An energy-saving fan adaptive energy efficiency optimization control method and system

Abstract

The present application relates to the technical field of fan control, and provides an energy-saving fan adaptive energy efficiency optimization control method and system, which comprises the following steps: comparing actual airflow parameter information with reference airflow parameter information, and identifying abnormal sign information of a cooling area in combination with server calculation load information; when the abnormal sign information is identified and the server calculation load information does not show an increase, analyzing airflow characteristic information of the cooling area to diagnose whether the root cause of heat dissipation abnormality is calculation load change or airflow passage obstruction; according to the root cause, if the root cause is calculation load change, adjusting fan operation parameter information to match heat dissipation demand; if the root cause is airflow passage obstruction, executing non-power promotion control measures to suppress heat dissipation deterioration and outputting maintenance prompt information. The present application has the effect of improving energy efficiency utilization rate and operation stability of a fan system.

Description

Technical Field

[0001] This invention relates to the technical field of wind turbine control, and specifically to an energy-saving wind turbine adaptive energy efficiency optimization control method and system. Background Technology

[0002] In industrial automation and smart building scenarios, wind turbine systems often face energy consumption fluctuations caused by dynamic changes in operating conditions. Traditional fixed-parameter control modes struggle to adapt to real-time changes in ambient temperature, humidity, and load demands, resulting in low energy efficiency. To address this challenge, energy-saving adaptive energy efficiency optimization control methods have been introduced. These systems aim to automatically adjust wind turbine operating parameters by monitoring environmental parameters and equipment status in real time and utilizing internal energy efficiency and operating condition correlation calculation logic, in order to minimize energy consumption while meeting performance requirements. However, in practical applications, these systems sometimes encounter complex situations not fully considered in the initial design, making it difficult to accurately determine the true cause of the anomaly and leading to inappropriate control decisions that affect the system's energy efficiency and stability.

[0003] Taking an information processing center as an example, its internal server racks are densely deployed, resulting in uneven heat distribution. Frequent daily maintenance activities also easily obstruct airflow paths due to temporary physical barriers. Existing adaptive fan control systems often fail to accurately identify the root cause of anomalies when faced with localized hotspots or obstructed airflow channels. For instance, when airflow is temporarily blocked, causing a localized temperature increase, the system may mistakenly interpret this as a sharp increase in server computing load, thus blindly increasing fan power. This misjudgment not only fails to effectively solve the heat dissipation problem but also leads to significant energy waste and may cause system oscillations, accelerating mechanical wear on the fan equipment and shortening its lifespan.

[0004] More specifically, when a temporary obstacle obstructs the cooling airflow within the information processing center, the temperature sensor readings in the affected area will rise abnormally. At this point, the adaptive control system, based on preset logic, interprets the temperature increase as an increase in computational load and rapidly increases the fan speed. However, due to the physical blockage of the airflow channel, even with the fan operating at high power, the cooling air cannot effectively reach the target cabinet, resulting in unresolved localized overheating and the entire fan system entering a state of excessive energy consumption. This ineffective power increase not only wastes energy but may also disrupt airflow circulation in the surrounding area, triggering a chain reaction that causes the system to frequently switch between different power levels, generating oscillations, further exacerbating energy consumption and damaging equipment.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] This application discloses an energy-saving adaptive energy efficiency optimization control method and system for wind turbines, which aims to solve the problem that existing wind turbine adaptive control systems cannot accurately identify the root cause of abnormalities when faced with local hot spots or obstructed airflow channels, thus leading to misjudgments and inappropriate control decisions, which in turn affect the system's energy efficiency and stability.

[0007] The technical solution of this application is as follows:

[0008] In a first aspect, this application discloses an adaptive energy efficiency optimization control method for energy-saving fans, comprising the following steps: acquiring fan operating parameter information, actual airflow parameter information of the cooling area, and server computing load information; determining the baseline airflow parameter information of the cooling area under preset operating conditions based on the fan operating parameter information; comparing the actual airflow parameter information with the baseline airflow parameter information, and identifying abnormal signs in the cooling area in conjunction with the server computing load information; when abnormal signs are identified and the server computing load information does not show an increase, analyzing the airflow characteristics of the cooling area to diagnose whether the root cause of the heat dissipation abnormality is a change in computing load or obstruction of the airflow channel; depending on the root cause, if the root cause is a change in computing load, adjusting the fan operating parameter information to match the heat dissipation requirements; if the root cause is obstruction of the airflow channel, implementing non-power boost control measures to suppress heat dissipation deterioration and outputting maintenance prompt information.

[0009] This technical solution effectively distinguishes whether the abnormal heat dissipation is caused by changes in computing load or by obstruction of airflow channels, thereby avoiding energy waste and system instability caused by blindly increasing fan power when airflow is obstructed in traditional systems, and achieving more precise energy efficiency optimization control.

[0010] Secondly, this application also discloses an energy-saving fan adaptive energy efficiency optimization control system for performing energy-saving fan adaptive energy efficiency optimization control, including: a parameter information acquisition module for acquiring fan operating parameter information, actual airflow parameter information of the cooling area, and server computing load information; an airflow parameter determination module for determining the baseline airflow parameter information of the cooling area under preset operating conditions based on the fan operating parameter information; an anomaly identification module for comparing the actual airflow parameter information with the baseline airflow parameter information and identifying anomaly signs in the cooling area in conjunction with the server computing load information; an airflow characteristic analysis module for analyzing the airflow characteristic information of the cooling area when anomaly signs are identified and the server computing load information does not show an increase, in order to diagnose whether the root cause of the heat dissipation anomaly is a change in computing load or obstruction of the airflow channel; and an optimization control execution module for adjusting the fan operating parameter information to match the heat dissipation requirements if the root cause is a change in computing load; and for executing non-power boost control measures to suppress heat dissipation deterioration and output maintenance prompt information if the root cause is obstruction of the airflow channel.

[0011] This application provides a system that can implement the above-mentioned method through this technical solution. Through modular design, the system can efficiently execute various control functions, thereby realizing adaptive energy efficiency optimization control of energy-saving fans and solving the problem that traditional systems cannot accurately determine the cause of abnormalities.

[0012] Beneficial Effects: The energy-saving fan adaptive energy efficiency optimization control method disclosed in this application acquires fan operating parameters, actual airflow parameters in the cooling area, and server computational load information. Based on this information, it determines benchmark airflow parameters and compares the actual airflow parameters with the benchmark airflow parameters. Combined with server computational load information, it identifies abnormal signs in the cooling area. When abnormal signs are identified but the server computational load has not increased, the method further analyzes the airflow characteristics of the cooling area to diagnose whether the root cause of the heat dissipation anomaly is a change in computational load or obstruction of the airflow path. Based on the diagnosis results, if it is a change in computational load, the fan operating parameters are adjusted to match the heat dissipation demand; if it is an obstruction of the airflow path, non-power-boosting control measures are implemented and a maintenance prompt is output.

[0013] Through the above technical solution, this application effectively solves the problem in existing wind turbine adaptive control systems that, when faced with local hotspots or obstructed airflow channels, cannot accurately identify the root cause of the anomaly, leading to misjudgments and blindly increasing wind turbine power. This method can accurately distinguish the type of heat dissipation anomaly, avoiding ineffective energy waste and system oscillation when airflow channels are obstructed, significantly improving the energy efficiency and operational stability of the wind turbine system. Compared with traditional fixed parameter control or simple adaptive control, the method of this application can respond more intelligently and precisely to complex changes in operating conditions, thereby minimizing energy consumption, extending equipment lifespan, and providing timely and accurate maintenance prompts for operation and maintenance personnel, resulting in significant economic and environmental benefits. Attached Figure Description

[0014] Figure 1 This is a flowchart of an energy-saving adaptive energy efficiency optimization control method for a wind turbine, as described in one embodiment of the present invention.

[0015] Figure 2 This is a flowchart of an energy-saving adaptive energy efficiency optimization control method for a wind turbine, as described in another embodiment of the present invention.

[0016] Figure 3 This is a block diagram of an energy-saving fan adaptive energy efficiency optimization control system according to another embodiment of the present invention;

[0017] Explanation of reference numerals in the attached figures:

[0018] 1. Energy-saving fan adaptive energy efficiency optimization control system; 11. Parameter information acquisition module; 12. Airflow parameter determination module; 13. Abnormal sign identification module; 14. Airflow characteristic analysis module; 15. Optimized control execution module. Detailed Implementation

[0019] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] This application proposes an adaptive energy efficiency optimization control method for energy-saving wind turbines, combining... Figure 1 As shown, it includes:

[0022] S1, obtain fan operating parameter information, actual airflow parameter information of the cooling area, and server calculation load information;

[0023] S2, based on the fan operating parameter information, determine the reference airflow parameter information of the cooling area under the preset operating conditions;

[0024] S3 compares the actual airflow parameter information with the reference airflow parameter information, and combines the server's calculated load information to identify abnormal signs in the cooling area;

[0025] S4. When abnormal signs are detected and the server's computing load information does not show an increase, analyze the airflow characteristics of the cooling area to diagnose whether the root cause of the heat dissipation abnormality is a change in computing load or obstruction of airflow channels.

[0026] S5. Based on the root cause, if the root cause is a change in the calculated load, adjust the fan operating parameters to match the heat dissipation requirements; if the root cause is an obstruction in the airflow channel, implement non-power boost control measures to suppress heat dissipation deterioration and output maintenance prompt information.

[0027] To better understand the energy-saving adaptive energy efficiency optimization control method for wind turbines proposed in this application, the key terms involved and their application environments are explained. Wind turbine operating parameter information is used to characterize the wind turbine's operating state and typically includes at least one or more parameters such as wind turbine speed, power, current, and voltage. These parameters directly reflect the wind turbine's air delivery capacity and energy consumption level under current operating conditions. By continuously acquiring wind turbine operating parameter information, the energy efficiency changes of the wind turbine at different operating stages can be understood, providing a basis for subsequent regulation and control.

[0028] The actual airflow parameters of the cooling zone refer to the real-time airflow-related state data acquired by multiple sensors deployed within the cooling zone. This information includes at least one or more of the following: supply air temperature, return air temperature, local wind speed, and regional pressure difference. These parameters reflect the airflow organization and heat dissipation effect within the cooling zone. They characterize whether the cooling airflow follows the designed path and expected intensity, and are crucial for judging heat dissipation efficiency.

[0029] Server compute load information is used to characterize the actual operating load of servers or computing devices within a cooling zone. It typically includes parameters such as CPU utilization, memory utilization, and storage or network I / O throughput. Server compute load information is highly correlated with server heat generation. By obtaining this information, it's possible to distinguish whether abnormal airflow is caused by changes in compute load, thus avoiding misjudgments that might result from relying solely on temperature or wind speed information.

[0030] Benchmark airflow parameters refer to the reference airflow parameter levels that the cooling zone should achieve under normal operating conditions, given fan operating parameters and preset environmental conditions. This benchmark airflow parameter information can be generated through historical operating data statistics, simulation model calculations, or preset rule tables, and serves as a comparison benchmark for actual airflow parameter information. By introducing benchmark airflow parameter information, the system gains an objective reference for judging "deviations from normal conditions."

[0031] Anomaly indicator information is used to characterize potential heat dissipation anomalies or operational risks in the cooling area. Conditions for its formation include, but are not limited to, a persistent or significant deviation between actual and baseline airflow parameters, or abnormal fluctuations in server computing load information that do not match changes in airflow parameters. Identification of anomaly indicator information provides triggering conditions for subsequent cause analysis and control strategy selection.

[0032] This method is typically deployed in HVAC systems of data centers, smart factories, or large buildings, and is implemented by integrating fan operation sensors, airflow and temperature sensors, server load monitoring interfaces, and a centralized control unit. Various sensors are responsible for real-time acquisition of fan operating status, airflow status in cooling areas, and server load status. The collected data is periodically transmitted to the control unit, which performs analysis, judgment, and control decisions, and outputs control commands to the fan actuators, thereby achieving adaptive adjustment of fan operation.

[0033] In the specific implementation process, the following steps are taken: first, acquire fan operating parameter information, actual airflow parameter information of the cooling area, and server computing load information. Fan operating parameter information can be obtained through speed sensors and electrical parameter acquisition modules installed on the fan itself; actual airflow parameter information of the cooling area can be obtained through temperature, wind speed, and differential pressure sensors deployed at the air inlets, outlets, or cold aisles of the server rack; server computing load information can be obtained by establishing a communication interface with the server management system or virtualization platform. The above information is collected synchronously or asynchronously at a preset sampling period and aggregated to the central control unit.

[0034] Subsequently, based on the acquired fan operating parameter information and combined with preset operating condition information, the corresponding baseline airflow parameter information is determined. This determination process can be based on a pre-established physical model, data-driven model, or operating condition mapping table, and is used to predict the normal return air temperature range, wind speed distribution level, or pressure difference range that the cooling area should reach under the current fan operating parameters and environmental conditions, thereby forming a comparable reference state.

[0035] Based on this, the actual airflow parameters are compared with the baseline airflow parameters, and server computational load information is used to identify any signs of anomalies. When the actual airflow parameters deviate from the baseline airflow parameters by more than a preset threshold, and this deviation is persistent over time, the system will further cross-validate the data using server computational load information. If the server computational load information does not show a corresponding increase, it indicates that the current airflow anomaly is not directly caused by changes in computational load, thus triggering further analysis of the airflow organization.

[0036] When anomalies are detected but the server's computational load does not show a significant increase, the system analyzes the airflow characteristics of the cooling area to determine whether the root cause of the heat dissipation anomaly is a change in computational load or obstruction of airflow channels. This analysis is achieved by comprehensively assessing temperature change trends, wind speed unevenness, and regional pressure difference changes at different locations. For example, if a localized area experiences a sustained increase in temperature, a significant decrease in wind speed, and a stable server load, it can be determined that the area may have issues such as obstructed airflow channels, airflow short-circuiting, or localized shading.

[0037] Finally, corresponding optimization control measures are implemented based on the root cause identified in the diagnosis. When the diagnostic results indicate that the change in heat dissipation demand is mainly caused by changes in server computing load, the system appropriately increases the fan speed or air delivery capacity by adjusting the fan operating parameters to match the new heat dissipation demand. When the diagnostic results indicate that the abnormal heat dissipation is mainly caused by obstructed airflow channels, the system avoids directly increasing the fan power and instead prioritizes non-power-boosting control strategies. For example, it coordinates and adjusts the operating parameters of fans in other areas or outputs maintenance prompts to instruct maintenance personnel to inspect and clean the airflow channels, thereby achieving effective energy consumption control while ensuring heat dissipation safety.

[0038] Optionally, steps to diagnose the root cause of abnormal heat dissipation that falls under the category of changes in computational load or obstruction of airflow channels include:

[0039] Acquire temperature data for the cooling zone;

[0040] Calculate the average rate of change of temperature data within a preset time window to obtain the average temperature change rate; monitor the average temperature change rate of the hot air channel.

[0041] When the average temperature change rate of the hot air channel continues to be higher than the preset threshold or when the deviation of the actual airflow parameter information from the reference airflow parameter information continues to increase, a slow change anomaly sign information is generated.

[0042] Obtain the average calculated load change information of all racks in the cooling area within a preset time window;

[0043] When a slow change anomaly is detected and the average calculated load change information does not show an increase, a fine-tuning command is sent to the wind turbine controller to slightly increase the wind turbine speed and maintain it for a preset duration.

[0044] Within the monitoring time window after the fan speed is fine-tuned, the average temperature change rate of the hot air channel is monitored and the unit power air volume information and corresponding ratio information before and after the fan adjustment are calculated.

[0045] Based on the average temperature change rate, unit power air volume information and corresponding ratio information, the root cause is diagnosed as the accumulation of subtle fluctuations in the calculated load or the increase in distributed airflow resistance.

[0046] When the root cause is the accumulation of minor fluctuations in the computational load, readjust the fan speed to match the heat dissipation requirements; when the root cause is the increase in distributed airflow resistance, stop increasing the fan power and output a maintenance alarm message.

[0047] Acquiring temperature data for the cooling area involves continuously and in real-time collecting temperature data at various locations within the cooling area using multiple temperature sensors deployed at different spatial positions. This data forms a set of temperature data reflecting the overall and local thermal state of the cooling area. This temperature data is not only used to characterize the current thermal state but also for subsequent analysis of temperature change trends, serving as a crucial basis for determining whether heat dissipation capacity is gradually deteriorating.

[0048] Based on this, the temperature data is processed within a preset time window, and the average rate of change of the temperature data within that time window is calculated to obtain the average temperature change rate. Simultaneously, the average temperature change rate of the hot air channel is monitored. By introducing a time window and a rate of change calculation method, the system focuses on the evolution trend of temperature over time, rather than relying solely on instantaneous temperature values. This allows it to identify slowly accumulating and gradually worsening heat dissipation anomalies. For example, the average rate of change can be calculated using linear regression, moving difference, or weighted average methods for the temperature series within the time window to reduce the interference of occasional fluctuations on the judgment results.

[0049] When the average temperature change rate of the hot air duct remains above a preset threshold for multiple consecutive time windows, or when the deviation of the actual airflow parameters from the baseline airflow parameters shows a continuously increasing trend, the system generates a gradual anomaly indicator. This gradual anomaly indicator is used to characterize a heat dissipation problem that is not an instantaneous occurrence but rather a state of gradual accumulation or deterioration over time. The corresponding preset threshold can be set based on historical operating data, system stability requirements, and safety margins. Simultaneously, the system acquires the average computational load change information of all cabinets within the cooling area within the same preset time window to determine whether the overall computational load change is likely the primary cause of the temperature change.

[0050] Upon detecting signs of a gradual anomaly and without a significant increase in the average calculated load change, the system sends a fine-tuning command to the fan controller to slightly increase the fan speed and maintain it for a preset duration. This fine-tuning is not intended to directly eliminate the anomaly, but rather a controlled probing intervention. By applying a small disturbance to the fan speed, an observable change in airflow is introduced without significantly increasing overall energy consumption, thus testing the cooling system's response to changes in airflow. The preset duration ensures that the system obtains sufficient temperature and airflow response data after the fan speed change, thereby avoiding misjudgments due to response lag.

[0051] Within the monitoring time window after the fan speed is fine-tuned, the system continuously monitors the average temperature change rate of the hot air channel and calculates the unit power airflow rate before and after the fan adjustment, along with their corresponding ratios. The unit power airflow rate is used to characterize the effective airflow capacity generated by the fan under unit energy consumption conditions; its changes reflect changes in fan efficiency and airflow channel resistance. By comparing the average temperature change rate and the unit power airflow rate ratio before and after the fine-tuning, the system status can be evaluated simultaneously from both thermal response and energy efficiency response dimensions.

[0052] Based on the aforementioned average temperature change rate, unit power airflow rate, and corresponding ratio information, the system diagnoses the root cause of heat dissipation anomalies, determining whether it stems from the accumulation of minor fluctuations in computational load or an increase in distributed airflow resistance. Specifically, when the average temperature change rate of the hot air channel rapidly decreases and stabilizes after a slight adjustment of the fan speed, and the unit power airflow rate ratio remains stable or slightly increases, it indicates that the system has a good response capability to changes in airflow, and the diagnosis is likely due to the accumulation of minor fluctuations in computational load over time. Conversely, when the temperature change rate does not decrease significantly, or decreases only to a limited extent, and the unit power airflow rate ratio decreases significantly, it indicates that airflow efficiency is impaired, and the diagnosis is likely due to an increase in distributed airflow resistance.

[0053] When the diagnostic results indicate that the root cause is the accumulation of subtle fluctuations in the computing load, the system readjusts the fan speed to further match the current heat dissipation requirements, thereby optimizing energy efficiency while ensuring cooling performance. When the diagnostic results indicate that the root cause is increased distributed airflow resistance, the system stops increasing the fan power and outputs a maintenance alarm message to prompt maintenance personnel to conduct physical inspections and repairs of the cooling area, so as to avoid blindly increasing the fan power under obstructed airflow conditions, which would lead to increased energy consumption and limited improvement in heat dissipation.

[0054] In some preferred embodiments, assuming that the average temperature change rate of the hot air aisle in a data center's cooling area has been slowly increasing over the past few hours and has exceeded a preset threshold, but the system does not detect a significant increase in the average computing load of all racks within the cooling area, the system identifies this slow-change anomaly. According to the solution of this application, the system first sends a fine-tuning command to the fan controller, for example, slightly increasing the fan speed by 2% and maintaining that speed for 30 minutes. During this monitoring time window, the system continuously monitors the average temperature change rate of the hot air aisle and the airflow per unit power.

[0055] If, within this time window, the average temperature change rate of the hot air channel rapidly decreases and stabilizes, and the ratio of the unit power airflow after fan adjustment to the value before adjustment remains above the preset threshold, the system diagnoses the root cause as the accumulation of minor fluctuations in the computational load and further adjusts the fan speed accordingly. If, within this time window, the temperature change rate does not decrease significantly, or the unit power airflow ratio decreases significantly, the system diagnoses the root cause as increased distributed airflow resistance and stops further increasing the fan power. Simultaneously, a maintenance alarm is sent to the operation and maintenance platform to avoid energy waste and promptly eliminate potential physical heat dissipation hazards.

[0056] Furthermore, in some embodiments of this application, it can be recognized that relying solely on the overall average computational load change information of the cooling area may not be able to reflect local computational load fluctuations or local temperature anomalies in a timely manner. Therefore, the aforementioned probing disturbance and response analysis mechanism introduced through fan fine-tuning can indirectly reveal local airflow organization anomalies or distributed resistance changes without significantly increasing system complexity. This allows for early identification of potential local heat dissipation risks under stable overall load conditions, avoiding unnecessary energy consumption caused by excessively increasing fan speed.

[0057] Optional, combined Figure 2 As shown, the steps for slightly increasing the fan speed and maintaining it for a preset duration include:

[0058] A1, obtain the sub-regional computational load information of each sub-region within the cooling area;

[0059] A2, calculate load information based on sub-regions, and evaluate the load change trend information of each sub-region;

[0060] A3. When a slow change anomaly is detected and the average calculated load change information does not show an increase, if the load change trend information shows that a sub-region has exceeded the limit within a preset time, the fan speed fine-tuning will be suspended and the load tracking monitoring process will be entered.

[0061] A4. If the load change trend information shows that the calculated load of each sub-area is in a stable state, then determine the fan speed increase range based on the overall heat dissipation demand of the cooling area, and send the fine-tuning instruction information corresponding to the fan speed increase range to the fan controller.

[0062] A5 continuously monitors the temperature change rate of each sub-region during the fine-tuning of the fan speed. Based on the temperature change rate of each sub-region, when the temperature rise rate of any sub-region exceeds the preset warning threshold, the fan speed increase is stopped.

[0063] Specifically, obtaining sub-regional computing load information within a cooling area involves dividing the entire cooling area into several logical or physical sub-regions and acquiring the computing load of servers or devices within each sub-region. For example, a data center can be divided into multiple rack rows, a single rack, or even specific unit areas within a rack. The computing load information of each sub-region can be comprehensively evaluated using metrics such as CPU utilization, memory usage, and network I / O of the servers within that area. Furthermore, assessing the load change trend of each sub-region based on its computing load information can be understood as analyzing the changes in the computing load of each sub-region over a period of time to determine whether it is in a stable state, slowly increasing, rapidly increasing, or decreasing state. For example, the trend can be quantified by calculating the average rate of change, standard deviation, or trend line slope of the sub-region's computing load within a preset time window.

[0064] In practical applications, when a slow-change anomaly is detected and the average computed load change information does not show an increase, if the load change trend information indicates that a sub-region's load increases beyond the preset limit within a certain time, the fan speed fine-tuning is paused and the load tracking monitoring process begins. This means that even if the overall average load does not increase significantly, if the load in a certain local area suddenly or continuously increases rapidly, the system will prioritize handling this local hotspot risk rather than blindly fine-tuning the overall fan speed. The load tracking monitoring process may include collecting load and temperature data for that sub-region more frequently and may trigger local heat dissipation optimization strategies for that sub-region. Furthermore, if the load change trend information shows that the computed load of each sub-region is stable, the fan speed increase range is determined based on the overall heat dissipation requirements of the cooling area, and a fine-tuning instruction corresponding to the fan speed increase range is sent to the fan controller. This indicates that, in the absence of local hotspot risk, the system will increase the fan speed by a more reasonable and precise range based on the heat dissipation requirements of the entire cooling area, avoiding excessive increases that lead to unnecessary energy consumption. The fan speed increase range can be dynamically calculated based on historical data, predictive models, or real-time heat dissipation requirements. Furthermore, during the fan speed fine-tuning process, the temperature change rate of each sub-zone is continuously monitored. Based on the temperature change rate of each sub-zone, the fan speed increase is stopped when the temperature rise rate of any sub-zone exceeds a preset warning threshold. This measure aims to provide a real-time safety assurance mechanism. Even if the load of each sub-zone is stable at the start of fine-tuning, unexpected rapid increases in local temperature may still occur during the fine-tuning process. By continuously monitoring and setting warning thresholds, inappropriate speed increases can be stopped in a timely manner, preventing local overheating and ensuring the safe operation of the system.

[0065] Optionally, the step of monitoring the average temperature change rate of the hot air passage includes:

[0066] Acquire data from various temperature sensors within the hot air duct;

[0067] Time compensation and spatial correction are performed on the data from each temperature sensor to obtain corrected temperature data;

[0068] Based on the corrected temperature data, calculate the local temperature gradient information and the local temperature change rate information;

[0069] The local temperature change rate information is compared with the overall average temperature change rate of the hot air channel to identify local hot spots.

[0070] When local hot spots exist, analyze the rate of temperature change in the local hot spots and determine the response of the local hot spots to fine-tuning of the fan speed.

[0071] Specifically, acquiring temperature sensor data within the hot air channel refers to collecting real-time temperature readings at various points using multiple temperature sensors deployed at different locations within the hot air channel. These sensors can include thermocouples, thermistors, or infrared temperature sensors, and their deployment locations should cover key areas of the hot air channel to provide comprehensive temperature distribution information.

[0072] The process involves performing time compensation and spatial correction on the data from each temperature sensor to obtain corrected temperature data. This can be understood as preprocessing the raw sensor data. Time compensation aims to eliminate errors caused by inconsistencies in the data acquisition times of different sensors, ensuring that all data are analyzed on the same time reference. Spatial correction aims to eliminate the influence of sensor installation location, environmental differences, and the sensor's own accuracy deviations on temperature readings. By establishing a mapping relationship between physical location and temperature readings, temperature data from different locations are unified to a comparable benchmark, thereby obtaining more accurate and representative temperature data.

[0073] In practical applications, calculating local temperature gradient information and local temperature change rate information based on corrected temperature data refers to using the corrected temperature data, through methods such as spatial interpolation or differential calculation, to evaluate the rate of temperature change in any local area within the hot air channel in space (temperature gradient) and in time (temperature change rate). Temperature gradient information can reveal the uneven distribution of heat in space, while temperature change rate information reflects the dynamic evolution of the thermal state in a local area.

[0074] Furthermore, the rate of local temperature change is compared with the overall average rate of temperature change in the hot air channel to identify local hotspots. This aims to distinguish between a general temperature increase and a localized abnormal temperature rise. When the rate of temperature change in a certain local area is significantly higher than the overall average rate of change in the hot air channel, it indicates that there may be a problem of excessively rapid heat accumulation or insufficient heat dissipation in that area, thus it is identified as a local hotspot.

[0075] Furthermore, when localized hotspot areas exist, analyzing the rate of temperature change in these areas and assessing their response to minor adjustments in fan speed aims to diagnose the root cause of the hotspot formation. By observing the speed and trend of temperature change in localized hotspot areas after minor adjustments to fan speed, the impact of fan adjustments on the heat dissipation of that area can be evaluated. For example, if the temperature in the hotspot area drops rapidly after adjustment, it may indicate insufficient airflow; if the temperature drops slowly or shows no significant change, it may point to deeper-level airflow obstruction or abnormal localized heat sources.

[0076] Optionally, the step of monitoring the average temperature change rate of the hot air passage includes:

[0077] Acquire data from various temperature sensors within the hot air channel, as well as real-time computing load information from each server;

[0078] Based on real-time load information, identify server information with high dynamic or sudden loads within the hot air channel;

[0079] When server information exists, estimate the heat change caused by load fluctuations based on temperature sensor data in the area where the server information is located.

[0080] The estimated heat change is subtracted from the temperature sensor data to obtain the load decoupling temperature data;

[0081] The average temperature change rate of the hot air duct is calculated based on the load decoupling temperature data.

[0082] Specifically, acquiring data from various temperature sensors within the hot air aisle refers to collecting real-time temperature readings at different locations using multiple temperature sensors deployed within the hot air aisle. These temperature sensors can include thermistors, thermocouples, or infrared temperature sensors, and their purpose is to provide fine-grained temperature distribution information. Meanwhile, real-time computing load information for each server refers to data obtained from the data center management system or the server's own monitoring agent, reflecting indicators such as current CPU utilization, memory usage, and I / O throughput.

[0083] Identifying servers with high dynamic or sudden loads within the hot air aisle can be understood as analyzing historical or real-time load data to identify servers with high frequency of load changes, large fluctuations, or instantaneous peaks. For example, a load change rate threshold or fluctuation variance threshold can be set; servers whose load changes exceed this threshold are marked as having high dynamic or sudden loads. The aim is to focus on the load sources that have the greatest impact on temperature data.

[0084] In practical applications, estimating the heat change caused by load fluctuations involves calculating the value based on the real-time load changes of identified servers with highly dynamic or sudden loads, combined with a pre-established server power consumption model or heat generation model. For example, a mapping relationship between server load and heat generation can be established through experiments or simulations. When the load changes, the corresponding heat change can be estimated based on this relationship. The purpose is to quantify the impact of load fluctuations on local temperature.

[0085] Furthermore, subtracting the estimated heat change from the temperature sensor data to obtain load decoupled temperature data involves subtracting the temperature increment corresponding to the heat change caused by load fluctuations from the original temperature sensor reading. For example, if the temperature sensor reading for a server area is T_raw, and the estimated heat change ΔQ caused by load fluctuations in that area is calculated, this can be converted into a temperature change ΔT_load using heat capacity and airflow models. Therefore, the load decoupled temperature data T_decoupled = T_raw - ΔT_load. The purpose is to eliminate the interference of load fluctuations on the temperature data, allowing the remaining temperature change to more accurately reflect airflow characteristics.

[0086] Therefore, calculating the average temperature change rate of the hot air channel based on load decoupling temperature data refers to calculating the average change rate of the temperature data after load decoupling within a preset time window. For example, methods such as moving average and exponential smoothing can be used to process multiple load decoupling temperature data points to obtain a more stable and accurate average temperature change rate. The aim is to provide a temperature change indicator that reflects the airflow channel conditions and is unaffected by instantaneous fluctuations in server load.

[0087] Optionally, the steps for calculating the average temperature change rate of the hot air passage include:

[0088] Acquire load decoupling temperature data for each heat dissipation area within the hot air channel;

[0089] Based on the physical dimensions, heat capacity, and airflow path information of each heat dissipation area, determine the thermal inertia parameters.

[0090] Based on thermal inertia parameter information, adjust the time window length and weight of each heat dissipation area;

[0091] Calculate the average temperature change rate of each heat dissipation area based on the adjusted time window length and weight;

[0092] The average temperature change rate of the hot air channel is calculated based on the average temperature change rate of each heat dissipation area.

[0093] Specifically, obtaining load-decoupled temperature data for each heat dissipation area within the hot air aisle refers to dividing the entire hot air aisle into several logical or physical heat dissipation areas and independently acquiring the temperature data for each area after load decoupling. These heat dissipation areas can be divided based on factors such as rack layout, equipment type, and airflow organization.

[0094] The process involves determining thermal inertia parameters based on the physical dimensions, heat capacity, and airflow path information of each heat dissipation area. This can be understood as modeling the inherent thermal characteristics of each heat dissipation area. Physical dimensions include the area's volume and surface area; heat capacity reflects the ability of the equipment and air within the area to absorb or release heat; and airflow path information describes the flow characteristics of the airflow within the area, such as velocity, direction, and the presence of obstructions. These factors collectively determine the speed and magnitude of the area's response to temperature changes, i.e., its thermal inertia. For example, an area containing many high-heat-capacity devices will experience relatively slow temperature changes and have high thermal inertia; conversely, an area with fast airflow and fewer devices will experience relatively rapid temperature changes and have low thermal inertia.

[0095] In practical applications, the time window length and weight of each heat dissipation region are adjusted based on thermal inertia parameter information. The aim is to make the temperature change rate calculation for each region more consistent with its own thermal response characteristics. For regions with high thermal inertia, a longer time window can be used to smooth short-term fluctuations and obtain a more stable average temperature change rate; for regions with low thermal inertia, a shorter time window can be used to improve sensitivity to rapid temperature changes. Furthermore, different weights can be assigned to each region based on its importance to overall heat dissipation or its thermal inertia, so that its differentiated contribution can be reflected in the subsequent calculation of the overall average temperature change rate.

[0096] Furthermore, based on the adjusted time window length and weight, the average temperature change rate of each heat dissipation area is calculated. This means that the load decoupled temperature data of each heat dissipation area is averaged within its own time window according to its own weight, thereby obtaining a more accurate and representative average temperature change rate for that area.

[0097] Therefore, the average temperature change rate of the hot air channel is calculated based on the average temperature change rate of each heat dissipation area. The purpose is to obtain a more comprehensive and accurate average temperature change rate that reflects the heat dissipation status of the entire hot air channel by comprehensively considering the independent thermal response of each sub-region.

[0098] Optionally, the steps for calculating the average temperature change rate of the hot air passage include:

[0099] Obtain the average temperature change rate of each heat dissipation area;

[0100] Identify information regarding fluctuations in the average temperature change rate;

[0101] Based on the fluctuation information, the average temperature change rate of each heat dissipation area is smoothed to suppress the impact of short-term drastic fluctuations.

[0102] Based on the physical location information, airflow direction information, and heat transfer path information of each heat dissipation area, determine the degree of mutual influence between heat dissipation areas;

[0103] The weights of the average temperature change rate of each heat dissipation area are adjusted based on the information on the degree of mutual influence.

[0104] The average temperature change rate of each heat dissipation area after smoothing is weighted and averaged according to the adjusted weights to obtain the overall average temperature change rate of the hot air channel.

[0105] Specifically, after obtaining the average temperature change rate for each heat dissipation area, it is necessary to identify the fluctuation information of these average temperature change rates. This fluctuation information can refer to drastic fluctuations or abnormal peaks in the temperature change rate over a short period of time. Identifying fluctuation information can be achieved through statistical analysis methods, such as calculating the standard deviation and variance, or setting dynamic thresholds to detect changes exceeding the normal range. The purpose is to quantify and locate instability in the data.

[0106] Furthermore, based on the identified fluctuation information, the average temperature change rate of each heat dissipation area is smoothed to suppress the impact of short-term, drastic fluctuations. Various techniques can be employed for smoothing, such as moving average filtering, exponential smoothing, and Kalman filtering. For example, a moving average value within a preset time window can be used instead of the instantaneous value, thereby eliminating the interference of instantaneous noise on the calculation of the average temperature change rate and making the data trend clearer and more stable.

[0107] Furthermore, based on the physical location information, airflow direction information, and heat transfer path information of each heat dissipation area, the degree of mutual influence between heat dissipation areas is determined. Physical location information includes the spatial coordinates of each cabinet, device, and temporary obstacle; airflow direction information can be obtained through airflow sensors or computational fluid dynamics (CFD) simulations; and heat transfer path information describes how heat is conducted, convectioned, or radiated between different areas. By comprehensively analyzing this information, a thermal coupling model can be constructed to quantify the impact of temperature changes in one area on temperature changes in other areas. For example, heat from an upstream area may directly affect the temperature of a downstream area through airflow.

[0108] Based on information about the degree of mutual influence, the weights of the average temperature change rate of each heat dissipation region are adjusted. Traditional weighted averages may only be based on region size or equipment density, while the scheme in this application considers the actual thermodynamic coupling between regions. For example, if one region has a significant thermal impact on another region, the weights of the two regions may be adjusted in tandem when calculating the overall average temperature change rate, or the weights of the affected regions may be corrected according to the degree of their impact to more accurately reflect the overall thermal state.

[0109] Finally, the average temperature change rate of each heat dissipation area after smoothing is weighted and averaged according to the adjusted weights to obtain the overall average temperature change rate of the hot air channel. This weighted averaging method not only considers the temperature change trend of each area itself, but also incorporates the thermodynamic interactions between areas, making the final overall average temperature change rate more accurate and representative.

[0110] Optionally, the steps for determining the degree of mutual influence between heat dissipation areas include:

[0111] Obtain spatial location information of each cabinet, equipment, and temporary obstacle within the cooling area;

[0112] Based on spatial location information, construct the three-dimensional physical layout information of the cooling area;

[0113] Obtain airflow velocity and direction information for each heat dissipation area;

[0114] Obtain temperature distribution information along the heat transfer path;

[0115] Based on three-dimensional physical layout information, airflow velocity and direction information, and temperature distribution information, the airflow path information and heat transfer path information of the cooling area are simulated.

[0116] Based on the simulated airflow path information and heat transfer path information, the degree of mutual influence between each heat dissipation area is determined.

[0117] Specifically, acquiring the spatial location information of each rack, device, and temporary obstacle within the cooling area refers to obtaining the precise three-dimensional coordinates and dimensions of all fixed and non-fixed physical entities (such as server racks, network equipment, power modules, and temporarily stacked items) within the data center cooling area using sensors (e.g., LiDAR, visual sensors, or ultrasonic sensors) or pre-defined layout drawings. This information forms the basis for constructing the physical model of the cooling area. Constructing the three-dimensional physical layout information of the cooling area based on spatial location information can be understood as using computer-aided design (CAD) software or 3D modeling tools to transform the acquired spatial location information into a digital three-dimensional model. This model can intuitively display the relative positions, sizes, and shapes of all physical entities within the cooling area, providing precise geometric boundary conditions for subsequent airflow and heat simulation. In practical applications, acquiring the airflow velocity and direction information of each heat dissipation area involves deploying airflow sensors (e.g., hot-wire anemometers, Pitot tubes, etc.) within the cooling area or using computational fluid dynamics (CFD) simulation software to measure and predict the airflow velocity vectors of each heat dissipation area (e.g., inside racks, cold aisles, hot aisles, etc.) in real time or periodically. This information is crucial for understanding airflow patterns within the cooling zone. Furthermore, obtaining temperature distribution information along the heat transfer path involves continuously monitoring and recording temperature field data within the cooling zone by deploying temperature sensors in key hotspot areas, along airflow paths, and on equipment surfaces. This data reflects heat transfer between different areas and is an important basis for evaluating heat dissipation efficiency and identifying potential hotspots. Therefore, based on three-dimensional physical layout information, airflow velocity and direction information, and temperature distribution information, the airflow path and heat transfer path information of the cooling zone can be simulated using computational fluid dynamics (CFD) and computational heat transfer (CHT) simulation techniques. This simulation process uses the obtained three-dimensional physical layout, airflow velocity and direction, and temperature distribution as input, and by solving the fluid dynamics and energy conservation equations, predicts the detailed airflow trajectory within the cooling zone and the heat transfer path from the heat source to the cooling medium. Its purpose is to reveal complex airflow patterns and heat distribution laws that are difficult to observe with the naked eye. Finally, based on the simulated airflow and heat transfer path information, the degree of mutual influence between the heat dissipation areas is determined. This involves quantifying the interactions between different heat dissipation areas through airflow and heat conduction / convection by analyzing the simulation results. For example, it's possible to calculate how much heat from one area is carried away by the airflow from another area, or how a temperature change in one area affects the temperature of adjacent areas. This degree of influence can be expressed as a coupling coefficient, heat exchange rate, or temperature gradient influence factor, etc.

[0118] Optionally, the steps for simulating airflow path information and heat transfer path information in the cooling zone include:

[0119] Acquire raw data from visual sensors, LiDAR, and thermal imagers;

[0120] The raw data is preprocessed to obtain preprocessed data; the preprocessing includes timestamp alignment, spatial coordinate system transformation, and data format unification.

[0121] Data quality assessment is performed on the preprocessed data to identify outliers or missing regions and obtain the corresponding data quality assessment results.

[0122] Based on the data quality assessment results, weighted fusion processing is performed on the preprocessed data;

[0123] The fused data is continuously compared with the preset layout benchmark model information to identify layout deviation information, and the fused data is corrected based on the layout deviation information.

[0124] Based on the corrected fusion data, the three-dimensional physical layout information is updated, and based on the updated three-dimensional physical layout information, the corresponding airflow path information and heat transfer path information are generated.

[0125] Raw data refers to unprocessed initial data acquired from various sensors. For example, visual sensors can provide images or video information of the cooling area to identify equipment location and obstacles; LiDAR can provide high-precision 3D point cloud data to accurately depict the structure and dimensions of the physical space; and thermal imagers can capture the temperature distribution within the area, revealing hotspots. These multi-source data collectively provide comprehensive and multi-dimensional input for subsequent simulation analysis.

[0126] Furthermore, preprocessing aims to eliminate noise, redundancy, and inconsistencies in the raw data, ensuring its accuracy and usability. Specifically, timestamp alignment ensures that data collected by different sensors at the same time can be correctly correlated, thus guaranteeing data temporal synchronization; spatial coordinate system transformation unifies data from different sensors into the same reference coordinate system for spatial fusion and analysis, eliminating coordinate inconsistencies caused by differences in sensor installation locations; and data format unification ensures compatibility of different data types in the processing flow, facilitating subsequent automated processing.

[0127] The purpose of data quality assessment is to identify potential outliers or missing regions in the preprocessed data. For example, sensor malfunctions may lead to invalid readings, obstructions may cause missing points in the lidar point cloud, or thermal imagers may have blind spots in certain areas. Through assessment, data quality results can be obtained, which provide important basis for subsequent data weighting, fusion, and correction.

[0128] Weighted fusion processing refers to the comprehensive integration of preprocessed data based on data quality assessment results. For example, higher-quality data with higher confidence levels can be assigned greater weight, while data with anomalies or missing information can have their weight reduced or be supplemented through interpolation, prediction, or other methods to generate more comprehensive and reliable fused data. This processing method can maximize the use of effective information and reduce the impact of low-quality data on the overall simulation results.

[0129] The preset layout baseline model information can be understood as the ideal or initial physical layout model of the cooling area, which includes the geometric information of fixed elements such as racks, equipment, and aisles. Continuously comparing the fused data with this baseline model allows for the dynamic identification of deviations between the actual layout and the baseline layout, such as equipment movement, the addition of temporary obstacles, or airflow baffles. Based on the identified layout deviations, the fused data can be corrected to ensure that the physical layout information used in the simulation accurately reflects the actual state of the cooling area in real time.

[0130] Finally, based on the corrected fused data, the three-dimensional physical layout information of the cooling area can be updated in real time. Based on this updated high-precision three-dimensional physical layout information, combined with airflow velocity and direction information and temperature distribution information along the heat transfer path obtained from other sensors, more accurate airflow path information and heat transfer path information can be generated.

[0131] This application also discloses an energy-saving wind turbine adaptive energy efficiency optimization control system, used to perform energy-saving wind turbine adaptive energy efficiency optimization control, combined with... Figure 3 As shown, the energy-saving fan adaptive energy efficiency optimization control system 1 includes:

[0132] The parameter information acquisition module 11 is used to acquire fan operating parameter information, actual airflow parameter information of the cooling area, and server calculation load information.

[0133] The airflow parameter determination module 12 is used to determine the reference airflow parameter information of the cooling area under preset operating conditions based on the fan operating parameter information;

[0134] The abnormal sign identification module 13 is used to compare the actual airflow parameter information with the reference airflow parameter information, and combine the server's calculated load information to identify abnormal sign information in the cooling area.

[0135] The airflow characteristic analysis module 14 is used to analyze the airflow characteristic information of the cooling area when abnormal signs are detected and the server computing load information does not show an increase, so as to diagnose whether the root cause of the heat dissipation abnormality is a change in computing load or obstruction of airflow channels.

[0136] The optimization control execution module 15 is used to adjust the fan operating parameters to match the heat dissipation requirements if the root cause is a change in the calculated load; if the root cause is an obstruction of the airflow channel, it executes non-power boost control measures to suppress heat dissipation deterioration and outputs maintenance prompt information.

[0137] To better understand the energy-saving adaptive energy efficiency optimization control system for wind turbines proposed in this application, it is necessary to explain the key functional modules involved and their collaborative relationships. The specific meanings and acquisition methods of wind turbine operating parameters, actual airflow parameters in the cooling zone, server computational load information, baseline airflow parameters, and anomaly indications have already been described in the above embodiments, and will not be repeated here. It is important to emphasize that this system does not process the above information in isolation, but rather achieves comprehensive judgment and adaptive control of cooling status, energy efficiency changes, and the causes of anomalies through the collaborative cooperation of various functional modules.

[0138] Specifically, the parameter information acquisition module is configured to acquire fan operating parameters, actual airflow parameters in the cooling area, and server computational load information. This module provides fundamental data input for subsequent analysis and decision-making, and can acquire data through various physical or logical interfaces. For example, it can directly connect to speed sensors, power meters, temperature sensors, wind speed sensors, or server monitoring units via hardware interfaces to acquire corresponding parameter information through analog or digital signals; alternatively, it can use software to acquire relevant data from distributed sensor nodes, server management systems, or centralized data platforms via network communication protocols. In some implementations, it can also support manual input or periodic configuration updates to adapt to different deployment environments. Regardless of the implementation method, the parameter information acquisition module provides the system with real-time or near-real-time information reflecting the fan operating status, cooling area airflow status, and computational load level.

[0139] The airflow parameter determination module is configured to determine the baseline airflow parameters of the cooling zone under preset operating conditions based on the fan's operating parameters. This module is used to construct a comparative reference for the actual operating state, and its output baseline airflow parameters characterize the ideal or normal airflow state that the cooling zone should exhibit under the current fan operating conditions and environmental conditions. This module can perform calculations based on preset physical models, empirical formulas, or models trained from historical data. For example, it can determine the corresponding baseline return air temperature, wind speed distribution, or pressure difference level based on parameters such as fan speed and ambient temperature. In system implementation, this module can operate as a calculation function module in the central control unit, or as an independent calculation unit providing baseline airflow parameter information to other modules.

[0140] The anomaly detection module is configured to compare actual airflow parameters with baseline airflow parameters and, in conjunction with server load information, identify anomalies in the cooling area. This module determines whether the current cooling state deviates from the expected operating state by analyzing the deviation between actual and baseline airflow parameters to identify significant differences. Simultaneously, the module incorporates server load information to attribute the identified deviations, distinguishing whether the anomaly is directly caused by changes in the load. As one implementation method, this module can employ threshold comparison, statistical analysis, or pattern recognition to comprehensively assess the magnitude, duration, and trend of the deviation, thereby generating anomaly detection information characterizing cooling anomalies.

[0141] The airflow characteristic analysis module is configured to analyze the airflow characteristics of the cooling area when the anomaly detection module identifies anomaly information and the server's computing load information does not show an increase. This analysis aims to diagnose whether the root cause of the heat dissipation anomaly is a change in computing load or obstruction of airflow channels. This module is used for further causal analysis of anomalies, focusing not on the existence of anomalies, but on the physical mechanisms underlying their occurrence. In specific implementations, this module can analyze airflow paths, local wind speed changes, and temperature gradients based on actual airflow parameters, temperature distribution information, and their spatial variations to determine whether there are airflow short-circuit, increased distributed resistance, or local obstruction. In some implementations, this module can also combine preset models or simulation calculation results to further verify the analysis conclusions, thereby improving the reliability of the diagnostic results.

[0142] The optimized control execution module is configured to perform corresponding control operations based on the diagnosed root cause. When the diagnostic result indicates that the root cause is a change in calculated load, the optimized control execution module adjusts the fan operating parameters to match the air supply capacity with the actual heat dissipation demand. When the diagnostic result indicates that the root cause is obstructed airflow, the optimized control execution module implements non-power-boosting control measures to suppress heat dissipation deterioration and outputs maintenance prompts. This module can generate corresponding control commands based on the diagnostic results of the airflow characteristic analysis module. For example, it can send new speed or power setpoints to the fan controller via the communication interface, or output maintenance prompts and alarms to the operation and maintenance management system, and coordinate with other areas to adjust fan operating parameters as needed. Through the control decisions and execution of this module, the system can avoid blindly increasing fan power when airflow is obstructed, thereby achieving a balance between energy efficiency optimization and operational safety.

[0143] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An adaptive energy efficiency optimization control method for energy-saving fans, characterized in that, include: Acquire fan operating parameter information, actual airflow parameter information of the cooling area, and server calculation load information; Based on the fan operating parameter information, the baseline airflow parameter information of the cooling area under preset operating conditions is determined; The actual airflow parameter information is compared with the reference airflow parameter information, and the abnormal signs in the cooling area are identified by combining the server's calculated load information. When the abnormal signs are detected and the server's computing load information does not show an increase, the airflow characteristics of the cooling area are analyzed to diagnose whether the root cause of the heat dissipation abnormality is a change in computing load or obstruction of airflow channels. Based on the root cause, if the root cause is a change in the calculated load, the fan operating parameters are adjusted to match the heat dissipation requirements; if the root cause is obstruction of the airflow channel, non-power boost control measures are implemented to suppress heat dissipation deterioration and output maintenance prompts.

2. The adaptive energy efficiency optimization control method for energy-saving fans according to claim 1, characterized in that, The steps for diagnosing the root cause of abnormal heat dissipation, which involves calculating load changes or obstructed airflow channels, include: Acquire temperature data for the cooling zone; Calculate the average rate of change of the temperature data within a preset time window to obtain the average temperature change rate; monitor the average temperature change rate of the hot air channel; When the average temperature change rate of the hot air channel continues to be higher than the preset threshold or when the deviation of the actual airflow parameter information from the reference airflow parameter information continues to increase, a slow change anomaly sign information is generated. Obtain the average calculated load change information of all racks in the cooling area within a preset time window; When the information of the slow change anomaly is detected and the information of the average calculated load change does not show an increase, a fine-tuning command is sent to the wind turbine controller to slightly increase the wind turbine speed and maintain it for a preset duration. Within the monitoring time window after the fan speed is fine-tuned, the average temperature change rate of the hot air channel is monitored and the unit power air volume information and corresponding ratio information before and after the fan adjustment are calculated. Based on the average temperature change rate, the unit power air volume information, and the corresponding ratio information, the root cause is diagnosed as the accumulation of subtle fluctuations in the computational load or an increase in distributed airflow resistance. When the root cause is the accumulation of minor fluctuations in the computational load, readjust the fan speed to match the heat dissipation requirements; when the root cause is the increase in distributed airflow resistance, stop increasing the fan power and output a maintenance alarm message.

3. The adaptive energy efficiency optimization control method for energy-saving fans according to claim 2, characterized in that, The step of slightly increasing the fan speed and maintaining it for a preset duration includes: Obtain sub-region computational load information for each sub-region within the cooling area; Calculate load information based on the sub-regions and evaluate the load change trend information of each sub-region; If the information of the slow change anomaly is detected and the average calculated load change information does not show an increase, and if the load change trend information shows that a sub-region has increased beyond the limit within a preset time, the execution of the fan speed fine adjustment will be suspended and the load tracking and monitoring process will be entered. If the load change trend information shows that the calculation load of each sub-region is in a stable state, then the fan speed increase is determined according to the overall heat dissipation demand of the cooling area, and the fine-tuning instruction information corresponding to the fan speed increase is sent to the fan controller. During the fine-tuning of the fan speed, the temperature change rate of each sub-region is continuously monitored. Based on the temperature change rate of each sub-region, when the temperature rise rate of any sub-region exceeds the preset warning threshold, the fan speed increase is stopped.

4. The adaptive energy efficiency optimization control method for an energy-saving fan according to claim 2, characterized in that, The step of monitoring the average temperature change rate of the hot air channel includes: Acquire data from various temperature sensors within the hot air duct; Time compensation and spatial correction are performed on the temperature sensor data to obtain corrected temperature data; Based on the corrected temperature data, calculate the local temperature gradient information and the local temperature change rate information; The local temperature change rate information is compared with the overall average temperature change rate of the hot air channel to identify local hot spot areas. When local hot spots exist, analyze the rate of temperature change in the local hot spots and determine the response of the local hot spots to fine-tuning of the fan speed.

5. The adaptive energy efficiency optimization control method for an energy-saving fan according to claim 2, characterized in that, The step of monitoring the average temperature change rate of the hot air channel includes: Acquire data from various temperature sensors within the hot air channel, as well as real-time computing load information from each server; Based on real-time load information, identify server information with high dynamic or sudden loads within the hot air channel; When the server information exists, estimate the heat change caused by load fluctuation based on the temperature sensor data of the area where the server information is located. The estimated heat change is subtracted from the temperature sensor data to obtain the load decoupling temperature data; Based on the load decoupling temperature data, the average temperature change rate of the hot air channel is calculated.

6. The adaptive energy efficiency optimization control method for an energy-saving fan according to claim 5, characterized in that, The step of calculating the average temperature change rate of the hot air channel includes: Acquire load decoupling temperature data for each heat dissipation area within the hot air channel; Based on the physical dimensions, heat capacity, and airflow path information of each heat dissipation area, determine the thermal inertia parameters. Based on the thermal inertia parameter information, adjust the time window length and weight of each heat dissipation area; Calculate the average temperature change rate of each heat dissipation area based on the adjusted time window length and weight; The average temperature change rate of the hot air channel is calculated based on the average temperature change rate of each heat dissipation area.

7. The adaptive energy efficiency optimization control method for an energy-saving fan according to claim 6, characterized in that, The step of calculating the average temperature change rate of the hot air channel includes: Obtain the average temperature change rate of each heat dissipation area; Identify information regarding fluctuations in the average temperature change rate; Based on the fluctuation information, the average temperature change rate of each heat dissipation area is smoothed to suppress the impact of short-term drastic fluctuations. Based on the physical location information, airflow direction information, and heat transfer path information of each heat dissipation area, determine the degree of mutual influence between heat dissipation areas; Based on the information on the degree of mutual influence, the weights of the average temperature change rate of each heat dissipation area are adjusted; The average temperature change rate of each heat dissipation area after smoothing is weighted and averaged according to the adjusted weights to obtain the overall average temperature change rate of the hot air channel.

8. The adaptive energy efficiency optimization control method for an energy-saving fan according to claim 7, characterized in that, The step of determining the degree of mutual influence between heat dissipation areas includes: Obtain spatial location information of each cabinet, equipment, and temporary obstacle within the cooling area; Based on the spatial location information, construct the three-dimensional physical layout information of the cooling area; Obtain airflow velocity and direction information for each heat dissipation area; Obtain temperature distribution information along the heat transfer path; Based on the three-dimensional physical layout information, the airflow velocity and direction information, and the temperature distribution information, the airflow path information and heat transfer path information of the cooling area are simulated. Based on the simulated airflow path information and heat transfer path information, the degree of mutual influence between each heat dissipation area is determined.

9. The adaptive energy efficiency optimization control method for an energy-saving fan according to claim 8, characterized in that, The steps for obtaining the airflow path information and heat transfer path information of the simulated cooling zone include: Acquire raw data from visual sensors, LiDAR, and thermal imagers; The original data is preprocessed to obtain preprocessed data; the preprocessing includes timestamp alignment, spatial coordinate system transformation, and data format unification. Data quality assessment is performed on the preprocessed data to identify outliers or missing regions and obtain the corresponding data quality assessment results. Based on the data quality assessment results, a weighted fusion process is performed on the preprocessed data; The fused data is continuously compared with the preset layout benchmark model information to identify layout deviation information, and the fused data is corrected based on the layout deviation information. Based on the corrected fusion data, the three-dimensional physical layout information is updated, and based on the updated three-dimensional physical layout information, corresponding airflow path information and heat transfer path information are generated.

10. An energy-saving fan adaptive energy efficiency optimization control system, used to execute energy-saving fan adaptive energy efficiency optimization control, characterized in that, include: The parameter information acquisition module is used to acquire fan operating parameter information, actual airflow parameter information of the cooling area, and server calculation load information; The airflow parameter determination module is used to determine the reference airflow parameter information of the cooling area under preset operating conditions based on the fan operating parameter information; An anomaly detection module is used to compare the actual airflow parameter information with the reference airflow parameter information, and combine the server's calculated load information to identify anomaly detection information in the cooling area. The airflow characteristic analysis module is used to analyze the airflow characteristic information of the cooling area when the abnormal signs are detected and the server computing load information does not show an increase, so as to diagnose whether the root cause of the heat dissipation abnormality is a change in computing load or obstruction of airflow channels. The optimized control execution module is used to adjust the fan operating parameters to match the heat dissipation requirements if the root cause is a change in the calculated load; if the root cause is an obstruction of the airflow channel, non-power boost control measures are implemented to suppress heat dissipation deterioration and output maintenance prompt information.

Images