to quantify the efficiency of the cooling system.

8. The data processing method of a data center energy carbon prediction hybrid model according to claim 6 is characterized by: The cooling system efficiency evaluation value is specifically expressed as: in, represents the cooling system efficiency evaluation value, represents the cooling demand assessment index, represents the data center environmental stability assessment index, represents the reliability evaluation value of the cooling equipment, Indicates the number of temperature anomaly areas. Indicates the number of critical temperature anomaly areas, represents the temperature anomaly amplitude of the xth temperature anomaly area, represents the critical temperature anomaly amplitude, Indicates the cooling system efficiency assessment impact factor corresponding to the set cooling demand assessment index, Indicates the impact factor of the cooling system efficiency assessment corresponding to the set data center environmental stability assessment. Indicates the cooling system efficiency assessment impact factor corresponding to the set cooling equipment reliability assessment value, Indicates the cooling system efficiency evaluation impact factor corresponding to the set number of temperature abnormal areas, It represents the cooling system efficiency evaluation influencing factor corresponding to the set temperature anomaly amplitude, x represents the number of each temperature anomaly area, x=1,2,3,...,y, y represents the total number of temperature anomaly areas.

9. A data processing system for a data center energy-carbon forecasting hybrid model, using the data processing method for a data center energy-carbon forecasting hybrid model as claimed in any one of claims 1 to 8, characterized in that: include: Data collection module, used to collect basic load data, environmental condition data and cooling equipment data of the data center; The regional division module is used to process the basic load data of the data center to obtain the cooling demand evaluation index, match the cooling demand evaluation index to obtain the initial cooling medium flow margin and perform regional division to obtain each target area; The cooling system efficiency analysis module is used to process the environmental condition data to obtain the data center environmental stability evaluation index, process the cooling equipment data to obtain the cooling equipment reliability evaluation value, combine the cooling demand evaluation index, and comprehensively analyze the cooling system efficiency evaluation value. According to the cooling system efficiency evaluation value, the cooling medium flow margin demand value is matched; The energy-carbon prediction hybrid model correction module is used to correct the prediction results of the energy-carbon prediction hybrid model according to the initial cooling medium flow margin and the cooling medium flow margin requirement value.

Images