Machine room energy-saving control method and system based on cloud computing

Abstract

The invention relates to the technical field of machine room energy consumption management, and discloses a machine room energy-saving control method and system based on cloud computing, and the method comprises the steps: collecting equipment operation data in real time through a multi-dimensional sensor network, building a power consumption prediction model and an energy distribution optimization model, and extracting key features through principal component analysis; and constructing a multiple regression model to determine an energy efficiency coefficient, and predicting a future power demand by adopting a long-short-term memory neural network. And an energy distribution scheme is optimized based on a genetic algorithm, and accurate power control and dynamic balance are realized. According to the invention, a self-adaptive energy consumption management decision mechanism is also established, the energy consumption trend is analyzed through a sliding window, an emergency adjustment program is started, state estimation is corrected by using a Kalman filtering algorithm, and parameters of a prediction model are dynamically adjusted. According to the method, intelligent management of energy utilization of the data center can be realized, the total energy consumption cost is effectively reduced, the energy utilization efficiency is improved, and technical support is provided for energy conservation and emission reduction of a large machine room.

Description

Technical Field

[0001] The present invention relates to the technical field of computer room energy consumption management, and in particular to a computer room energy-saving control method and system based on cloud computing. Background Art

[0002] With the rapid development of technologies like cloud computing and artificial intelligence, the power consumption of equipment in data centers is increasing exponentially. Energy costs now account for a significant portion of data center operating expenses, necessitating the urgent need for precise and efficient energy management technologies. Current mainstream energy management solutions for data centers rely primarily on traditional monitoring systems and static energy allocation strategies, which are insufficient to cope with the complex and ever-changing operating conditions of equipment.

[0003] Furthermore, the diverse array of equipment in a computer room environment, with complex and ever-changing operating states, makes traditional monitoring methods incapable of creating a complete digital representation of equipment energy consumption. This leads to an information gap between energy consumption data and actual equipment status. This lack of digital representation further restricts the ability to deeply analyze equipment power variations and energy conversion efficiency, making it difficult for managers to accurately grasp the true energy consumption characteristics and trends of different equipment under various operating conditions. This results in the use of a crude static allocation model, which is unable to dynamically optimize energy allocation based on real-time operating conditions, resulting in significant energy waste.

[0004] Therefore, existing technologies generally face problems such as incomplete data collection, lack of depth in energy consumption analysis, and delayed decision-making responses, which makes it difficult to achieve refined management of computer room energy consumption. Summary of the Invention

[0005] The present invention provides a computer room energy-saving control method and system based on cloud computing to achieve refined management of computer room energy consumption.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a computer room energy-saving control method based on cloud computing, comprising: Obtain real-time data and historical power consumption changes of equipment in the computer room, establish a basic data set including power peak, average, and fluctuation range, and obtain the equipment operating parameter matrix; Obtaining the device load according to the device operating parameter matrix and comparing it with a preset load threshold, obtaining the device energy conversion efficiency coefficient based on the comparison result and predicting the power demand of each device, and if the predicted power exceeds the current power supply capacity, obtaining the power consumption prediction matrix and determining the optimal power allocation plan for each device according to the power consumption prediction matrix; A real-time energy scheduling system is established based on the optimal power allocation plan. If the actual power consumption of one of the devices exceeds the preset range, the operating parameters of the computer room equipment are adjusted, the energy configuration status is obtained and continuously monitored, and if abnormal fluctuations are found, the energy allocation parameters are updated and an adaptive energy consumption management decision-making mechanism is obtained.

[0007] Preferably, the acquisition of real-time data of equipment in the computer room and historical changes in power consumption, establishing a basic data set including power peak, average, and fluctuation range, and obtaining an equipment operating parameter matrix includes: The power consumption, temperature distribution and load status of the equipment in the computer room are collected in real time through a multi-dimensional sensor network to obtain raw data streams covering various working conditions. The raw data streams are stored in a pre-established database in a time series format to obtain structured equipment operation records; According to the structured equipment operation record, the historical change trajectory is segmented, and the power peak, average power and fluctuation range in each time period are calculated to form the first basic data; If the fluctuation amplitude exceeds a preset fluctuation threshold, the first basic data is marked as abnormal, and the operating parameters under the abnormal working condition are obtained by combining the load state and temperature distribution data; The basic data set after abnormal marking is compared with the working condition coverage requirements, and the equipment operation parameter matrix is ​​sorted out for subsequent monitoring.

[0008] Preferably, the device load is obtained according to the device operation parameter matrix and compared with a preset load threshold, the energy conversion efficiency coefficient of the device is obtained according to the comparison result, and the power demand of each device is predicted and calculated, and if the predicted power exceeds the current power supply capacity, the power consumption prediction matrix is ​​obtained and the optimal power allocation scheme for each device is determined according to the power consumption prediction matrix, including: Based on the equipment operating parameter matrix, key feature vectors are extracted to construct a multivariate regression model between equipment status and energy consumption level. If the equipment load rate exceeds a preset load threshold, it is marked as a high energy consumption state, and the energy conversion efficiency coefficient of each type of equipment under different load conditions is determined; A device power consumption prediction model is established using the energy conversion efficiency coefficient and the power demand of each device is predicted and calculated. If the predicted power exceeds the current power supply capacity, a load balancing mechanism is triggered to obtain a power consumption prediction matrix including time and device dimensions. An energy allocation optimization model is constructed based on the power consumption prediction matrix. The objective function is set to minimize the overall energy consumption cost. The constraints include the minimum operating power requirement of the equipment and the upper limit of the power supply. The optimal power allocation plan for each device is determined by calculation.

[0009] Preferably, the key feature vectors are extracted based on the device operation parameter matrix, and a multivariate regression model between the device state and the energy consumption level is constructed. If the device load rate exceeds a preset load threshold, it is marked as a high energy consumption state, and the energy conversion efficiency coefficient of each type of device under different load conditions is determined, including: Standardizing the power changes and operating parameters in the device operating parameter matrix, obtaining data content corresponding to the load, obtaining a second basic data set, and performing data comparison with the load threshold; If exceeded, the corresponding device state is marked as the high energy consumption state, and the marked state data group is determined; Performing hierarchical processing on the data under different load conditions to obtain the energy consumption distribution corresponding to the device state and obtain a classified efficiency data set; The efficiency data set is associated with the power change to obtain the energy conversion efficiency coefficient.

[0010] Preferably, the energy conversion efficiency coefficient is used to establish a device power consumption prediction model and predict the power demand of each device. If the predicted power exceeds the current power supply capacity, the load balancing mechanism is triggered to obtain a power consumption prediction matrix including time dimension and device dimension, including: Based on the energy conversion efficiency coefficient, combined with the time dimension and the equipment dimension, historical operation records are classified and sorted to obtain characteristic content related to power demand, thereby obtaining a third basic data set; By merging the third basic data set and combining the data time series characteristics, the power demand in the future period is analyzed and processed to obtain a predicted power demand distribution result; Performing data verification based on the power demand distribution result and in combination with the power supply capacity limit, triggering the load balancing mechanism if it exceeds the limit, and obtaining an adjusted power allocation strategy; According to the power allocation strategy, in combination with the time dimension and the device dimension, a prediction matrix containing multi-dimensional information is constructed to obtain the power consumption prediction matrix.

[0011] Preferably, the energy allocation optimization model is constructed based on the power consumption prediction matrix, the objective function is set to minimize the overall energy consumption cost, the constraints include the minimum operating power requirement of the equipment and the upper limit of the power supply, and the optimal power allocation plan for each device is determined by calculation, including: Classify and process the predicted power data of each device according to the power consumption prediction matrix, obtain power demand distribution information related to the operation of the device, and obtain a classified power demand set; Based on the power demand set combined with the preset supply upper limit and power requirements, verify whether the power demand of each device matches the constraint conditions; if the predicted power of one device exceeds the power supply upper limit, adjust the power allocation ratio of the device to determine an adjusted power demand list; Processing the power allocation plan for each device through the power demand list, combining the objective function and the energy consumption cost data, and obtaining a preliminary allocation result that satisfies the minimum cost; According to the preliminary allocation result and in combination with the actual requirements of the equipment operation, a power allocation plan including the optimization of each device is generated to obtain the optimal power allocation plan.

[0012] Preferably, the real-time energy scheduling system is established according to the optimal power allocation scheme, and if the actual power consumption of one device exceeds a preset range, the operating parameters of the equipment in the computer room are adjusted to obtain the energy configuration status, including: Continuously collect the actual power consumption of each device and determine whether the actual power consumption exceeds a preset range. If so, automatically adjust the operating parameters, generate an adjusted parameter configuration plan based on the requirements of intelligent management, and determine a temporary operating mode suitable for each device; According to the adjusted parameter configuration plan, each device is accurately allocated, the energy configuration data feedback is updated in real time, and a power allocation list that conforms to the current device status is obtained; The power distribution list is compared with the real-time monitoring data to determine the power control strategy of each device and obtain the energy configuration status.

[0013] Preferably, if abnormal fluctuations are found, the energy allocation parameters are updated and an adaptive energy consumption management decision mechanism is obtained, including: If abnormal fluctuations are found, the trend of energy consumption changes is analyzed, and combined with the preset threshold range, it is determined whether the conditions for initiating emergency regulation are met and the specific status of the triggering conditions is determined; According to the specific status of the trigger condition, the allocation strategy information that matches the current fluctuation characteristics is extracted from the historical experience database, an adjusted parameter plan is generated, the updated configuration data is obtained and fed back to the monitoring process to determine whether the preset stability standard is met and obtain the final operating state.

[0014] Preferably, the method further comprises: Establishing a closed-loop feedback control system based on the adaptive energy consumption management decision-making mechanism, comparing the degree of deviation between actual energy consumption and expected targets, dynamically adjusting the weight parameters of the prediction model, and determining a continuously optimized energy management and control plan; Specifically: According to the energy consumption management decision mechanism, actual energy consumption data is obtained, compared with the expected target, and whether it exceeds the preset threshold range is determined to obtain an initial deviation result. If it exceeds, real-time information is collected and combined with historical operation data to obtain an adjusted status record; Extracting a matching solution from a pre-established database according to the adjusted status record and determining an updated weight value; The energy management and control plan is dynamically adjusted through the updated weight value. For the goal of continuous optimization, the adjusted plan data is fed back to the closed-loop process to determine whether it meets the preset conditions and obtain the final configuration result.

[0015] In the second aspect, a computer room energy-saving control system based on cloud computing includes: The detection end is used to obtain real-time data and historical changes in power consumption of equipment in the computer room, establish a basic data set including power peak, average, and fluctuation range, and obtain the equipment operation parameter matrix; The processing end obtains the device load based on the device operating parameter matrix and compares it with a preset load threshold. Based on the comparison result, the energy conversion efficiency coefficient of the device is obtained and the power demand of each device is predicted and calculated. If the predicted power exceeds the current power supply capacity, the power consumption prediction matrix is ​​obtained and the optimal power allocation plan for each device is determined based on the power consumption prediction matrix. The decision-making end is used to establish a real-time energy scheduling system based on the optimal power allocation plan. If the actual power consumption of one of the devices exceeds the preset range, the operating parameters of the computer room equipment are adjusted, the energy configuration status is obtained and continuously monitored. If abnormal fluctuations are found, the energy allocation parameters are updated and an adaptive energy consumption management decision-making mechanism is obtained.

[0016] Compared with the existing technology, the present invention provides a computer room energy-saving control method and system based on cloud computing, which has the following beneficial effects: 1. The present invention uses a multi-dimensional sensor network to collect power consumption, temperature distribution, and load status of computer room equipment in real time, and covers raw data streams under various working conditions. This multi-dimensional, multi-working condition data collection method can more completely construct a digital representation of the energy consumption status of the equipment compared to traditional single data collection methods. For example, traditional methods may only collect power consumption data of the equipment, while the present invention also combines information such as temperature and load status, making data collection more comprehensive, thereby solving the information gap problem caused by insufficient data collection in the existing technology, and realizing a complete digital representation of the energy consumption status of the computer room equipment, providing a comprehensive data foundation for subsequent accurate energy consumption analysis and management.

[0017] 2. The present invention obtains the equipment operation parameter matrix by establishing a basic data set including power peak, average value, and fluctuation amplitude, and then extracts key feature vectors based on the equipment operation parameter matrix to construct a multivariate regression model between the equipment state and the energy consumption level. In this way, the energy conversion efficiency coefficient of the equipment under different load conditions can be deeply analyzed. For example, the traditional method may simply count the average power consumption of the equipment, while the present invention can dig out the complex relationship between the equipment load rate and energy consumption through the multivariate regression model, and then determine the energy conversion efficiency of the equipment under various load conditions. This solves the problem that the existing technology is difficult to deeply analyze the power change law and energy conversion efficiency of the equipment, and provides strong support for accurate energy consumption control.

[0018] 3. The present invention establishes a real-time energy scheduling system. If the actual power consumption of a device exceeds a preset range, the operating parameters of the equipment in the computer room are adjusted, and continuous monitoring is performed. At the same time, by constructing an adaptive energy consumption management decision-making mechanism, energy allocation parameters can be updated promptly when abnormal fluctuations are detected. For example, when the actual power consumption of a device exceeds a preset range, the system can quickly respond and adjust the operating parameters, rather than requiring the long time required to detect and handle energy consumption anomalies as in traditional methods. This effectively solves the problem of delayed decision-making response in existing technologies, improves the speed of decision-making response, and can promptly adjust energy allocation according to the actual operating status of the device, reducing energy waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flow chart of a computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention; Figure 2 This is a flow chart of another computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention; Figure 3 This is a flow chart of another computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention; Figure 4 This is a flow chart of another computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention; Figure 5 This is a flow chart of another computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention; Figure 6 This is a flow chart of another computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention; Figure 7 This is a flow chart of another computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention; Figure 8 This is a flow chart of another computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention; Figure 9 This is another flow chart of a computer room energy-saving control method based on cloud computing provided by an embodiment of the present invention. Figure 10 This is a schematic diagram of the structure of a computer room energy-saving control system based on cloud computing provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] Reference Figure 1 The first embodiment of the present invention provides a flow chart of a computer room energy-saving control method based on cloud computing, comprising the following steps: S1, obtain the real-time data of the equipment in the computer room and the historical change trajectory of power consumption, establish a basic data set including power peak, average, and fluctuation amplitude, and obtain the equipment operation parameter matrix; S2, obtaining the device load based on the device operating parameter matrix and comparing it with a preset load threshold, obtaining the device energy conversion efficiency coefficient based on the comparison result, and predicting the power demand of each device. If the predicted power exceeds the current power supply capacity, obtaining a power consumption prediction matrix and determining the optimal power allocation plan for each device based on the power consumption prediction matrix; S3, establish a real-time energy scheduling system based on the optimal power allocation plan. If the actual power consumption of one of the devices exceeds the preset range, adjust the operating parameters of the equipment in the computer room, obtain the energy configuration status and conduct continuous monitoring. If abnormal fluctuations are found, update the energy allocation parameters and obtain an adaptive energy consumption management decision-making mechanism.

[0022] In step S1, the real-time data of the equipment in the computer room and the historical change trajectory of power consumption are obtained, a basic data set including power peak value, average value and fluctuation range is established, and the equipment operation parameter matrix is ​​obtained.

[0023] refer to Figure 2 , step S1 includes: S11, using a multi-dimensional sensor network to collect real-time data on the power consumption, temperature distribution, and load status of the equipment in the computer room, obtaining raw data streams covering various working conditions, and storing the raw data streams in a pre-established database in a time series format to obtain structured equipment operation records; S12, segmenting the historical change trajectory according to the structured equipment operation record, calculating the power peak, average power, and fluctuation amplitude in each time period to form the first basic data; S13, if the fluctuation amplitude exceeds a preset fluctuation threshold, marking the first basic data as abnormal, and obtaining operating parameters under abnormal conditions in combination with the load state and temperature distribution data; S14, comparing the abnormally marked basic data set with the working condition coverage requirements, and organizing the equipment operation parameter matrix for subsequent monitoring.

[0024] For example, in equipment monitoring scenarios in computer rooms, real-time data collection of equipment power, temperature, and load status through a multi-dimensional sensor network is a key component. Sensors can be deployed near each key device in the room, recording data streams in real time. For example, every minute, power values, temperature values, and load percentage are collected, generating time series data that is stored in a database. This approach ensures data continuity and integrity, providing a reliable foundation for subsequent analysis.

[0025] Specifically, segmented processing of historical traces can divide a day's operating data into 24 hourly time periods, calculating the peak power, average power, and fluctuation range within each time period. For example, between 9:00 AM and 10:00 AM, a device's peak power is 5,000 watts, its average power is 4,500 watts, and its fluctuation range is 10%. If the preset fluctuation range threshold is 8%, the data for that time period will be marked as abnormal. This segmented analysis helps quickly locate problem periods and improves troubleshooting efficiency.

[0026] For example, after an anomaly is flagged, potential risk points can be further analyzed by combining load status and temperature distribution data. For example, if power fluctuations are abnormal during a certain period, while the load factor reaches 90%, and the temperature distribution shows that the core temperature of the equipment reaches 50°C, well above the normal range of 40°C, then the risk of overload or poor heat dissipation can be determined. This multi-dimensional data correlation analysis can more accurately identify the root cause of the problem and prevent equipment damage.

[0027] Specifically, when constructing a comprehensive matrix, multiple parameters such as power, temperature, and load can be organized into a data matrix. For example, each row of the matrix represents a point in time, and each column corresponds to a parameter, such as power values ​​in the first column and temperature values ​​in the second. This structured organization makes the correlations between data clear at a glance, facilitating subsequent monitoring model training and anomaly prediction, significantly improving the intelligence of the monitoring system.

[0028] For example, if data is missing for certain extreme operating conditions, such as operating parameters under high temperature and high load, during the comparison of operating condition coverage requirements, targeted supplemental test data can be added to ensure that the complete data set covers all possible scenarios. This approach effectively reduces monitoring blind spots, improves the system's adaptability to complex operating conditions, and ensures the safe and stable operation of equipment in the computer room.

[0029] Specifically, the implementation of this approach has brought beneficial results, including improved data analysis efficiency, reduced equipment failure rates, and optimized resource allocation. Through real-time data collection and structured storage, data processing time can be reduced by over 30%. Through anomaly tagging and risk assessment, the early warning rate for equipment failures has increased to 90%. Through a comprehensive matrix and complete data set, the monitoring system's coverage of complex operating conditions reaches 95%, providing strong technical support for computer room management.

[0030] In step S2, the device load is obtained according to the device operation parameter matrix and compared with the preset load threshold. The energy conversion efficiency coefficient of the device is obtained according to the comparison result and the power demand of each device is predicted and calculated. If the predicted power exceeds the current power supply capacity, the power consumption prediction matrix is ​​obtained and the optimal power allocation plan for each device is determined according to the power consumption prediction matrix.

[0031] In the field of computer room equipment, the power consumption prediction matrix is ​​a model or tool used to analyze and predict the power consumption of equipment or systems. Through multi-dimensional data association and algorithmic modeling, it helps managers optimize energy use, plan capacity, and reduce operating costs.

[0032] refer to Figure 3 , step S2 includes: S21, extracting key eigenvectors based on the device operating parameter matrix, constructing a multivariate regression model between device status and energy consumption level, marking a device as being in a high energy consumption state if its load rate exceeds a preset load threshold, and determining the energy conversion efficiency coefficient of each type of device under different load conditions; S22: Establishing a device power consumption prediction model based on the energy conversion efficiency coefficient and predicting the power demand of each device. If the predicted power exceeds the current power supply capacity, a load balancing mechanism is triggered to obtain a power consumption prediction matrix including time and device dimensions. S23, constructing an energy allocation optimization model based on the power consumption prediction matrix, setting the objective function to minimize the overall energy consumption cost, and the constraints including the minimum operating power requirement of the equipment and the upper limit of the power supply, and determining the optimal power allocation plan for each device through calculation.

[0033] refer to Figure 4 , S21 includes: S211, standardizing the power changes and operating parameters in the device operating parameter matrix, obtaining data corresponding to the load, obtaining a second basic data set, and comparing the data with the load threshold; S212, if exceeded, marking the corresponding device state as the high energy consumption state, and determining the marked state data group; S213, performing hierarchical processing on the data under different load conditions, obtaining energy consumption distribution corresponding to the device state, and obtaining a classified efficiency data set; S214: Associating the efficiency data set with the power change to obtain the energy conversion efficiency coefficient.

[0034] For example, in the scenario of equipment monitoring in a computer room, specific implementation methods can be developed from multiple dimensions for the processing and analysis of the equipment operating parameter matrix, combined with key data such as power changes and load conditions to ensure the comprehensiveness and accuracy of data processing.

[0035] For example, to standardize multidimensional data, collected data such as power, temperature, and load factor are uniformly formatted. For example, suppose a server in a computer room has power data in watts, load factor expressed as a percentage, and temperature in degrees Celsius. Standardization converts this data into a uniform numerical range, facilitating subsequent comparison and analysis. This approach eliminates dimensional differences and ensures comparability across different types of data.

[0036] For example, when comparing load rate values ​​against thresholds, you can set the load rate threshold to 85%. If a device's load rate reaches 88% during a specific time period, the system automatically marks it as high energy consumption. This marking mechanism helps quickly identify potentially high-risk devices, providing a basis for subsequent management.

[0037] For example, by layering state data groups, energy consumption distribution under different load conditions can be broken down. For example, under high-energy consumption conditions with a load factor exceeding 80%, the energy consumption distribution of a device shows that most of its energy consumption is concentrated in the processor module. However, when the load factor is below 50%, the energy consumption distribution is more balanced. This categorized analysis clearly reveals the energy consumption characteristics of a device under different operating conditions, providing a reference for optimizing operational strategies.

[0038] For example, when processing efficiency data sets, the power fluctuations are correlated with eigenvectors, analyzing the relationship between power fluctuations and parameters such as load factor and temperature. For example, if a device experiences frequent power fluctuations and a simultaneous temperature increase when the load factor is high, it can be inferred that the power fluctuations are likely related to insufficient heat dissipation capacity. This comprehensive state correspondence provides a deeper understanding of device operating patterns.

[0039] For example, to determine the final reference data set, the above analysis results can be integrated into a structured dataset that includes the power, energy consumption distribution, and efficiency characteristics of the device under different load conditions. For example, if a server's energy efficiency is only 70% at a 90% load, but reaches 85% at a 60% load, the server's operating load can be adjusted based on the reference data set to improve overall efficiency.

[0040] For example, in practice, the above method can be further optimized from a data visualization perspective, using charts to display the corresponding relationship between load rate and power changes, providing an intuitive presentation of equipment status. This approach helps managers quickly understand equipment operating conditions and improve decision-making efficiency. Through this multi-dimensional analysis and processing, the level of refinement of equipment monitoring in the computer room is enhanced, data utilization is significantly improved, and reliable support for equipment management is provided.

[0041] refer to Figure 5 , S22 includes: S221: Classify and organize historical operation records based on the energy conversion efficiency coefficient, in combination with the time dimension and the device dimension, obtain characteristic content related to power demand, and obtain a third basic data set; S222, analyzing and processing the power demand in the future time period by merging the third basic data set and combining the data time series characteristics, and obtaining a predicted power demand distribution result; S223, performing data verification based on the power demand distribution result and the power supply capacity limit, and triggering the load balancing mechanism if the limit is exceeded to obtain an adjusted power allocation strategy; S224 : Construct a prediction matrix containing multi-dimensional information according to the power allocation strategy in combination with the time dimension and the device dimension to obtain the power consumption prediction matrix.

[0042] For example, in the context of equipment monitoring in a computer room, the application of energy conversion efficiency coefficients can be analyzed by combining historical operating records with information from both time and device dimensions. By categorizing and organizing historical operating records, we can extract features related to power demand. For example, suppose a server's operating data from the past month shows an energy conversion efficiency coefficient of 0.75 during peak hours and 0.85 during off-peak hours. Combining this with time-based data reveals a close correlation between efficiency fluctuations and operating time periods. This data can be categorized by time period and device type to form an initial data set, laying the foundation for subsequent predictions.

[0043] For example, when combining time series features with the initial data set to forecast power demand, we analyze the distribution of power demand in future time periods. For example, based on historical data, we predict that during the upcoming holiday peak season, the power demand of a server cluster may reach 5,000 watts, a 20% increase compared to normal times. This forecasting process comprehensively considers historical operating patterns and temporal characteristics to ensure that the results are close to actual demand.

[0044] Suppose the total power supply capacity of the computer room is limited to 8,000 watts, while the predicted power demand is 8,500 watts, exceeding the preset threshold by 500 watts. In this case, the system automatically triggers the load balancing mechanism and adjusts the power distribution strategy, shifting some of the load to backup equipment or operating during off-peak hours to avoid the risk of power shortages.

[0045] For example, when generating power allocation policies, a multidimensional prediction matrix is ​​constructed by combining time and device information. Suppose, after adjustments, a server's peak load is limited to 70%, its power allocation is 3,000 watts, and some tasks are scheduled for the off-peak hours of the early morning. This generates a prediction matrix that incorporates not only power data but also device operating status and time distribution information, providing a comprehensive reference for managers.

[0046] For example, during the process of generating the final power consumption forecast, the data can be further refined. Suppose the forecast matrix shows that a device's average power consumption over the next week will be 2500 watts, with potential short-term peaks during specific time periods. By comparing this data with historical records, proactive countermeasures can be formulated, such as adjusting the device's operating plan or increasing cooling support, to ensure stable device operation. This integration of multi-dimensional information helps improve the accuracy and practicality of forecasts, providing strong support for computer room management.

[0047] refer to Figure 6 , S23 includes: S231, classifying the predicted power data of each device according to the power consumption prediction matrix, obtaining power demand distribution information related to the operation of the device, and obtaining a classified power demand set; S232, based on the power demand set combined with the preset supply upper limit and power requirements, verifying whether the power demand of each device matches the constraint conditions; if the predicted power of a device exceeds the power supply upper limit, adjusting the power allocation ratio of the device, and determining an adjusted power demand list; S233, processing the power allocation plan for each device based on the power demand list, the objective function, and the energy consumption cost data to obtain a preliminary allocation result that satisfies the minimum cost; S234: Based on the preliminary allocation result and in combination with the actual operation requirements of the equipment, a power allocation plan including the optimized power allocation of each equipment is generated to obtain the optimal power allocation plan.

[0048] For example, in the scenario of equipment management in a computer room, the power consumption prediction matrix can be used to categorize predicted power data. For example, suppose a computer room contains multiple server clusters. The prediction matrix shows that some devices have high power demands during specific time periods, while others are relatively stable. This data can be categorized by device type and time period to generate power demand distribution information. This results in a categorized power demand set. For example, the predicted power demand for server A during peak periods is 4000 watts, and during off-peak periods it is 2000 watts.

[0049] For example, during the power demand set verification process, the system analyzes whether the power requirements of each device meet the constraints, combining the preset supply cap and power requirements. For example, suppose the total power cap for a computer room is 10,000 watts, and a server cluster's peak power consumption is predicted to reach 11,000 watts, exceeding the cap by 1,000 watts. The system then identifies the exceeding device and adjusts its power allocation, for example, reducing the cluster's power allocation from 100% to 90%. This creates an adjusted power demand list, ensuring that the total power demand remains within the supply cap.

[0050] For example, when optimizing allocation plans based on a power demand list, the power allocation plan for each device is processed based on the objective function and energy cost data. Assuming the objective function is cost-minimization, and the operating cost of a particular device is higher during peak hours and lower during off-peak hours, the system will tend to shift some of the load to off-peak hours. Preliminary allocation results might indicate that Server B's power allocation during peak hours should be adjusted from 3000 watts to 2000 watts, with the remaining load scheduled for the early morning hours.

[0051] For example, further optimization of the preliminary allocation results, combined with the actual operational needs of the devices, generates final energy allocation data. For example, suppose Server C's power is initially limited to 2500 watts, but its actual operational requirements require at least 2800 watts. The system adjusts the allocation ratios of other devices based on their priorities to ensure that Server C's operational needs are met. The final energy allocation data not only includes the optimized power plan for each device but also time distribution information, providing a comprehensive reference for managers.

[0052] For example, in the aforementioned process, each step of classification, verification, adjustment, and optimized allocation is closely aligned with the actual operating characteristics of the equipment in the computer room, ensuring that the power allocation plan meets both supply constraints and equipment requirements. This approach helps improve resource utilization efficiency, reduce operating costs, and ensure stable equipment operation during peak periods, providing reliable support for computer room management.

[0053] In step S3, a real-time energy scheduling system is established based on the optimal power allocation scheme. If the actual power consumption of one of the devices exceeds the preset range, the operating parameters of the computer room equipment are adjusted, the energy configuration status is obtained and continuously monitored, and if abnormal fluctuations are found, the energy allocation parameters are updated and an adaptive energy consumption management decision mechanism is obtained.

[0054] In the field of computer room equipment, energy configuration status refers to a comprehensive description of the distribution, usage, and real-time operational status of electrical energy within a data center or computer room. It involves the deployment and monitoring of key elements such as the power system, power distribution units, equipment loads, and backup energy sources, ensuring efficient, stable, and secure energy support for computer room equipment operation.

[0055] refer to Figure 7 , S3 includes: S31, continuously collecting actual power consumption of each device, determining whether the actual power consumption exceeds a preset range, and if so, automatically adjusting the operating parameters, generating an adjusted parameter configuration plan in accordance with intelligent management requirements; S32, accurately allocating power to each device based on the adjusted parameter configuration plan, updating energy configuration data feedback in real time, and obtaining a power allocation list that meets the current device status; S33, comparing the power distribution list with the real-time monitoring data, determining the power control strategy of each device, and obtaining the energy configuration status.

[0056] For example, in the case of equipment management in a computer room, real-time monitoring requires continuous acquisition of power consumption information for each device. Imagine a computer room with multiple server clusters, recording actual power consumption data for each device every minute. A server's power consumption suddenly rises from the expected 3,000 watts to 3,800 watts during a specific period. Given a ±10% tolerance, with an upper limit of 3,300 watts, the system determines that the device's power consumption exceeds the tolerance range and generates a preliminary determination. This approach allows for timely detection of abnormal conditions, providing a basis for subsequent adjustments.

[0057] If the server in question experiences abnormal power consumption due to excessive load, the system analyzes device status data and decides to reduce its operating frequency or limit non-critical tasks. This generates an adjusted parameter configuration to keep power consumption below 3200 watts. This temporary operating mode ensures basic device functionality while avoiding the risk of overload.

[0058] For example, when implementing precise allocation, power is dynamically allocated to devices based on the adjusted parameter configuration. For example, if the total power supply limit for a computer room is 12,000 watts and the current total demand is 11,800 watts, the system, by updating energy configuration data in real time, reduces the power of the server that exceeds the tolerance from 3,800 watts to 3,200 watts. The saved 600 watts is then allocated to other devices in need, forming a new power allocation list. This dynamic balancing strategy ensures the rational use of overall power resources.

[0059] For example, the power allocation list is compared with real-time monitoring data to further optimize energy allocation. If the comparison reveals that a device's power consumption is still close to the upper tolerance limit after adjustment, the system will analyze its load trends based on historical data to determine whether the power allocation ratio needs to be further reduced or some tasks should be transferred to other devices. Ultimately, the power control strategy for each device is determined. This approach continuously optimizes the configuration status and adapts to dynamic changes in device operation.

[0060] For example, when generating the final configuration state under dynamic balancing, the system comprehensively considers the real-time status and historical data of all devices. For example, if a server cluster's adjusted power allocation is 2500 watts, but monitoring data indicates an impending increase in workload, the system will reserve a 300-watt power buffer to ensure no anomalies during peak load. This proactive adjustment strategy effectively improves device operational stability.

[0061] For example, from a holistic perspective, each of the aforementioned steps closely integrates with the actual operational characteristics of the equipment in the computer room, forming a complete closed-loop management process, from data collection to parameter adjustment, to power distribution and optimized configuration. This approach maximizes energy efficiency while ensuring normal equipment operation, providing reliable support for computer room management.

[0062] refer to Figure 8 , S3 includes: S34, if abnormal fluctuations are found, the trend of energy consumption changes is analyzed and, based on the preset threshold range, it is determined whether the conditions for initiating emergency regulation are met; S35, if satisfied, extract the allocation strategy information that matches the current fluctuation characteristics from the historical experience database, generate an adjusted parameter plan, obtain the updated configuration data and feed it back to the monitoring process to determine whether the preset stability standard is met and obtain the final operating state.

[0063] For example, in a computer room energy management scenario, real-time energy consumption information for each device is extracted to meet the need for continuous monitoring. Consider a computer room with multiple server clusters, where the operating status of each device is recorded hourly, including key indicators such as voltage, current, and power. Preliminary analysis revealed that the energy consumption of a particular server increased significantly from 2000 watts to 2600 watts within a short period of time. The system flagged this as a potential risk point. This approach enables rapid identification of anomalies, providing a foundation for subsequent action.

[0064] For example, the initial results of abnormal fluctuations will be followed by a detailed analysis of energy consumption trends. For example, if the preset energy consumption fluctuation threshold is plus or minus 15%, that is, the upper limit of the server is 2300 watts, and the current 2600 watts exceeds the range, the system will determine that the conditions for initiating emergency adjustment are met. This detailed analysis helps clarify whether the fluctuation requires intervention, avoiding misjudgment or excessive adjustment.

[0065] For example, after determining the trigger condition, the system extracts allocation strategy information with similar fluctuation characteristics from a historical experience database. If the database records a case in which energy consumption was controlled by reducing the operating frequency under similar circumstances, the system will generate an adjustment plan based on this information, reducing the frequency of the server by 20% from the full load state, and reducing energy consumption to within 2200 watts. This adaptive decision-making allows for flexible response to actual conditions, ensuring the rationality of configuration data.

[0066] For example, updated configuration data dynamically adjusts energy allocation. For example, if the total power supply capacity of a computer room is 10,000 watts and the current total demand is 9,800 watts, the system will reduce the power of the server from 2,600 watts to 2,200 watts, leaving 400 watts available for other devices. This adjustment information is then fed back to the monitoring process to determine whether stability standards have been met. This dynamic adjustment mechanism effectively balances resource allocation.

[0067] For example, during the final operational status verification process, the system continuously monitors the effects of adjustments. Assuming that after the adjustment, the server's energy consumption remains stable at 2150 watts and the overall computer room operating status exhibits no new fluctuations, the system determines that the preset stability standard has been met. This closed-loop feedback mechanism enables timely verification of adjustment results, ensuring long-term operational reliability.

[0068] For example, in response to complex fluctuations, the system can also predict potential risks using historical data. For example, if a server's energy consumption remains stable after adjustments, but historical data shows that its workload often surges during specific periods, the system will reserve 200 watts of buffer capacity in advance. This proactive strategy can further enhance refined management.

[0069] For example, the system regularly updates the strategy information in its database to meet the management mechanism's objectives. If a more optimal frequency control ratio is accumulated through repeated adjustments, the system will prioritize it for subsequent fluctuation management. This continuous optimization approach continuously improves the adaptability and efficiency of energy allocation.

[0070] refer to Figure 9 , the method further comprises: Establishing a closed-loop feedback control system based on the adaptive energy consumption management decision-making mechanism, comparing the degree of deviation between actual energy consumption and expected targets, dynamically adjusting the weight parameters of the prediction model, and determining a continuously optimized energy management and control plan; Specifically: According to the energy consumption management decision mechanism, actual energy consumption data is obtained, compared with the expected target, and whether it exceeds the preset threshold range is determined to obtain an initial deviation result. If it exceeds, real-time information is collected and combined with historical operation data to obtain an adjusted status record; Extracting a matching solution from a pre-established database according to the adjusted status record and determining an updated weight value; The energy management and control plan is dynamically adjusted through the updated weight value. For the goal of continuous optimization, the adjusted plan data is fed back to the closed-loop process to determine whether it meets the preset conditions and obtain the final configuration result.

[0071] For example, in the energy management scenario of a computer room, actual energy consumption data can be obtained from equipment operation records to implement energy management decision-making mechanisms. Suppose a server cluster's daily energy consumption records show an average power of 3,000 watts, while the preset target is 2,800 watts. The discrepancy is 200 watts. If the preset threshold range is plus or minus 100 watts, the initial deviation indicates that the range has been exceeded. This comparison provides a basis for subsequent adjustments.

[0072] For example, real-time device status information is collected to detect threshold-exceeding conditions. For example, if a server's current load suddenly increases, the system can adjust its estimated status based on historical energy consumption data under similar loads. If it finds that actual energy consumption may have been affected by a temporary task, the adjusted status record will indicate that the fluctuation is short-term. This correction helps more accurately determine whether intervention is necessary.

[0073] For example, after obtaining the adjusted status record, the system analyzes the current configuration of the prediction weights. Suppose the current weights are biased toward the historical average, but the current deviation is characterized by a sudden load. The system extracts matching solutions from the database and decides to adjust the weights to a model that prioritizes real-time data. The updated weights are more reflective of the current status. This dynamic adjustment ensures the targeted nature of the predictions.

[0074] For example, based on the updated weights, the energy management plan is dynamically adjusted. For example, if the total power supply capacity of a computer room is 10,000 watts and the current total demand is 9,500 watts, the system will reallocate the excess 200 watts of power to reduce the operating frequency of the server, reducing the estimated energy consumption to 2,850 watts. This adjustment effectively balances overall resource allocation.

[0075] For example, to achieve continuous optimization, adjusted solution data is fed back into the closed-loop process. Assuming that the server energy consumption remains within the target range after adjustment and overall data center operations are not affected, the system determines that the preset conditions have been met and confirms the final configuration. This closed-loop mechanism ensures the reliability of adjustments.

[0076] For example, looking at it from another perspective, the system can also use historical solutions in the database to supplement adjustments based on the specific characteristics of energy consumption deviations. If lowering fan speed has previously reduced energy consumption in similar situations, the system can use this as a backup solution, flexibly applying it based on the current weighting. This combination of multiple solutions improves management adaptability.

[0077] For example, in actual operation, the correction process for status estimates can incorporate additional information such as device operating time. For example, if a server has been running continuously for longer than expected, the system will further analyze whether the energy consumption deviation is related to hardware aging and adjust the status record accordingly. This detailed analysis provides a more comprehensive basis for subsequent decision-making.

[0078] Reference Figure 10 The embodiment of the present invention provides a computer room energy-saving control system based on cloud computing, including: The detection end is used to obtain real-time data and historical changes in power consumption of equipment in the computer room, establish a basic data set including power peak, average, and fluctuation range, and obtain the equipment operation parameter matrix; The processing end obtains the device load based on the device operating parameter matrix and compares it with a preset load threshold. Based on the comparison result, the energy conversion efficiency coefficient of the device is obtained and the power demand of each device is predicted and calculated. If the predicted power exceeds the current power supply capacity, the power consumption prediction matrix is ​​obtained and the optimal power allocation plan for each device is determined based on the power consumption prediction matrix. The decision-making end is used to establish a real-time energy scheduling system based on the optimal power allocation plan. If the actual power consumption of one of the devices exceeds the preset range, the operating parameters of the computer room equipment are adjusted, the energy configuration status is obtained and continuously monitored. If abnormal fluctuations are found, the energy allocation parameters are updated and an adaptive energy consumption management decision-making mechanism is obtained.

[0079] It should be noted that the cloud computing-based computer room energy-saving control system provided in an embodiment of the present invention is used to execute all the process steps of the cloud computing-based computer room energy-saving control method in the above embodiment. The working principles and beneficial effects of the two correspond one to one, so they will not be repeated here.

[0080] The embodiment of the present invention further provides a terminal device. The terminal device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a computer room energy-saving control program based on cloud computing. When the processor executes the computer program, the steps in the above-mentioned embodiments of the computer room energy-saving control method based on cloud computing are implemented, such as Figure 1 Alternatively, the processor implements the functions of the modules / units in the above-mentioned system embodiments when executing the computer program.

[0081] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0082] The terminal device may be a computing device such as a desktop computer, laptop, PDA, or smart tablet. The terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the aforementioned components are merely examples of terminal devices and do not constitute a limitation of the terminal device. The terminal device may include more or fewer components than those described above, or a combination of certain components or different components. For example, the terminal device may also include input / output devices, network access devices, buses, and the like.

[0083] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor. The processor is the control center of the terminal device and connects various parts of the entire terminal device using various interfaces and lines.

[0084] The memory can be used to store the computer programs and / or modules. The processor implements the various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and accessing the data stored in the memory. The memory may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or an image playback function); the data storage area may store data generated based on the use of the mobile phone (such as audio data, a phone book, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0085] If the module / unit integrated into the terminal device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0086] It should be noted that the system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the system embodiment provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0087] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A computer room energy-saving control method based on cloud computing, characterized in that: include: Obtain real-time data and historical power consumption changes of equipment in the computer room, establish a basic data set including power peak, average, and fluctuation range, and obtain the equipment operating parameter matrix; Obtaining the device load based on the device operating parameter matrix and comparing it with a preset load threshold, obtaining the device energy conversion efficiency coefficient based on the comparison result and predicting the power demand of each device, and if the predicted power exceeds the current power supply capacity, obtaining the power consumption prediction matrix and determining the optimal power allocation plan for each device based on the power consumption prediction matrix; A real-time energy scheduling system is established based on the optimal power allocation plan. If the actual power consumption of one of the devices exceeds the preset range, the operating parameters of the computer room equipment are adjusted, the energy configuration status is obtained and continuously monitored, and if abnormal fluctuations are found, the energy allocation parameters are updated and an adaptive energy consumption management decision-making mechanism is obtained.

2. The computer room energy-saving control method based on cloud computing according to claim 1 is characterized in that: The real-time data of the equipment in the computer room and the historical change trajectory of power consumption are obtained, a basic data set including power peak, average, and fluctuation range is established, and the equipment operation parameter matrix is ​​obtained, including: The power consumption, temperature distribution and load status of the equipment in the computer room are collected in real time through a multi-dimensional sensor network to obtain raw data streams covering various working conditions. The raw data streams are stored in a pre-established database in a time series format to obtain structured equipment operation records; According to the structured equipment operation record, the historical change trajectory is segmented, and the power peak, average power and fluctuation range in each time period are calculated to form first basic data; If the fluctuation amplitude exceeds a preset fluctuation threshold, the first basic data is marked as abnormal, and the operating parameters under the abnormal working condition are obtained by combining the load state and temperature distribution data; The basic data set after abnormal marking is compared with the working condition coverage requirements, and the equipment operation parameter matrix is ​​sorted out for subsequent monitoring.

3. The computer room energy-saving control method based on cloud computing according to claim 1 is characterized in that: The device load is obtained according to the device operation parameter matrix and compared with a preset load threshold, the energy conversion efficiency coefficient of the device is obtained according to the comparison result, and the power demand of each device is predicted and calculated, and if the predicted power exceeds the current power supply capacity, the power consumption prediction matrix is ​​obtained and the optimal power allocation scheme of each device is determined according to the power consumption prediction matrix, including: Based on the equipment operating parameter matrix, key feature vectors are extracted to construct a multivariate regression model between equipment status and energy consumption level. If the equipment load rate exceeds a preset load threshold, it is marked as a high energy consumption state, and the energy conversion efficiency coefficient of each type of equipment under different load conditions is determined; A device power consumption prediction model is established using the energy conversion efficiency coefficient and the power demand of each device is predicted and calculated. If the predicted power exceeds the current power supply capacity, a load balancing mechanism is triggered to obtain a power consumption prediction matrix including time and device dimensions. An energy allocation optimization model is constructed based on the power consumption prediction matrix. The objective function is set to minimize the overall energy consumption cost. The constraints include the minimum operating power requirement of the equipment and the upper limit of the power supply. The optimal power allocation plan for each device is determined by calculation.

4. The computer room energy-saving control method based on cloud computing according to claim 3 is characterized in that: The key feature vectors are extracted based on the device operation parameter matrix, and a multivariate regression model is constructed between the device state and the energy consumption level. If the device load rate exceeds a preset load threshold, it is marked as a high energy consumption state, and the energy conversion efficiency coefficient of each type of device under different load conditions is determined, including: Standardizing the power changes and operating parameters in the device operating parameter matrix, obtaining data content corresponding to the load, obtaining a second basic data set, and performing data comparison with the load threshold; If exceeded, the corresponding device state is marked as the high energy consumption state, and the marked state data group is determined; Performing hierarchical processing on the data under different load conditions to obtain the energy consumption distribution corresponding to the device state and obtain a classified efficiency data set; The efficiency data set is associated with the power change to obtain the energy conversion efficiency coefficient.

5. The computer room energy-saving control method based on cloud computing according to claim 3 is characterized in that: The device power consumption prediction model is established based on the energy conversion efficiency coefficient and the power demand of each device is predicted and calculated. If the predicted power exceeds the current power supply capacity, the load balancing mechanism is triggered to obtain a power consumption prediction matrix including time dimension and device dimension, including: Based on the energy conversion efficiency coefficient, combined with the time dimension and the equipment dimension, historical operation records are classified and sorted to obtain characteristic content related to power demand, thereby obtaining a third basic data set; By merging the third basic data set and combining the data time series characteristics, the power demand in the future period is analyzed and processed to obtain a predicted power demand distribution result; Performing data verification based on the power demand distribution result and in combination with the power supply capacity limit, triggering the load balancing mechanism if it exceeds the limit, and obtaining an adjusted power allocation strategy; According to the power allocation strategy, in combination with the time dimension and the device dimension, a prediction matrix containing multi-dimensional information is constructed to obtain the power consumption prediction matrix.

6. The computer room energy-saving control method based on cloud computing according to claim 3 is characterized in that: The energy allocation optimization model is constructed based on the power consumption prediction matrix, and the objective function is set to minimize the overall energy consumption cost. The constraints include the minimum operating power requirement of the equipment and the upper limit of the power supply. The optimal power allocation plan for each device is determined by calculation, including: Classify and process the predicted power data of each device according to the power consumption prediction matrix, obtain power demand distribution information related to the operation of the device, and obtain a classified power demand set; Based on the power demand set combined with the preset supply upper limit and power requirements, verify whether the power demand of each device matches the constraint conditions; if the predicted power of one device exceeds the power supply upper limit, adjust the power allocation ratio of the device to determine an adjusted power demand list; Processing the power allocation plan for each device through the power demand list, combining the objective function and the energy consumption cost data, and obtaining a preliminary allocation result that satisfies the minimum cost; According to the preliminary allocation result and in combination with the actual requirements of the equipment operation, a power allocation plan including the optimization of each device is generated to obtain the optimal power allocation plan.

7. The computer room energy-saving control method based on cloud computing according to any one of claims 1 to 6, characterized in that: The real-time energy scheduling system is established according to the optimal power allocation scheme, and if the actual power consumption of one device exceeds a preset range, the operating parameters of the equipment in the computer room are adjusted to obtain the energy configuration status, including: Continuously collect the actual power consumption of each device and determine whether the actual power consumption exceeds a preset range. If so, automatically adjust the operating parameters and generate an adjusted parameter configuration plan based on the requirements of intelligent management; According to the adjusted parameter configuration plan, each device is accurately allocated, the energy configuration data feedback is updated in real time, and a power allocation list that conforms to the current device status is obtained; The power distribution list is compared with the real-time monitoring data to determine the power control strategy of each device and obtain the energy configuration status.

8. The computer room energy-saving control method based on cloud computing according to any one of claims 1 to 6, characterized in that: If abnormal fluctuations are found, the energy allocation parameters are updated and an adaptive energy consumption management decision mechanism is obtained, including: If abnormal fluctuations are found, the trend of energy consumption changes is analyzed, and combined with the preset threshold range, it is determined whether the conditions for initiating emergency regulation are met and the specific status of the triggering conditions is determined; According to the specific status of the trigger condition, the allocation strategy information that matches the current fluctuation characteristics is extracted from the historical experience database, an adjusted parameter plan is generated, the updated configuration data is obtained and fed back to the monitoring process to determine whether the preset stability standard is met and obtain the final operating status.

9. The computer room energy-saving control method based on cloud computing according to any one of claims 1 to 6, characterized in that: The method further comprises: Establishing a closed-loop feedback control system based on the adaptive energy consumption management decision-making mechanism, comparing the degree of deviation between actual energy consumption and expected targets, dynamically adjusting the weight parameters of the prediction model, and determining a continuously optimized energy management and control plan; Specifically: According to the energy consumption management decision mechanism, actual energy consumption data is obtained, compared with the expected target, and whether it exceeds the preset threshold range is determined to obtain an initial deviation result. If it exceeds, real-time information is collected and combined with historical operation data to obtain an adjusted status record; Extracting a matching solution from a pre-established database according to the adjusted status record and determining an updated weight value; The energy management and control plan is dynamically adjusted through the updated weight value. For the goal of continuous optimization, the adjusted plan data is fed back to the closed-loop process to determine whether it meets the preset conditions and obtain the final configuration result.

10. A computer room energy-saving control system based on cloud computing, characterized in that: include: The detection end is used to obtain real-time data and historical changes in power consumption of equipment in the computer room, establish a basic data set including power peak, average, and fluctuation range, and obtain the equipment operation parameter matrix; The processing end obtains the device load based on the device operating parameter matrix and compares it with a preset load threshold. Based on the comparison result, the energy conversion efficiency coefficient of the device is obtained and the power demand of each device is predicted and calculated. If the predicted power exceeds the current power supply capacity, the power consumption prediction matrix is ​​obtained and the optimal power allocation plan for each device is determined based on the power consumption prediction matrix. The decision-making end is used to establish a real-time energy scheduling system based on the optimal power allocation plan. If the actual power consumption of one of the devices exceeds the preset range, the operating parameters of the computer room equipment are adjusted, the energy configuration status is obtained and continuously monitored. If abnormal fluctuations are found, the energy allocation parameters are updated and an adaptive energy consumption management decision-making mechanism is obtained.

Images