AI optimization control method and system for efficient machine room of central air conditioner

By real-time monitoring and prediction of the room temperature, adjusting the air conditioning control strategy, and setting the buffer time threshold, the hysteresis and fluctuations of the temperature control of the traditional room is solved, and more efficient and accurate temperature management is achieved, reducing operation and maintenance costs.

CN119063188AInactive Publication Date: 2024-12-03GUANGZHOU SJ ENERGY SAVING TECH CO LTD

Patent Information

Application Number
CN202411562263.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-05
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional computer room temperature control relies on manual experience and simple temperature sensors, making it difficult to respond to temperature changes accurately in real time, resulting in excessive temperature fluctuations and lack of predictive capabilities for future temperature changes, resulting in temperature control hysteresis.

Method used

Adopt AI optimization control method for centralized air conditioning high-efficiency computer room, by monitoring temperature parameters and equipment load information in real time, establishing a temperature prediction model, predicting future temperature changes, and adjusting the air conditioning control strategy based on the prediction results, setting the buffer time threshold for alarm and optimization, regularly Generate system operation report and analyze adjustment strategies.

Benefits of technology

It improves the efficiency of the air conditioning system, reduces operating costs, enhances the accuracy and prediction capabilities of temperature control, reduces operation and maintenance interference, promptly detects potential problems, optimizes temperature adjustment strategies, and avoids equipment overheating.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119063188A_ABST
    Figure CN119063188A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of central air conditioner control, in particular to a central air conditioner efficient machine room AI optimization control method and system. Temperature parameters inside and outside the efficient machine room are monitored in real time and stored in a database; the method comprises the steps of acquiring temperature parameters inside and outside a machine room in real time, acquiring equipment load information, extracting a characteristic variable rule according to the temperature parameters outside the machine room and the equipment load information in different time periods, predicting the temperature in the machine room in different time periods in the future, and adjusting an air conditioner control strategy according to a prediction result. Characteristic variable laws are extracted, complex temperature change conditions are converted into specific analyzable data through the prediction model, and the characteristic variables can capture the laws such as the trend, periodicity and seasonality of temperature change, so that the temperature prediction model can more accurately predict the temperatures in the machine room in different time periods in the future; reliable reference is provided for adjusting an air conditioner control strategy in advance, the efficiency of an air conditioner system is improved, and the operation cost is reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of centralized air-conditioning control, and in particular to an AI optimization control method and system for a centralized air-conditioning high-efficiency computer room. Background Art

[0002] With the rapid development of information technology, various computer rooms play a vital role in modern society. Whether it is a data center computer room, a communication computer room or an enterprise computer room, a stable temperature environment needs to be maintained to ensure the normal operation and service life of the equipment.

[0003] In traditional computer room temperature control, the air conditioning system is often adjusted based on manual experience and simple temperature sensor feedback. However, manual adjustment is difficult to accurately respond to temperature changes in real time, and it is easy to cause excessive temperature fluctuations, posing a threat to the stability of the equipment. For example, when the equipment load in the computer room suddenly increases, the temperature may rise rapidly. If the air conditioning cooling power cannot be adjusted in time, it may cause the equipment to overheat, reduce performance or even damage the equipment. On the other hand, traditional temperature control methods lack the ability to predict future temperature changes and cannot prepare in advance. They often make passive adjustments after the temperature has changed, resulting in a lag in temperature control.

[0004] Therefore, an AI optimization control method and system for a centralized air-conditioning efficient computer room is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide an AI optimization control method and system for a centralized air-conditioning high-efficiency computer room to solve the problems raised in the above-mentioned background technology.

[0006] In order to solve the above technical problems, one of the purposes of the present invention is to provide an AI optimization control method for a centralized air-conditioning high-efficiency computer room, comprising the following steps: S1. Real-time monitoring of temperature parameters inside and outside the high-efficiency computer room and storage in the database; S2. Obtain equipment load information, extract characteristic variable rules based on the external temperature parameters of the computer room and equipment load information at different time periods, predict the temperature inside the computer room at different time periods in the future, and adjust the air conditioning control strategy based on the prediction results; S3. Based on the adjustment result, the internal temperature of the computer room is monitored and a buffer time threshold is set. If the temperature parameter of the computer room within the buffer time threshold is not within the set standard range, an alarm is issued and optimization is performed. If the environmental parameter within the buffer time threshold is within the set standard range, it indicates that the temperature adjustment in the computer room is normal. S4. Generate system operation reports regularly, analyze the effects of the computer room temperature adjustment strategies in the system operation reports, and optimize the temperature adjustment strategies.

[0007] As a further improvement of the technical solution, in S1, a temperature sensor is used to collect the internal and external temperatures of the computer room, and the temperature parameters collected by the sensor are regularly stored in a database.

[0008] As a further improvement of the technical solution, the analysis of characteristic variable rules of temperature parameters of the computer room in different time periods in S2 includes the following steps: S2.1. Obtain the load information of the equipment in the computer room at different time periods by connecting to the monitoring system of the equipment in the computer room; S2.2, extract characteristic variables related to the temperature inside the computer room by combining the external temperature parameters of the computer room and the equipment load information; S2.3. Establish a temperature prediction model for the computer room, predict the temperature in the computer room at different time periods in the future, and adjust the air conditioning control strategy based on the prediction results.

[0009] As a further improvement of the technical solution, the specific steps of establishing the prediction model in S2.3 are: Data collection and organization: Obtain historical room temperature data from the database, including temperature records at different time points, and pre-process the data to remove outliers and noise; Feature engineering: Extract the room temperature parameter data corresponding to different time periods in the database, and divide the data into training sets and test sets according to a certain ratio; Model training: Select the support vector regression model, use the training set to train the model, evaluate and optimize the trained model; use the test set to evaluate the trained model, input the current feature variables, and use the trained model to predict the room temperature at different times in the future; Result analysis and adjustment: Analyze the difference between the predicted results and the actual temperature, evaluate the accuracy and reliability of the model, and adjust and optimize the model based on the analysis results.

[0010] As a further improvement of the technical solution, setting the buffer time threshold in S3 includes the following steps: S3.1. Set the standard temperature range of the computer room according to the actual situation of the computer room and the requirements for temperature stability; S3.2. Set a reasonable buffer time threshold, monitor the internal temperature of the computer room, and use the temperature range judgment algorithm to determine whether the temperature parameters of the computer room are within the set standard range.

[0011] As a further improvement of the technical solution, the conditional judgment algorithm is used to judge whether the temperature parameter of the computer room is within the set standard range in S3, which is as follows: Assume that the standard temperature range of the equipment room is set to , the current room temperature is , you can use the following conditions to judge: if < or > , then the room temperature parameters are not within the set standard range; Set the tolerance to , then the standard temperature range of the equipment room becomes , then the room temperature parameters are not within the set standard range; if < or > , the room temperature parameters are not within the set standard range.

[0012] As a further improvement of the technical solution, the temperature adjustment strategy is optimized in S4, including the following steps: S4.1. Generate system operation reports according to the set cycle. The report content should include the changes in temperature in the computer room, the adjustment of air conditioning control strategy, and energy consumption; S4.2. Analyze the data in the system operation report, evaluate the effectiveness of the temperature control strategy, and optimize the temperature control strategy based on the analysis results.

[0013] As a further improvement of the present technical solution, a human-computer interaction control panel is adopted in S4 to provide the user with an intuitive and convenient operation interface, and the air conditioning control strategy can be adjusted through the human-computer interaction interface.

[0014] The second object of the present invention is to provide a centralized air-conditioning high-efficiency computer room AI optimization control system, including any one of the centralized air-conditioning high-efficiency computer room AI optimization control methods described above, including a temperature monitoring unit, a prediction unit, a monitoring unit and an optimization analysis unit; The temperature monitoring unit is used to monitor the temperature parameters inside and outside the high-efficiency machine room in real time and store them in the database; The prediction unit is used to obtain equipment load information, extract characteristic variable rules according to the external temperature parameters of the computer room and the equipment load information at different time periods, predict the temperature inside the computer room at different time periods in the future, and adjust the air conditioning control strategy according to the prediction results; The monitoring unit is used to monitor the internal temperature of the computer room based on the adjustment result, set a buffer time threshold, and if the temperature parameter of the computer room within the buffer time threshold is not within the set standard range, an alarm prompt is issued and optimization is performed; if the environmental parameter within the buffer time threshold is within the set standard range, it indicates that the temperature adjustment in the computer room is normal; The optimization and analysis unit is used to regularly generate a system operation report, analyze the effect of the temperature adjustment strategy of the computer room in the system operation report, and optimize the temperature adjustment strategy.

[0015] Compared with the prior art, the present invention has the following beneficial effects: In this AI optimization control method for a high-efficiency computer room with centralized air conditioning, the temperature parameters and equipment load information inside and outside the computer room are monitored in real time, the characteristic variable rules are extracted, and the complex temperature changes are converted into specific analyzable data through the prediction model. These characteristic variables can capture the trend, periodicity and seasonality of temperature changes, so that the temperature prediction model can more accurately predict the temperature in the computer room at different time periods in the future, providing a reliable reference for adjusting the air conditioning control strategy in advance, improving the efficiency of the air conditioning system and reducing operating costs.

[0016] In the AI ​​optimization control method for the high-efficiency computer room of centralized air conditioning, by setting the buffer time threshold, a slight error in the temperature sensor or a short-term environmental interference may lead to a false alarm, causing unnecessary interference to the operation and maintenance personnel. Setting the buffer time threshold can filter out these short-term abnormal situations, improve the accuracy of the alarm, and ensure that the operation and maintenance personnel can concentrate on dealing with abnormal problems.

[0017] In this centralized air-conditioning high-efficiency computer room AI optimization control method, the collected data is collated and summarized to generate a system operation report in a clear and easy-to-read format. The report is in the form of tables, charts or text descriptions through the human-computer interactive control panel, so that users can quickly understand the computer room temperature, air conditioning control strategy and energy consumption. Through the system operation report, managers can fully understand the changes in the temperature in the computer room, the adjustment of the air conditioning control strategy and energy consumption. This helps to timely discover potential problems, such as excessive temperature fluctuations, energy waste caused by unreasonable control strategies, etc., so as to take targeted measures for optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is the overall flow chart of the present invention; Figure 2 It is a flow chart of characteristic variable regularity of temperature parameters of a computer room of the present invention; Figure 3 A flowchart of setting a buffer time threshold value of the present invention; Figure 4 It is a flowchart of the optimization and adjustment process of the present invention; Figure 5 It is a schematic diagram of the working principle of the present invention.

[0019] The meaning of each number in the figure is: 100. Temperature monitoring unit; 200. Prediction unit; 300. Monitoring unit; 400. Optimization analysis unit. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] like Figure 1-Figure 5 As shown, one of the purposes of the present invention is to provide an AI optimization control method for a centralized air-conditioning high-efficiency computer room, comprising the following steps: S1. Real-time monitoring of temperature parameters inside and outside the high-efficiency computer room and storage in the database; In S1, temperature sensors are used to collect the internal and external temperatures of the computer room, and the temperature parameters collected by the sensors are regularly stored in the database.

[0022] Use temperature sensors to monitor the temperature inside and outside the computer room in real time. The internal temperature refers to the air temperature inside the computer room, while the external temperature refers to the ambient temperature outside the computer room. The monitored temperature data is stored in the database. The temperature sensor can collect temperature parameters inside and outside the computer room in real time to ensure the timeliness of the data. The system can respond quickly to temperature changes, adjust the air conditioning control strategy in time, and improve the accuracy of temperature control. The temperature parameters collected by the sensor are regularly stored in the database to ensure the traceability and reliability of the data. Even in the event of system failure or data loss, historical data can be restored from the database for analysis and troubleshooting.

[0023] S2. Obtain equipment load information, extract characteristic variable rules based on the external temperature parameters of the computer room and equipment load information at different time periods, predict the temperature inside the computer room at different time periods in the future, and adjust the air conditioning control strategy based on the prediction results; The analysis of characteristic variables of temperature parameters in different periods in the computer room in S2 includes the following steps: S2.1. Obtain the load information of the equipment in the computer room at different time periods by connecting to the monitoring system of the equipment in the computer room; S2.2, extract characteristic variables related to the temperature inside the computer room by combining the external temperature parameters of the computer room and the equipment load information; S2.3. Establish a temperature prediction model for the computer room, predict the temperature in the computer room at different time periods in the future, and adjust the air conditioning control strategy based on the prediction results.

[0024] By connecting to the monitoring system of the equipment in the computer room, the load information of the equipment in the computer room at different time periods is obtained. The equipment load information includes the operating power parameters of the equipment, which reflects the working status of the equipment and the heat generated. Combined with the external temperature parameters of the computer room and the equipment load information, the characteristic variables related to the temperature in the computer room are extracted. For example, different external temperatures and equipment load states make the temperature in the computer room different in different time periods. Through the characteristic variables such as outdoor temperature changes and equipment load information in the computer room, the changing trend and influencing factors of the temperature in the computer room can be reflected. The appropriate machine learning or deep learning algorithm is selected to establish a temperature prediction model in the computer room. The model is trained using historical data and the parameters of the model are adjusted to improve the accuracy of the prediction. According to the current external temperature parameters of the computer room and the equipment load information, the trained prediction model is used to predict the temperature in the computer room at different time periods in the future. According to the prediction results, the control strategy of the air conditioner is adjusted, such as adjusting the set temperature, wind speed, and operation mode of the air conditioner, so as to keep the temperature in the computer room within a suitable range.

[0025] Taking into account the two important factors of the external temperature of the computer room and the equipment load, we can more comprehensively understand the reasons for the temperature changes in the computer room. The changes in the external temperature will affect the internal temperature through the cooling system of the computer room, while the equipment load directly generates heat. Extracting the characteristic variables related to the temperature in the computer room can deeply analyze the complex relationship between these factors and the temperature of the computer room, and provide a basis for accurate temperature prediction. By extracting characteristic variables, complex temperature changes can be converted into specific analyzable data. These characteristic variables can capture the trends, periodicity and seasonality of temperature changes, so that the temperature prediction model can more accurately predict the temperature in the computer room at different time periods in the future, providing a reliable reference for adjusting the air conditioning control strategy in advance, improving the efficiency of the air conditioning system, and reducing operating costs.

[0026] Specific steps for establishing the prediction model in S2.3: Data collection and organization: Obtain historical room temperature data from the database, including temperature records at different time points, and pre-process the data to remove outliers and noise; Feature engineering: Extract the room temperature parameter data corresponding to different time periods in the database, and divide the data into training sets and test sets according to a certain ratio; Model training: Select the support vector regression model, use the training set to train the model, evaluate and optimize the trained model; use the test set to evaluate the trained model, input the current feature variables, and use the trained model to predict the room temperature at different times in the future; Result analysis and adjustment: Analyze the difference between the predicted results and the actual temperature, evaluate the accuracy and reliability of the model, and adjust and optimize the model based on the analysis results.

[0027] During specific use, database management software is used to read data and specify the required time range and data granularity. For example, hourly temperature records for the past year can be obtained, and statistical methods or data visualization techniques can be used to identify outliers. By calculating the mean and standard deviation of the temperature data, data that exceeds a certain range, such as the mean plus or minus three times the standard deviation, is considered an outlier, and the time information is converted into features that can be used for model training. For example, hours can be extracted as features of the time scale. Time features can help the model capture seasonal and periodic changes in temperature. These statistical features can reflect the distribution and variation of temperature and provide relevant features for the model. Use random sampling to divide the data into training sets and test sets, ensuring that the division process is random to avoid data bias. The commonly used division ratio is 70% of the data as the training set and 30% of the data as the test set. The temperature data and feature variables in the training set are sorted to ensure that the data format meets the requirements of the model. Use the training set to train the SVR model. The functions provided by the machine learning library model can be used to implement the training of the SVR model. The parameters of the hyperplane will be continuously adjusted to minimize the prediction error on the training set. The data in the test set is input into the trained SVR model to obtain the predicted value of the model. According to the error analysis and evaluation results, the parameters of the SVR model are adjusted to improve the accuracy and reliability of the model. The parameters can be adjusted step by step to observe the performance changes of the model and find the optimal parameter combination.

[0028] By extracting characteristic variables, complex temperature changes can be converted into specific analyzable data. These characteristic variables can capture the trends and periodicity of temperature changes, allowing the temperature prediction model to more accurately predict the temperature in the computer room at different time periods in the future, providing a reliable reference for adjusting the air conditioning control strategy in advance. Based on the prediction results, the operating mode and parameters of the air conditioner can be reasonably adjusted in advance to improve the efficiency of temperature regulation in the computer room and reduce operating costs.

[0029] S3. Based on the adjustment result, the internal temperature of the computer room is monitored and a buffer time threshold is set. If the temperature parameter of the computer room within the buffer time threshold is not within the set standard range, an alarm is issued and optimization is performed. If the environmental parameter within the buffer time threshold is within the set standard range, it indicates that the temperature adjustment in the computer room is normal. Setting the buffer time threshold in S3 includes the following steps: S3.1. Set the standard temperature range of the computer room according to the actual situation of the computer room and the requirements for temperature stability; S3.2. Set a reasonable buffer time threshold, monitor the internal temperature of the computer room, and use the temperature range judgment algorithm to determine whether the temperature parameters of the computer room are within the set standard range.

[0030] In S3, a conditional judgment algorithm is used to determine whether the temperature parameter of the equipment room is within the set standard range, as follows: Assume that the standard temperature range of the equipment room is set to , the current room temperature is , you can use the following conditions to judge: if < or > , then the room temperature parameters are not within the set standard range; Set the tolerance to , then the standard temperature range of the equipment room becomes , then the room temperature parameters are not within the set standard range; if < or > , the room temperature parameters are not within the set standard range.

[0031] Tolerance determines an acceptable range of fluctuations around a standard temperature range. It reflects the system's tolerance to temperature changes. For example, without tolerance, the temperature must be strictly within the standard range [20°C, 25°C] to be considered normal. But after introducing tolerance, such as a tolerance of 1°C, the temperature can fluctuate within a wider range of [19°C, 26°C] without being considered abnormal.

[0032] During specific use, a temperature sensor is installed inside the computer room to monitor the temperature in the computer room in real time. The location of the sensor should be selected at a location that can represent the overall temperature of the computer room, such as the center of the computer room or the equipment concentration area. By setting a standard temperature range for the computer room, for example, the standard temperature range of the computer room is 20-25°C, when the prediction model predicts that the temperature in the computer room will rise or fall in the next period of time, the air conditioning strategy is adjusted in advance to cool down or heat up the inside of the computer room. Since a certain time period is required in the process of regulating the temperature inside the computer room, a buffer threshold time is set, such as 10 minutes, and the temperature is measured inside the computer room after the regulation is completed. If the environmental parameters are within the buffer time threshold, If the number is within the set standard range, it means that the temperature adjustment in the computer room is normal. If the computer room temperature parameter is not within the set standard range within the buffer time threshold, an alarm prompt will be issued so that relevant personnel can take timely measures to intervene and adjust the temperature inside the computer room, so that the temperature adjustment process in the computer room is always in a smooth state, avoiding the internal temperature of the computer room from fluctuating and affecting the performance of the equipment in the computer room. At the same time, by setting the buffer time threshold, a slight error in the temperature sensor or a short-term environmental interference may cause a false alarm, causing unnecessary interference to the operation and maintenance personnel. Setting the buffer time threshold can filter out these short-term abnormal situations, improve the accuracy of the alarm, and ensure that the operation and maintenance personnel can concentrate on dealing with abnormal problems.

[0033] S4. Generate system operation reports regularly, analyze the effects of the computer room temperature adjustment strategies in the system operation reports, and optimize the temperature adjustment strategies.

[0034] The temperature regulation strategy is optimized in S4, including the following steps: S4.1. Generate system operation reports according to the set cycle. The report content should include the changes in temperature in the computer room, the adjustment of air conditioning control strategy, and energy consumption; S4.2. Analyze the data in the system operation report, evaluate the effectiveness of the temperature control strategy, and optimize the temperature control strategy based on the analysis results.

[0035] The S4 uses a human-machine interactive control panel to provide users with an intuitive and convenient operating interface, and adjust the air-conditioning control strategy through the human-machine interactive interface.

[0036] When used specifically, the generation cycle of the system operation report is determined according to the actual operation of the computer room and management needs. For example, you can choose to generate reports every day, week or month, obtain the temperature change data in the computer room from the system, ensure the accuracy and completeness of the temperature data, including the temperature values, maximum temperature, minimum temperature, and average temperature in different time periods, and record the adjustment of the air conditioning control strategy, including the time of adjustment, the adjusted parameters such as set temperature, wind speed, etc., and the reasons for the adjustment. By collating and summarizing the collected data, a system operation report is generated in a clear and easy-to-read format. The report is in the form of tables, charts or text descriptions through the human-computer interactive control panel, so that users can quickly understand the temperature of the computer room, the air conditioning control strategy and energy consumption. Through the system operation report, managers can fully understand the temperature changes in the computer room, the adjustment of the air conditioning control strategy and the energy consumption, which helps to timely discover potential problems, such as excessive temperature fluctuations, energy waste caused by unreasonable control strategies, etc., so as to take targeted measures for optimization.

[0037] The second object of the present invention is to provide a centralized air-conditioning high-efficiency computer room AI optimization control system, including any one of the centralized air-conditioning high-efficiency computer room AI optimization control methods described above, including a temperature monitoring unit 100, a prediction unit 200, a monitoring unit 300 and an optimization analysis unit 400; The temperature monitoring unit 100 is used to monitor the temperature parameters inside and outside the high-efficiency machine room in real time and store them in the database; The prediction unit 200 is used to obtain equipment load information, extract characteristic variable rules according to the external temperature parameters of the computer room and the equipment load information at different time periods, predict the temperature inside the computer room at different time periods in the future, and adjust the air conditioning control strategy according to the prediction results; The monitoring unit 300 is used to monitor the internal temperature of the computer room based on the adjustment result, set a buffer time threshold, and if the temperature parameter of the computer room within the buffer time threshold is not within the set standard range, an alarm prompt is issued and optimization is performed; if the environmental parameter within the buffer time threshold is within the set standard range, it indicates that the temperature adjustment in the computer room is normal; The optimization and analysis unit 400 is used to regularly generate a system operation report, analyze the effect of the temperature adjustment strategy of the computer room in the system operation report, and optimize the temperature adjustment strategy.

[0038] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. AI optimization control method for high-efficiency central air-conditioning room, characterized by: The following steps are involved: S1. Real-time monitoring of temperature parameters inside and outside the high-efficiency computer room and storage in the database; S2. Obtain equipment load information, extract characteristic variable rules based on the external temperature parameters of the computer room and equipment load information at different time periods, predict the temperature inside the computer room at different time periods in the future, and adjust the air conditioning control strategy based on the prediction results; S3. Based on the adjustment result, the internal temperature of the computer room is monitored and a buffer time threshold is set. If the temperature parameter of the computer room within the buffer time threshold is not within the set standard range, an alarm is issued and optimization is performed. If the environmental parameter within the buffer time threshold is within the set standard range, it indicates that the temperature adjustment in the computer room is normal. S4. Generate system operation reports regularly, analyze the effects of the computer room temperature adjustment strategies in the system operation reports, and optimize the temperature adjustment strategies.

2. The AI ​​optimization control method for a centralized air-conditioning high-efficiency computer room according to claim 1 is characterized in that: In S1, a temperature sensor is used to collect the internal and external temperatures of the computer room, and the temperature parameters collected by the sensor are regularly stored in a database.

3. The AI ​​optimization control method for a centralized air-conditioning high-efficiency computer room according to claim 1 is characterized in that: The analysis of characteristic variable rules of temperature parameters of the computer room in different time periods in S2 includes the following steps: S2.

1. Obtain the load information of the equipment in the computer room at different time periods by connecting to the monitoring system of the equipment in the computer room; S2.2, extract characteristic variables related to the temperature inside the computer room by combining the external temperature parameters of the computer room and the equipment load information; S2.

3. Establish a temperature prediction model for the computer room, predict the temperature in the computer room at different time periods in the future, and adjust the air conditioning control strategy based on the prediction results.

4. The AI ​​optimization control method for a centralized air-conditioning high-efficiency computer room according to claim 3 is characterized in that: The specific steps of establishing the prediction model in S2.3 are: Data collection and organization: Obtain historical room temperature data from the database, including temperature records at different time points, and pre-process the data to remove outliers and noise; Feature engineering: Extract the room temperature parameter data corresponding to different time periods in the database, and divide the data into training sets and test sets according to a certain ratio; Model training: Select the support vector regression model, use the training set to train the model, evaluate and optimize the trained model; use the test set to evaluate the trained model, input the current feature variables, and use the trained model to predict the room temperature at different times in the future; Result analysis and adjustment: Analyze the difference between the predicted results and the actual temperature, evaluate the accuracy and reliability of the model, and adjust and optimize the model based on the analysis results.

5. The AI ​​optimization control method for a centralized air-conditioning high-efficiency computer room according to claim 1 is characterized in that: The buffer time threshold is set in S3, including the following steps: S3.

1. Set the standard temperature range of the computer room according to the actual situation of the computer room and the requirements for temperature stability; S3.

2. Set a reasonable buffer time threshold, monitor the internal temperature of the computer room, and use the temperature range judgment algorithm to determine whether the temperature parameters of the computer room are within the set standard range.

6. The AI ​​optimization control method for a centralized air-conditioning high-efficiency computer room according to claim 5 is characterized in that: In S3, whether the temperature parameter of the equipment room is within the set standard range is judged by using a conditional judgment algorithm, which is specifically as follows: Assume that the standard temperature range of the equipment room is set to , the current room temperature is , you can use the following conditions to judge: if < or > , then the room temperature parameters are not within the set standard range; Set the tolerance to , then the standard temperature range of the equipment room becomes , then the room temperature parameters are not within the set standard range; if < or > , the room temperature parameters are not within the set standard range.

7. The AI ​​optimization control method for a centralized air-conditioning high-efficiency computer room according to claim 1 is characterized in that: The temperature regulation strategy is optimized in S4, comprising the following steps: S4.

1. Generate system operation reports according to the set cycle. The report content should include the changes in temperature in the computer room, the adjustment of air conditioning control strategy, and energy consumption; S4.

2. Analyze the data in the system operation report, evaluate the effectiveness of the temperature control strategy, and optimize the temperature control strategy based on the analysis results.

8. The AI ​​optimization control method for a centralized air-conditioning high-efficiency computer room according to claim 7 is characterized in that: The S4 adopts a human-machine interactive control panel to provide users with an intuitive and convenient operation interface, and adjust the air conditioning control strategy through the human-machine interactive interface.

9. Used to implement a centralized air-conditioning high-efficiency computer room AI optimization control system, comprising the centralized air-conditioning high-efficiency computer room AI optimization control method according to any one of claims 1 to 8, characterized in that: It comprises a temperature monitoring unit (100), a prediction unit (200), a monitoring unit (300) and an optimization analysis unit (400); The temperature monitoring unit (100) is used to monitor the temperature parameters inside and outside the high-efficiency machine room in real time and store them in a database; The prediction unit (200) is used to obtain equipment load information, extract characteristic variable rules according to the external temperature parameters of the computer room and the equipment load information at different time periods, predict the temperature inside the computer room at different time periods in the future, and adjust the air conditioning control strategy according to the prediction results; The monitoring unit (300) is used to monitor the temperature inside the computer room based on the adjustment result, set a buffer time threshold, and if the temperature parameter of the computer room within the buffer time threshold is not within the set standard range, an alarm prompt is issued and optimization is performed; if the environmental parameter within the buffer time threshold is within the set standard range, it indicates that the temperature adjustment in the computer room is normal; The optimization analysis unit (400) is used to regularly generate a system operation report, analyze the effect of the temperature adjustment strategy of the computer room in the system operation report, and optimize the temperature adjustment strategy.

Citation Information

Patent Citations

  • Metal surface defect detection device and method based on stereo camera machine vision

    CN115060648A

  • Vehicle-mounted air conditioner operation control system based on Internet

    CN116691282A

  • Central air-conditioning water system group control method

    CN117490193A

  • Low-cost commissioning method for the air-conditioning systems in existing large public buildings

    US20210123625A1

  • Machine room temperature control method and apparatus, and electronic device and storage medium

    WO2024016586A1

Cited By

  • Air conditioner dynamic temperature difference AI adjusting system based on Internet of Things

    CN119554733A

  • Thermal power plant steam turbine bearing temperature adjusting method, system, equipment and medium

    CN119712252A

  • Machine room air conditioner, control method and device thereof, storage medium and program product

    CN120684773A

  • Efficient machine room energy-saving control system based on global optimization

    CN122015237A

  • Intelligent temperature control energy-saving method for linkage of machine room server cluster and refrigeration air conditioner

    CN122069697A