Intelligent control method of air conditioning system for server room temperature control

Abstract

The application relates to the technical field of data regulation, in particular to an intelligent control method for an air conditioning system for temperature control of a server room, which comprises the following steps: acquiring temperature data, wind direction data and load power data of each monitoring position in a time monitoring window of the server room; acquiring temperature data and total load data of a corresponding channel position of each air conditioner; obtaining a thermal regulation demand index according to the fluctuation of the temperature data, the difference of the wind direction data of each monitoring position, and the change coordination between the temperature data and the load power data; obtaining a thermal radiation characteristic index according to the change coordination of the temperature data of adjacent channel positions and the change coordination of the total load data of the adjacent channel positions; and regulating the air conditioner temperature according to the thermal regulation demand index and the thermal radiation characteristic index corresponding to each air conditioner. The application makes the temperature regulation result more reasonable, realizes dynamic thermal balance among different channels and maintains the temperature stability of the server room.

Description

Technical Field

[0001] This invention relates to the field of data control technology, specifically to an intelligent control method for an air conditioning system used for temperature control in server rooms. Background Technology

[0002] With the rapid development of information technology and cloud computing industries, the scale of data centers and server rooms continues to expand, the power density of equipment inside the server rooms is steadily increasing, and heat is highly concentrated in the server racks, resulting in a significant increase in the heat load of the server room environment. To ensure the long-term stable operation of servers, the server room needs to be maintained within a constant and reasonable temperature and humidity range; if the temperature control system responds slowly or the control is uneven, it will cause local overheating or condensation, which will lead to abnormal server operation or even hardware damage.

[0003] Traditional server room temperature control systems often employ constant temperature settings or timed start / stop control modes. However, in high-density server deployment scenarios, uneven heat distribution within server racks is common, leading to significant temperature differences and airflow short-circuiting issues in different areas. While existing server room temperature control and air conditioning systems possess some automation and intelligent control capabilities, their core control logic remains primarily based on fixed threshold adjustments and single-parameter feedback closed loops. The feedback signals from existing systems are mostly derived from the average temperature of the server room or data from single-point sensors, making it difficult to accurately reflect the thermal distribution characteristics of each rack and aisle. In actual operation, the temperature control method of the air conditioning system still adjusts based on the overall average temperature, lacking a feature mapping mechanism that fuses multi-parameter features, resulting in poor rationality in temperature regulation. Summary of the Invention

[0004] To address the technical problem that existing methods, which rely on overall average temperature for regulation and lack a feature mapping mechanism for multi-parameter feature fusion, resulting in poor temperature control rationality, the present invention aims to provide an intelligent control method for air conditioning systems used in server room temperature control. The specific technical solution adopted is as follows:

[0005] Acquire temperature, wind direction, and load power data for each monitoring location in the server room within a time monitoring window; acquire temperature and total load data for each air conditioner's corresponding channel location.

[0006] Based on the temperature data fluctuations at each monitoring location within the time monitoring window and the differences in wind direction data between different monitoring locations, combined with the synergistic changes between temperature data and load power data, the thermal control demand index for the server room is obtained.

[0007] Based on the coordination of temperature data changes between adjacent channel locations within the time monitoring window, and the coordination of total load data changes between adjacent channel locations, the thermal radiation characteristic index of each air conditioning channel is obtained.

[0008] The air conditioner temperature is regulated based on the heat regulation demand index and heat radiation characteristic index corresponding to each air conditioner.

[0009] Preferably, the step of obtaining the thermal regulation demand index for the server room based on the temperature data fluctuations at each monitoring location within the time monitoring window, the differences in wind direction data between different monitoring locations, and the synergistic effect of changes in temperature data and load power data, specifically includes:

[0010] Based on the degree of temperature fluctuation of each monitoring location within the time monitoring window, and the changing trend of temperature differences between adjacent times, the temperature non-uniformity coefficient of each monitoring location is obtained.

[0011] Based on the differences in wind direction data between adjacent monitoring locations at the same time within the time monitoring window, and combined with the temperature non-equilibrium coefficient, the degree of consistency in airflow organization is obtained.

[0012] The thermal sensitivity is obtained by considering the difference between the rate of change of temperature data and the rate of change of load power data at each monitoring location within the time monitoring window, combined with the difference between the magnitude of change of temperature data and the magnitude of change of load power data at adjacent times.

[0013] Based on the negative correlation coefficient of the airflow organization consistency and the degree of thermal sensitivity, the thermal control demand index of the server room is determined.

[0014] Preferably, the step of obtaining the temperature non-uniformity coefficient for each monitoring location based on the fluctuation of temperature data at each monitoring location within the time monitoring window and the changing trend of temperature differences between adjacent times specifically includes:

[0015] Obtain the fluctuation coefficient of all temperature data at a selected monitoring location within a time monitoring window, where the selected monitoring location refers to any monitoring location.

[0016] The absolute value of the temperature difference between the selected monitoring location and the adjacent previous time point within the time monitoring window is obtained as the temperature difference data.

[0017] The temperature non-uniformity coefficient of the selected monitoring location is determined based on the ratio between the temperature difference data of each moment and the adjacent previous moment within the time monitoring window at the selected monitoring location, and the fluctuation coefficient.

[0018] Preferably, the step of obtaining the airflow organization consistency based on the difference in wind direction data between adjacent monitoring locations at the same time within the time monitoring window, combined with the temperature non-uniformity coefficient, specifically includes:

[0019] Obtain the wind direction angle between the selected monitoring location and adjacent monitoring locations at each same time.

[0020] The mean value of the cosine of the wind direction angle corresponding to each monitoring location at each time is used as the first characteristic coefficient, and the negative correlation coefficient of the mean value of the temperature non-uniformity coefficient of all monitoring locations is used as the second characteristic coefficient; the product of the first characteristic coefficient and the second characteristic coefficient is the degree of consistency of the airflow organization.

[0021] Preferably, the step of obtaining the thermal sensitivity based on the difference between the rate of change of temperature data and the rate of change of load power data at each monitoring location within the time monitoring window, combined with the difference between the magnitude of change of temperature data and the magnitude of change of load power data at adjacent time points, specifically includes:

[0022] Based on the difference in temperature data at adjacent times within the time monitoring window at the selected monitoring location, a first slope value and a first difference value are determined for each two adjacent times. Based on the difference in load power data at adjacent times within the time monitoring window at the selected monitoring location, a second slope value and a second difference value are determined for each two adjacent times.

[0023] The degree of thermal sensitivity is determined based on the difference between the first slope value and the second slope value corresponding to the same time sequence at the selected monitoring location, as well as the difference between the first difference value and the second difference value.

[0024] Preferably, the step of obtaining the thermal radiation characteristic index of each air conditioner channel based on the coordination of temperature data changes between adjacent channel locations within the time monitoring window and the coordination of total load data corresponding to adjacent channel locations specifically includes:

[0025] Based on the rate of change of temperature data at each channel location within the time monitoring window, the moment of temperature abrupt change at each channel location is obtained;

[0026] Obtain the first and second adjacent positions of the target channel position, where the first and second adjacent positions are both channel positions adjacent to the target channel position; wherein the target channel position is any channel position.

[0027] Based on the relative distribution of temperature change moments corresponding to the first adjacent position and the target channel position, and the relative distribution of temperature change moments corresponding to the second adjacent position and the target channel position, the first distribution coefficient of the target channel position is obtained.

[0028] Based on the relative distribution of the fluctuation degree of the total load data corresponding to the first adjacent position and the target channel position, and the relative distribution of the fluctuation degree of the total load data corresponding to the second adjacent position and the target channel position, the second distribution coefficient of the target channel position is obtained.

[0029] The product of the first distribution coefficient and the second distribution coefficient at each channel location is used as the thermal radiation characteristic index of each air-conditioning channel.

[0030] Preferably, obtaining the temperature abrupt change time at each channel location based on the rate of change of temperature data within the time monitoring window specifically includes:

[0031] Obtain the slope value corresponding to the temperature data of every two adjacent moments within the time monitoring window for the target channel location. Arrange all slope values ​​in a set order to obtain a slope value sequence. Obtain the first-order difference value of the slope value sequence. Take the slope value corresponding to the maximum value of the first-order difference value as the dividing point. Take all moments corresponding to the slope value that is greater than or equal to the dividing point as the temperature change moment of the target channel location.

[0032] Preferably, obtaining the first distribution coefficient of the target channel position based on the relative distribution of the temperature change times corresponding to the first adjacent position and the target channel position, and the relative distribution of the temperature change times corresponding to the second adjacent position and the target channel position, specifically includes:

[0033] Obtain the first ratio of the temperature change time corresponding to the first adjacent position and the target channel position, obtain the second ratio of the temperature change time corresponding to the second adjacent position and the target channel position, and sum the first ratio and the second ratio of all times as the first distribution coefficient of the target channel position.

[0034] Preferably, the step of obtaining the second distribution coefficient of the target channel position based on the relative distribution of the fluctuation degree of the total load data corresponding to the first adjacent position and the target channel position, and the relative distribution of the fluctuation degree of the total load data corresponding to the second adjacent position and the target channel position, specifically includes:

[0035] The variance of all total load data for each channel location within the time monitoring window is used as the fluctuation coefficient.

[0036] Obtain the third ratio of the fluctuation coefficient between the target channel position and the first adjacent position, obtain the fourth ratio of the fluctuation coefficient between the target channel position and the second adjacent position, and use the sum of the third and fourth ratios as the second distribution coefficient of the target channel position.

[0037] Preferably, the step of regulating the air conditioner temperature based on the thermal regulation demand index and thermal radiation characteristic index corresponding to each air conditioner specifically includes:

[0038] The normalized coefficient of the product of the thermal regulation demand index and the thermal radiation characteristic index corresponding to each air conditioner is used as the intelligent regulation coefficient of each air conditioner.

[0039] The difference between the current target temperature value set for each air conditioner and the mean value of all temperature data at all monitoring locations corresponding to the air conditioner within the time monitoring window is obtained as the temperature deviation of each air conditioner.

[0040] The product of the intelligent control coefficient and the temperature deviation is added to the target temperature value to obtain the controlled target temperature value for each air conditioner. This controlled target temperature value is used to adjust the set temperature of the air conditioner.

[0041] The embodiments of the present invention have at least the following beneficial effects:

[0042] This invention first acquires multi-dimensional parameters for each monitoring location, simultaneously acquiring multi-dimensional parameters for each air conditioning channel, providing a data foundation for subsequent feature analysis of the coupling relationships between multiple dimensions and types of parameters. Then, considering the overall temperature distribution of the server room, feature analysis is performed on temperature and airflow direction between different monitoring locations, combined with the synergistic relationship between temperature and load changes, to comprehensively assess the degree of temperature control required in the current server room scenario. This integration of multi-dimensional feature analysis results avoids the irrationality of single-temperature judgments, initially determining the degree of temperature control required, and providing decision-making data for subsequent precise control of the air conditioning system. Furthermore, taking the channel locations corresponding to the air conditioners as the analysis object, based on temperature data changes between adjacent channel locations and total load data changes, the synergy of heat flow transfer between server room channels and the correlation of load fluctuations are quantified, integrating dual-dimensional synergistic features to reflect the overall heat dissipation status, providing a basis for inter-channel correlation in subsequent coordinated control of the air conditioning system. Finally, combining the feature analysis results from both aspects, the temperature of each air conditioner is dynamically adjusted to avoid the accumulation of local thermal disturbances in a single channel, making the temperature control results more reasonable, achieving dynamic thermal balance between different channels, and maintaining temperature stability in the server room. Attached Figure Description

[0043] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart of the steps of an intelligent control method for an air conditioning system for temperature control in a server room, provided by the present invention.

[0045] Figure 2 This is a schematic diagram showing the location of the access channels in the server room provided by the present invention;

[0046] Figure 3This is a flowchart of the steps for obtaining the thermal control demand index of a server room provided by the present invention;

[0047] Figure 4 This is a flowchart of the steps for obtaining the thermal radiation characteristic index of each air conditioner channel provided by the present invention. Detailed Implementation

[0048] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent control method for an air conditioning system for temperature control in a server room according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0050] The following description, in conjunction with the accompanying drawings, details a specific scheme for an intelligent control method of an air conditioning system for temperature control in a server room provided by the present invention.

[0051] Please see Figure 1 The diagram illustrates a flowchart of a method for intelligent control of an air conditioning system for temperature control in a server room, according to an embodiment of the present invention. The method includes the following steps:

[0052] Step S100: Obtain temperature data, wind direction data, and load power data for each monitoring location in the server room within the time monitoring window; obtain temperature data and total load data for each air conditioner's corresponding channel location.

[0053] Monitoring locations are set up in multiple different locations within the server room. By using a zoned deployment method, a thermal distribution monitoring model for the server room space is established. The method for setting up the monitoring locations needs to be based on the actual size of the server room and the actual distribution of the server racks.

[0054] Specifically, in the server room, monitoring points are deployed at the front (intake side) and rear (exhaust side) of each rack column. The number of monitoring points is determined by the rack column length, with one point for every two or three racks, and all points are installed at a uniform height. Each monitoring point corresponds to a local area, allowing the acquisition of the total load power of all devices within that local area as load power data. For example, a pre-defined area centered on a monitoring point can be used as the local area corresponding to that monitoring point. The distance between adjacent monitoring points or the number of racks can be determined by the actual rack column length. The area corresponding to the local area can be set based on the server room area in the specific implementation scenario, for example, 20-30 square meters, without any restrictions.

[0055] Then, the channel is positioned in the center of a ventilation duct corresponding to a refrigeration and air conditioning unit, such as... Figure 2 The total load data corresponding to the location of the triangle in the diagram represents the sum of the load power of all devices on both sides of the ventilation channel. A ventilation channel corresponding to a cooling air conditioner can be an intake channel formed by cabinets arranged face-to-face, or an exhaust channel formed by cabinets arranged back-to-back. Therefore, the monitoring locations set in this embodiment mainly cover the server rack area and ventilation channels of the server room.

[0056] In other embodiments, monitoring positions can be set at the air supply outlets and the ceiling return air area. The air supply outlets refer to the exit areas where the precision air conditioner supplies cold air to the cold aisle; monitoring positions are generally located at the air conditioner outlets. The ceiling return air area refers to the high-altitude area below the ceiling of the server room where hot air accumulates, typically in the top area of ​​the server room. One monitoring position is set for each local area. Considering objective factors such as cost, implementers can set up the monitoring positions according to the specific implementation scenario, ensuring coverage of key local areas within the server room.

[0057] In other embodiments, the server room can be divided into several local areas according to a preset size, and a monitoring position can be set in each local area. In this embodiment, the server room is treated as a whole area for feature analysis. In other embodiments, the implementer can choose according to the specific implementation scenario. If the server room area is large, the server room can be divided into different areas, and each different area can be treated as a whole area for feature analysis. The area division method is a well-known technology and will not be described in detail here.

[0058] As a specific example, the duration of the time monitoring window can be set to 5 minutes. The time monitoring window contains several data collection moments, and the time interval between adjacent moments is equal. There is no restriction here, and the implementer can set the data collection frequency according to the specific real-time scenario.

[0059] Temperature sensors are installed at each monitoring location to collect temperature data in real time, and wind speed sensors, such as ultrasonic anemometers, are installed to collect wind direction angles as wind direction data. The total actual load power of all devices within the local area of ​​the monitoring location is also acquired in real time as the load power data for that location. For example, the real-time power of each rack or server node can be collected through the power monitoring unit in a data center management system or server power management system. All collected data is then aggregated and synchronized in time via a central control server or edge computing unit for subsequent analysis and processing. It should be understood that after data acquisition, preprocessing operations, such as noise reduction, can be performed on the data. Data noise reduction methods are well-known techniques and are not limited here.

[0060] Step S200: Based on the temperature data fluctuations at each monitoring location within the time monitoring window and the differences in wind direction data between different monitoring locations, and combined with the synergistic changes between temperature data and load power data, the thermal regulation demand index of the server room is obtained.

[0061] The main purpose of the thermal control demand index is to quantify the intensity of temperature control demand in server rooms. It mainly measures the temperature distribution imbalance at monitoring locations within the server room, assesses airflow heat dissipation capacity and temperature response sensitivity to load, and ultimately integrates the results of multi-dimensional feature analysis to preliminarily determine the degree of temperature control demand, providing decision-making data for subsequent precise control of the air conditioning system.

[0062] First, temperature imbalance is the core cause of temperature control regulation, and airflow organization is the key guarantee for heat dissipation. The comprehensive analysis of the characteristics of the two aspects can fully reflect whether temperature imbalance can be effectively alleviated by airflow. The specific implementation process is shown in step S201.

[0063] Firstly, the temperature imbalance coefficient at a single monitoring location takes into account both the fluctuation amplitude and the trend of change. The degree of fluctuation in temperature data directly reflects the dispersion of heat distribution in the current area (the greater the fluctuation, the more obvious the local hotspots). The changing trend of temperature differences between adjacent times can reflect whether the imbalance is worsening. Simply looking at the fluctuation amplitude cannot determine the dynamic evolution of the imbalance, and simply looking at the trend cannot quantify the intensity of the imbalance. Therefore, the product of the two needs to be used as the temperature imbalance coefficient to fully characterize the temperature imbalance state at a single location. Specific implementation steps are detailed in steps S2011 to S2013.

[0064] Secondly, the consistency of airflow organization combines wind direction differences and temperature imbalance coefficients. Wind direction differences between adjacent monitoring locations reflect the continuity of airflow direction (the smaller the wind direction difference, the smoother the airflow). However, the actual value of airflow needs to be considered in conjunction with temperature imbalance. If a region has a high temperature imbalance coefficient (significant hotspots) but small wind direction differences (smooth airflow), it indicates great heat dissipation potential and high-quality airflow organization. If the temperature imbalance is high and the wind direction differences are large (chaotic airflow), heat dissipation is difficult. Therefore, it is necessary to calculate the consistency of airflow organization by weighting the wind direction consistency with the reciprocal of the temperature imbalance coefficient, thus achieving a coupled assessment of temperature imbalance and heat dissipation capacity. Specific implementation steps are detailed in steps S2014 to S2015.

[0065] Then, load power is the core source of heat in the computer room. The sensitivity of temperature to load changes directly determines the urgency of regulation. The higher the sensitivity, the easier it is for load fluctuations to cause temperature exceedances, and the stronger the need for regulation. Thermal sensitivity is mainly measured from two dimensions: the difference in rate of change and the difference in amplitude of change. The difference in rate of change reflects the synchronicity of the changing trends of the two (the smaller the difference in rate of change, the more timely the temperature changes with the load, and the more sensitive the response); the difference in amplitude of change reflects the intensity of the impact of load disturbance on temperature (the smaller the difference in amplitude of change, the more significant the temperature change caused by a unit load change, and the stronger the coupling). By comprehensively analyzing the differences in these two aspects, the thermal sensitivity is evaluated, and the coupled response characteristics of temperature and load are fully characterized. The specific method is described in step S202.

[0066] Finally, the thermal regulation demand index is obtained by combining the consistency of airflow organization and the degree of thermal sensitivity, reflecting the superimposed effect of heat dissipation difficulty and response urgency.

[0067] In this regard, such as Figure 3 As shown, the method for obtaining the thermal control demand index of the server room can be implemented by steps S201 to S203.

[0068] Step S201: Based on the fluctuation of temperature data at each monitoring location within the time monitoring window and the changing trend of temperature differences between adjacent times, obtain the temperature non-uniformity coefficient for each monitoring location; based on the difference in wind direction data between adjacent monitoring locations at the same time within the time monitoring window, and combined with the temperature non-uniformity coefficient, obtain the airflow organization consistency.

[0069] The main objectives of this step are: firstly, to analyze the fluctuation range and trend of each monitoring location to quantify the temperature non-equilibrium coefficient; and secondly, to analyze the differences in wind direction data between adjacent monitoring locations at the same time, and, in conjunction with the temperature non-equilibrium coefficient, to quantify the consistency of airflow organization in the current server room.

[0070] Step S2011: Obtain the fluctuation coefficient of all temperature data at the selected monitoring location within the time monitoring window. The selected monitoring location refers to any monitoring location.

[0071] Specifically, the fluctuation coefficient reflects the degree of fluctuation of temperature data at a monitoring location at all times within a time monitoring window. As a specific example, this embodiment calculates the variance of all temperature data at the selected monitoring location within the time monitoring window as the fluctuation coefficient.

[0072] Step S2012: Obtain the absolute value of the temperature data difference between the selected monitoring location and the adjacent previous time point within the time monitoring window as the temperature difference data for each time point.

[0073] As a specific example, for a selected monitoring location, the absolute value of the difference between the temperature data at time t and the temperature data at time t-1 is taken as the temperature difference data at time t. Thus, there is a temperature difference data between each two adjacent times. If the adjacent previous time cannot be obtained at the first time within the monitoring time window, it is not considered.

[0074] Step S2013: Based on the ratio between the temperature difference data of the selected monitoring location at each time point within the time monitoring window and the adjacent previous time point, and the fluctuation coefficient, determine the temperature non-uniformity coefficient of the selected monitoring location.

[0075] Specifically, taking any monitoring location as an example, this embodiment will use the first... If a number of monitoring locations are selected, the method for obtaining the temperature non-uniformity coefficient of the selected monitoring locations can be expressed by the formula: ,in, This represents the temperature non-uniformity coefficient at the selected monitoring location. Indicates the first The selected monitoring location is also known as the selected monitoring location. Indicates the first The fluctuation coefficient of all temperature data at each monitoring location within the time monitoring window. Indicates the first Temperature difference data at time t+1 within the time monitoring window for each monitoring location. Indicates the first Temperature difference data at time t for each monitoring location within the time monitoring window. This indicates the number of moments contained within the time monitoring window.

[0076] Volatility coefficient The fluctuation coefficient reflects the static dispersion of temperature at the selected monitoring location within the time monitoring window. A larger fluctuation coefficient indicates greater temperature fluctuations at the monitoring location within the time window, unstable local heat distribution, and the presence of instantaneous hot spots or sudden temperature drops, representing the intensity of temperature imbalance. Conversely, a smaller fluctuation coefficient indicates that the temperature remains relatively stable within the time monitoring window, with a lower risk of static imbalance.

[0077] Temperature difference data This reflects the dynamic changes in temperature at the selected monitoring location within the time monitoring window. Furthermore, it shows the ratio of temperature differences between adjacent time points. This ratio reflects the drastic changes in temperature data. A larger ratio indicates more dramatic changes in temperature differences and a greater trend of imbalance; a smaller ratio indicates slower changes in temperature differences and a less pronounced trend of imbalance. The mean value reflects the overall situation within the current time monitoring window.

[0078] It should be noted that, to ensure the calculation results are meaningful, in this embodiment, when performing fractional operations, if the denominator is 0, a parameter adjustment factor can be added to the denominator to prevent the denominator from being 0. This parameter adjustment factor is a very small positive number. For example, the value of this parameter adjustment factor can be 0.01. Its specific value can be set by the implementer according to the actual situation, and this application embodiment does not impose specific limitations.

[0079] The temperature non-uniformity coefficient reflects the degree of non-uniformity and fluctuation in temperature data at a single monitoring location within a time monitoring window.

[0080] In other embodiments, to avoid the situation where the denominator is 0, making the fraction calculation meaningless, the method for obtaining the temperature non-uniformity coefficient of the selected monitoring location can also be expressed by the formula: , This represents the preset parameter tuning factor, which takes the value of a very small positive number, such as 0.01.

[0081] Step S2014: Obtain the wind direction angle between the selected monitoring location and the adjacent monitoring location at each same time.

[0082] As a concrete example, in the server room, along the airflow direction from the vents, adjacent monitoring locations are used as reference monitoring locations. The angle between the wind direction of each monitoring location and the reference monitoring location at the same moment is taken as the corresponding wind direction angle. For example... Figure 2As shown, for ease of understanding, this is described using a two-dimensional image space. The air conditioner is located at the top, so the airflow direction of the current server room vents is vertically downward. For any monitoring position, an adjacent monitoring position can be obtained by moving vertically downward. It should be understood that monitoring positions where the wind direction angle cannot be obtained are not analyzed. It should also be noted that the angle between two directions generally ranges from 0° to 90°.

[0083] Step S2015: The mean value of the cosine of the wind direction angle corresponding to each monitoring location at each time moment is taken as the first characteristic coefficient, and the negative correlation coefficient of the mean value of the temperature non-uniformity coefficient of all monitoring locations is taken as the second characteristic coefficient; the product of the first characteristic coefficient and the second characteristic coefficient is the degree of consistency of the airflow organization.

[0084] Firstly, the larger the wind direction angle corresponding to each monitoring location at each moment, the smaller the cosine value of the wind direction angle. The average value of the cosine value of the wind direction angle of all monitoring locations at all moments is used to obtain the first characteristic coefficient, which reflects the consistency of airflow direction in the current server room scenario. The larger the value of the first characteristic coefficient, the higher the consistency of airflow organization in the server room.

[0085] Secondly, the mean value of the temperature non-uniformity coefficient at all monitoring locations is calculated to reflect the overall degree of temperature imbalance in the current server room scenario. The smaller the value, the greater the degree of temperature uniformity. In this embodiment, the negative correlation coefficient is represented by a negative exponential function, that is, the negative correlation coefficient is... As the second characteristic coefficient, where This represents an exponential function with base e. This represents the average of the temperature non-uniformity coefficients across all monitoring locations.

[0086] When the value of the second characteristic coefficient is larger and the value of the first characteristic coefficient is larger, it indicates that the overall temperature non-uniformity in the current server room scenario is low, that is, the temperature distribution is relatively uniform, and there is a relatively consistent wind direction. This indicates that there is high-quality airflow organization, that is, the corresponding airflow organization has high consistency.

[0087] The consistency of airflow organization reflects the degree of coupling between the spatial-temporal coherence of airflow direction and the temperature field equilibrium, and reflects the spatial transmission coherence of air as a medium for carrying cold energy.

[0088] Step S202: Based on the difference between the rate of change of temperature data and the rate of change of load power data at each monitoring location within the time monitoring window, and combined with the difference between the magnitude of change of temperature data and the magnitude of change of load power data at adjacent times, the thermal sensitivity is obtained.

[0089] Temperature evolution over a given region is determined not only by air transport but also by multiple coupled factors, including heat source intensity, local heat dissipation conditions, and air conditioning response behavior. Therefore, further dynamic coupling analysis is needed to obtain an actionable assessment of the control impact. The main purpose of this step is to evaluate the sensitivity of the system load to airflow disturbances and power changes by analyzing the coupling strength and response delay between different types of data at the same monitoring location.

[0090] Specifically, the first step is to determine the first slope value and the first difference value corresponding to each two adjacent times based on the difference in temperature data of the selected monitoring location at adjacent times within the time monitoring window.

[0091] Specifically, the first slope value refers to the slope value corresponding to the temperature data of each monitoring location at every two adjacent moments within the time monitoring window, reflecting the rate of change of the temperature data, and the first difference value refers to the temperature difference corresponding to the temperature data of each monitoring location at every two adjacent moments within the time monitoring window.

[0092] As a specific example, this implementation takes the selected monitoring location as an example. For the time monitoring window of the selected monitoring location, the temperature data at each moment is subtracted from the temperature data of the adjacent previous moment and divided by the time interval between the two adjacent moments to obtain the slope value corresponding to each pair of adjacent temperature data, which is the first slope value.

[0093] In other embodiments, the slope of the data point at each time point on the curve can also be obtained by curve fitting. For example, the time series can be used as the abscissa and the temperature data corresponding to the time series can be used as the ordinate. The least squares method can be used to fit the curve and obtain the slope value of the data point at each time point on the fitted curve. This method is a well-known technique and will not be described in detail here.

[0094] Furthermore, for the time monitoring window of the selected monitoring location, the absolute value of the difference between the temperature data at each moment and the temperature data at the adjacent previous moment is taken as the corresponding temperature difference, which is also the first difference value.

[0095] The second step is to determine the second slope value and the second difference value corresponding to each two adjacent moments based on the difference in load power data of the selected monitoring location at adjacent moments within the time monitoring window.

[0096] Specifically, the second slope value refers to the slope value corresponding to the load power data of each monitoring location at every two adjacent moments within the time monitoring window, reflecting the rate of change of the load power data. The second difference value refers to the load difference corresponding to the load power data of each monitoring location at every two adjacent moments within the time monitoring window.

[0097] It should be understood that the method for calculating the slope value of the load power data at each monitoring location within the time monitoring window for every two adjacent moments is the same as the method for calculating the slope value of the temperature data in the aforementioned steps, and will not be repeated here.

[0098] As a specific example, for a time monitoring window at a selected monitoring location, the absolute value of the difference between the load power data at each moment and the load power data at the adjacent previous moment is taken as the corresponding load difference, which is also the second difference value.

[0099] It should be noted that data from the previous time step, for which the adjacent time step cannot be obtained, will not be analyzed. Furthermore, to avoid the influence of different units on subsequent calculation results, this embodiment includes normalization processing of the first slope value, the first difference value, the second slope value, and the second difference value before performing the thermal sensitivity analysis. The normalization method can be the minimization normalization method, which is a well-known technique and is not limited here; implementers can choose according to the specific implementation scenario.

[0100] The third step is to determine the degree of thermal sensitivity based on the difference between the first slope value and the second slope value corresponding to the same time sequence at the selected monitoring location, as well as the difference between the first difference value and the second difference value.

[0101] Specifically, the difference in absolute value between the first slope value and the second slope value at the same time for the selected monitoring location is calculated to obtain the difference in the trend of change; the difference between the first difference value and the second difference value at the same time for the selected monitoring location is calculated as the difference in the magnitude of change. The degree of thermal sensitivity is negatively correlated with the difference in the trend of change, and the degree of thermal sensitivity is positively correlated with the difference in the magnitude of change.

[0102] More specifically, in this embodiment, the first If a number of monitoring locations are selected, the method for obtaining the thermal sensitivity of the selected monitoring locations can be expressed by the formula:

[0103]

[0104] in, Indicates the thermal sensitivity of the selected monitoring location. Indicates the first Selecting a monitoring location is equivalent to choosing a monitoring location. This indicates the number of moments contained within the time monitoring window; This represents the first slope value at the selected monitoring location at time n. This represents the second slope value at the selected monitoring location at time n. This represents the first difference value at the selected monitoring location at time n. This represents the second difference value at the selected monitoring location at time n.

[0105] It should be noted that, to ensure the calculation results are meaningful, in this embodiment, when performing fractional operations, if the denominator is 0, a parameter adjustment factor can be added to the denominator to prevent the denominator from being 0. This parameter adjustment factor is a very small positive number. For example, the value of this parameter adjustment factor can be 0.01. Its specific value can be set by the implementer according to the actual situation, and this application embodiment does not impose specific limitations.

[0106] This indicates the difference in trends; the smaller the value, the more synchronized the trends of temperature and load power at the corresponding moment. This represents the relative difference between the magnitude of temperature change and the magnitude of load change. The larger the value, the more significant the deviation of the temperature change from the power change, meaning the stronger the response of temperature to power change. The product of these two values ​​is used to calculate the overall trend synchronization and significant magnitude, reflecting the local coupling sensitivity at each moment.

[0107] A higher thermal sensitivity value indicates a high degree of synchronization between temperature and load power changes, with the temperature change being significantly greater than the load power change. This means the temperature responds quickly and strongly to load power disturbances, resulting in high coupled thermal sensitivity. Conversely, a lower thermal sensitivity value indicates a weaker correlation between temperature and power changes, resulting in low coupled thermal sensitivity.

[0108] Step S203: Based on the negative correlation coefficient of the airflow organization consistency and the thermal sensitivity, determine the thermal control demand index of the server room.

[0109] The thermal regulation demand index reflects the combined effect of heat dissipation difficulty and response urgency. The consistency of airflow organization is negatively correlated with regulation difficulty; higher consistency indicates smoother heat dissipation channels, making it easier to alleviate temperature imbalances through existing airflow, thus reducing regulation difficulty. Conversely, chaotic airflow can lead to continuous temperature increases even with low thermal sensitivity due to obstructed heat dissipation, making regulation more difficult. Therefore, a negative correlation coefficient (e.g., the reciprocal) should be applied to the consistency of airflow organization to quantify heat dissipation difficulty as an additive indicator. Thermal sensitivity is positively correlated with response urgency; higher thermal sensitivity indicates a greater susceptibility to temperature increases due to load changes, corresponding to a higher regulation demand.

[0110] As a specific example, the product of the reciprocal of the airflow organization consistency and the mean of the thermal sensitivity of all monitoring locations is used as the thermal control demand index for the server room. It should be understood that this embodiment obtains the negative correlation coefficient of airflow organization consistency by taking the reciprocal. To ensure the calculation result is meaningful, when performing fractional operations, if the denominator is 0, a parameter adjustment factor can be added to the denominator to prevent the denominator from being zero. This parameter adjustment factor is a very small positive number. For example, the value of this parameter adjustment factor can be 0.01. Its specific value can be set by the implementer according to the actual situation, and this application embodiment does not impose specific limitations. In other embodiments, the negative correlation coefficient can also be obtained by using the negative exponent of the natural constant, which will not be elaborated here.

[0111] The larger the negative correlation coefficient of the consistency of airflow organization (difficulty in heat dissipation), the higher the average thermal sensitivity (urgency of regulation). The larger the product of the two, the more likely the area is to experience temperature increases due to load fluctuations and to have difficulty dissipating heat naturally through airflow, thus the higher the regulation requirement for the corresponding area. The smaller the product, the more likely the heat dissipation is smooth or the temperature is not sensitive to the load, thus the lower the regulation requirement.

[0112] Step S300: Based on the coordination of temperature data changes in adjacent channel locations within the time monitoring window and the coordination of total load data changes in adjacent channel locations, obtain the thermal radiation characteristic index of each air conditioning channel.

[0113] The main purpose of the thermal radiation characteristic index is to quantify the synergy of heat flow transfer between computer room channels and the correlation with load fluctuations, capture the synchronous correlation of temperature changes between channels, measure the degree of response of adjacent load fluctuations, and finally integrate the two-dimensional synergistic characteristics to reflect the overall heat dissipation status, providing a basis for the inter-channel correlation for subsequent coordinated control of the air conditioning system.

[0114] In this regard, such as Figure 4 As shown, the method for obtaining the thermal radiation characteristic index of each air conditioner channel is implemented by steps S301 to S305.

[0115] Step S301: Based on the rate of change of temperature data at each channel location within the time monitoring window, obtain the time of temperature abrupt change at each channel location.

[0116] Specifically, the first step is to obtain the slope value corresponding to the temperature data of the target channel location at every two adjacent moments within the time monitoring window.

[0117] It should be noted that the slope value of the temperature data at the target channel location at each moment is obtained using the same method as in step S202, and will not be repeated here. If the adjacent time step cannot be obtained for the first time step, no analysis will be performed. The slope value reflects the rate of change of the temperature data at the target channel location at each moment, that is, it reflects the drastic degree of temperature change.

[0118] The second step is to arrange all slope values ​​in a set order to obtain a slope value sequence, obtain the first-order difference value of the slope value sequence, take the slope value corresponding to the maximum value of the first-order difference value as the dividing point, and take all times corresponding to the slope value being greater than or equal to the dividing point as the temperature change time of the target channel position.

[0119] It should be understood that the moment of temperature abrupt change represents the moment when the slope increases significantly.

[0120] As a specific example, in this embodiment, the order can be set from small to large slope values. Its core advantage is to focus on the intensity classification of temperature change rate. Using the strongest temperature change rate in all monitoring periods as the benchmark for abrupt change judgment, it can more objectively define the critical standard of the "abrupt change level" slope and improve the consistency of abrupt change time judgment between different channels.

[0121] It should be noted that the first-order difference value corresponds to two slope values, and the maximum value of the first-order difference is... This represents the inflection point where the rate of change experiences the most dramatic jump. The maximum value of the first-order difference. The two slope values ​​are associated and are represented as follows: and , representing the rate of change before and after the inflection point, respectively. Treating both as boundary points means defining the abrupt change process as starting from the slope value... It begins to appear, up to the slope value. This is a phase that has been confirmed and strengthened.

[0122] Temperature abrupt changes do not occur instantaneously, but rather develop gradually until the intensity increases. All moments with slope values ​​greater than or equal to the cutoff point are considered moments of temperature abrupt changes. This criterion captures both the starting point of the abrupt change trend (the moment corresponding to the cutoff point with a smaller slope value) and the moment when the abrupt change is clearly confirmed (the moment corresponding to the cutoff point with a larger slope value and subsequent moments with a sustained high slope), effectively avoiding the possibility of missing early signs of abrupt changes due to using only a single threshold.

[0123] Step S302: Obtain the first adjacent position and the second adjacent position of the target channel position, where the first adjacent position and the second adjacent position are both channel positions adjacent to the target channel position; wherein the target channel position is any channel position.

[0124] Specifically, the first adjacent position refers to the position of the channel in the channel adjacent to the left of the target channel position, and the second adjacent channel position refers to the position of the channel in the channel adjacent to the right of the target channel position. The consideration of adjacent channels is to quantify the range of thermal influence of the current channel on the surrounding area. If adjacent channels are ignored, the spatial characteristics of heat flow radiation transfer will be underestimated.

[0125] It should be noted that the method for obtaining the temperature change time of the first adjacent position and the second adjacent position is the same as the method for obtaining the temperature change time of the target channel position.

[0126] Step S303: Based on the relative distribution of the temperature change times corresponding to the first adjacent position and the target channel position, and the relative distribution of the temperature change times corresponding to the second adjacent position and the target channel position, the first distribution coefficient of the target channel position is obtained.

[0127] Specifically, the first ratio of the temperature change time corresponding to the first adjacent position and the target channel position is obtained, the second ratio of the temperature change time corresponding to the second adjacent position and the target channel position is obtained, and the sum of the first ratio and the second ratio of all times is used as the first distribution coefficient of the target channel position.

[0128] The first distribution coefficient needs to integrate the relationship between the abrupt change time of the target channel and the adjacent channels, and calculate the time ratio of the abrupt change time of the first adjacent position, the second adjacent position and the target channel. The first distribution coefficient analyzes the temperature delay characteristics between the target channel and the adjacent channels, reflects the coupling strength of the temperature abrupt change between adjacent channels in time, and essentially characterizes the temporal correlation of thermal disturbances between channels.

[0129] As a concrete example, if we take the i-th channel position as the target channel position, the method for obtaining the first distribution coefficient of the target channel position can be expressed by the formula:

[0130]

[0131] in, The first distribution coefficient represents the target channel position, where i represents the i-th channel position, i+1 represents the i+1-th channel position (which is also the second adjacent position), and i-1 represents the i-1-th channel position (which is also the first adjacent position). This represents the time of the z-th temperature abrupt change at the second adjacent position. This represents the time of the z-th temperature abrupt change at the first adjacent position. This represents the z-th time of temperature abrupt change at the target channel location. This represents the minimum total number of temperature abrupt change moments at the target channel location, the first adjacent location, and the second adjacent location.

[0132] Indicates the first ratio. The second ratio indicates that when the first ratio is less than 1 and the second ratio is less than 1, it means that the temperature change at the target channel location is later than the temperature change at the adjacent channel location. The adjacent channel experiences thermal disturbance first, and the heat flow radiates from both sides to the current target channel.

[0133] When the first ratio is greater than 1 and the second ratio is greater than 1, it indicates that the temperature change at the target channel location occurs earlier than the temperature change at the adjacent channel location. The current target channel experiences thermal disturbance first, and the heat flow radiates from the current target channel to both sides.

[0134] The first distribution coefficient reflects the overall temporal correlation between the target channel position and the positions of the channels on the left and right. A larger value of the first distribution coefficient indicates that the temperature abrupt change at the target channel position generally occurs earlier than at adjacent channel positions, meaning the thermal disturbance mainly originates from the target channel position and radiates to both sides, thus indicating a higher urgency for the current target channel to be controlled. Conversely, a smaller value of the first distribution coefficient indicates that the temperature abrupt change at the target channel position generally occurs later than at adjacent channel positions, meaning the thermal disturbance mainly originates from adjacent channel positions and radiates towards the center, thus indicating a lower urgency for the target channel to be controlled.

[0135] It should be noted that when the channel corresponding to the leftmost air conditioner cannot obtain the adjacent channel to its left, the first ratio is set to 0 when calculating the first distribution coefficient. For the same reason, when the channel corresponding to the rightmost air conditioner cannot obtain the adjacent channel to its right, the second ratio is set to 0 when calculating the first distribution coefficient. Therefore, when synthesizing the temporal correlation characteristics between two adjacent channels, an additive approach is used to avoid excluding edge channels from quantization.

[0136] Step S304: Based on the relative distribution of the fluctuation degree of the total load data corresponding to the first adjacent position and the target channel position, and the relative distribution of the fluctuation degree of the total load data corresponding to the second adjacent position and the target channel position, the second distribution coefficient of the target channel position is obtained.

[0137] Specifically, the first step is to obtain the variance of all total load data for each channel location within the time monitoring window as the fluctuation coefficient.

[0138] It should be understood that the fluctuation coefficient at each channel location characterizes the stability and fluctuation intensity of heat generated by the load within the channel. A larger fluctuation coefficient indicates a greater ability for the channel to generate its own thermal disturbances, resulting in more severe load fluctuations and making it more likely to become a source of heat radiation. A smaller fluctuation coefficient indicates a more stable load power and more uniform energy injection within the channel, making it less likely to generate thermal disturbances and more likely to passively receive radiation from neighboring areas.

[0139] The second step is to obtain the third ratio of the fluctuation coefficient between the target channel position and the first adjacent position, and the fourth ratio of the fluctuation coefficient between the target channel position and the second adjacent position. The sum of the third and fourth ratios is used as the second distribution coefficient of the target channel position.

[0140] The second distribution coefficient needs to be compared with the load fluctuation intensity ratio between the target channel location and the adjacent channel location. Its core function is to compare the load power variance between the target channel and the adjacent channels to determine whether the target channel has the ability to actively generate thermal disturbance. The more intense the load fluctuation, the easier it is to become a source of heat radiation.

[0141] As a concrete example, if we take the i-th channel position as the target channel position, then the method for obtaining the second distribution coefficient of the target channel position can be expressed by the formula:

[0142]

[0143] in, The second distribution coefficient represents the target channel position, where i represents the i-th channel position, i+1 represents the i+1-th channel position (i.e., the second adjacent position), and i-1 represents the i-1-th channel position (i.e., the first adjacent position). The fluctuation coefficient representing the target channel position. This represents the fluctuation coefficient at the second adjacent position. This represents the fluctuation coefficient of the first adjacent position.

[0144] It should be noted that, to ensure the calculation results are meaningful, in this embodiment, when performing fractional operations, if the denominator is 0, a parameter adjustment factor can be added to the denominator to prevent the denominator from being 0. This parameter adjustment factor is a very small positive number. For example, the value of this parameter adjustment factor can be 0.01. Its specific value can be set by the implementer according to the actual situation, and this application embodiment does not impose specific limitations.

[0145] The third ratio, The fourth ratio, when the third ratio Greater than 1, fourth ratio A value greater than 1 indicates a larger load fluctuation at the target channel location, making it more prone to thermal disturbances that may radiate to adjacent areas. When the third ratio... Less than 1, fourth ratio When the value is less than 1, it indicates that the load fluctuation amplitude at the target channel location is smaller, the adjacent channel locations are more likely to generate thermal disturbances, and the target channel location is less likely to become a radiation source.

[0146] The smaller the value of the second distribution coefficient, the weaker the load fluctuation at the target channel location is compared to the left and right neighboring areas, making it difficult for the target channel to actively generate thermal disturbances, and the heat flow mainly comes from the radiation of the neighboring areas; the larger the value of the second distribution coefficient, the stronger the load fluctuation at the target channel location is compared to the left and right neighboring areas, making it easier for the target channel to actively generate thermal disturbances, and it becomes the heat radiation source of the neighboring areas on both sides.

[0147] It should be noted that, similar to the first distribution coefficient, when the channel corresponding to the leftmost air conditioner cannot have its adjacent channel to the left, the third ratio is set to 0 when calculating the second distribution coefficient. For the same reason, when the channel corresponding to the rightmost air conditioner cannot have its adjacent channel to the right, the fourth ratio is set to 0 when calculating the second distribution coefficient. Therefore, when synthesizing the load intensity correlation characteristics between two adjacent channels, an additive approach is used to avoid excluding edge channels from quantization.

[0148] Step S305: The product of the first distribution coefficient and the second distribution coefficient for each channel location is used as the thermal radiation characteristic index for each air-conditioning channel.

[0149] For the target channel location, the temporal coupling degree and load fluctuation intensity are combined. The larger the product, the more tightly coupled the target channel is with the neighboring area in time and the relatively high load power fluctuation of the target channel. This indicates the existence of strong local heat radiation and coupling transmission phenomena, that is, the higher the heat flow radiation index of the target channel.

[0150] Step S400: Adjust the air conditioner temperature according to the heat regulation demand index and heat radiation characteristic index corresponding to each air conditioner.

[0151] Specifically, the normalized coefficient of the product of the thermal regulation demand index and the thermal radiation characteristic index corresponding to each air conditioner is used as the intelligent regulation coefficient corresponding to each air conditioner.

[0152] It should be understood that the Norm linear normalization function or the minimax normalization method can be used for normalization. Implementers can also choose other appropriate normalization functions according to the specific implementation scenario. This is a well-known technique and is not restricted here.

[0153] The thermal regulation demand index reflects the urgency and difficulty of cooling in the current scenario, while the thermal radiation characteristic index reflects the intensity and propagation capability of thermal disturbances in the current scenario. By multiplying these two characteristic analysis results, the degree of temperature regulation required in the current scenario is quantified. A larger intelligent regulation coefficient indicates a greater degree of temperature regulation required, while a smaller intelligent regulation coefficient indicates a lesser degree of temperature regulation required.

[0154] Furthermore, the difference between the current target temperature value set for each air conditioner and the average value of all temperature data at all monitoring locations within the time monitoring window is obtained as the temperature deviation of each air conditioner. The product of the intelligent control coefficient and the temperature deviation is then added to the target temperature value to obtain the adjusted target temperature value for each air conditioner. This adjusted target temperature value is used to adjust the set temperature of the air conditioner.

[0155] For any given air conditioner, the temperature deviation is calculated by subtracting the average temperature data from all monitoring locations within the corresponding time monitoring window from the currently set target temperature. This temperature deviation reflects the difference between the currently set temperature and the actual temperature in the current scenario. The product of the intelligent control coefficient and the temperature deviation reflects the required temperature correction. Adding this correction to the target temperature value yields the adjusted result. Ultimately, the adjusted target temperature can be adjusted via the air conditioning control module. It's understood that the target temperature set for the server room can be directly obtained from the air conditioning system.

[0156] It should be noted that in this embodiment, all adjacent monitoring positions on both sides of the channel corresponding to each air conditioner are taken as the range that the air conditioner can cover during the air supply process. Therefore, when adjusting the temperature, the actual temperature of the channel corresponding to the air conditioner is represented by the average of the temperature data of all monitoring positions on both sides of the channel.

[0157] The air conditioning system continuously monitors data and then updates the intelligent control coefficient based on the new monitoring data. By dynamically adjusting the supply air temperature as needed, it avoids overcooling, achieves dynamic thermal balance between different areas, and maintains a stable temperature in the server room.

[0158] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An intelligent control method for an air conditioning system used for temperature control in a server room, characterized in that, The method includes the following steps: Acquire temperature, wind direction, and load power data for each monitoring location in the server room within a time monitoring window; acquire temperature and total load data for each air conditioner's corresponding channel location. Based on the temperature data fluctuations at each monitoring location within the time monitoring window and the differences in wind direction data between different monitoring locations, combined with the synergistic changes between temperature data and load power data, the thermal control demand index for the server room is obtained. Based on the coordination of temperature data changes between adjacent channel locations within the time monitoring window, and the coordination of total load data changes corresponding to adjacent channel locations, the thermal radiation characteristic index of each air conditioning channel is obtained, including: Based on the rate of change of temperature data at each channel location within the time monitoring window, the moment of temperature abrupt change at each channel location is obtained; Obtain the first and second adjacent positions of the target channel position, where the first and second adjacent positions are both channel positions adjacent to the target channel position; wherein the target channel position is any channel position. Based on the relative distribution of temperature change moments corresponding to the first adjacent position and the target channel position, and the relative distribution of temperature change moments corresponding to the second adjacent position and the target channel position, the first distribution coefficient of the target channel position is obtained. Based on the relative distribution of the fluctuation degree of the total load data corresponding to the first adjacent position and the target channel position, and the relative distribution of the fluctuation degree of the total load data corresponding to the second adjacent position and the target channel position, the second distribution coefficient of the target channel position is obtained. The product of the first and second distribution coefficients at each channel location is used as the thermal radiation characteristic index of each air-conditioning channel. The air conditioner temperature is regulated based on the heat regulation demand index and heat radiation characteristic index corresponding to each air conditioner.

2. The intelligent control method for an air conditioning system for temperature control in a server room according to claim 1, characterized in that, The thermal control demand index for the server room is obtained by considering the temperature fluctuations at each monitoring location within the time monitoring window, the differences in wind direction data between different monitoring locations, and the synergistic effects of changes in temperature and load power data. Specifically, this includes: Based on the degree of temperature fluctuation of each monitoring location within the time monitoring window, and the changing trend of temperature differences between adjacent times, the temperature non-uniformity coefficient of each monitoring location is obtained. Based on the differences in wind direction data between adjacent monitoring locations at the same time within the time monitoring window, and combined with the temperature non-equilibrium coefficient, the degree of consistency in airflow organization is obtained. The thermal sensitivity is obtained by considering the difference between the rate of change of temperature data and the rate of change of load power data at each monitoring location within the time monitoring window, combined with the difference between the magnitude of change of temperature data and the magnitude of change of load power data at adjacent times. Based on the negative correlation coefficient of the airflow organization consistency and the degree of thermal sensitivity, the thermal control demand index of the server room is determined.

3. The intelligent control method for an air conditioning system for temperature control in a server room according to claim 2, characterized in that, The temperature non-uniformity coefficient for each monitoring location is obtained based on the fluctuation of temperature data at each monitoring location within the time monitoring window and the changing trend of temperature differences between adjacent times. Specifically, this includes: Obtain the fluctuation coefficient of all temperature data at a selected monitoring location within a time monitoring window, where the selected monitoring location refers to any monitoring location. The absolute value of the temperature difference between the selected monitoring location and the adjacent previous time point within the time monitoring window is obtained as the temperature difference data. The temperature non-uniformity coefficient of the selected monitoring location is determined based on the ratio between the temperature difference data of each moment and the adjacent previous moment within the time monitoring window at the selected monitoring location, and the fluctuation coefficient.

4. The intelligent control method for an air conditioning system for temperature control in a server room according to claim 3, characterized in that, The method of determining the consistency of airflow organization based on the difference in wind direction data between adjacent monitoring locations at the same time within the time monitoring window, combined with the temperature non-uniformity coefficient, specifically includes: Obtain the wind direction angle between the selected monitoring location and adjacent monitoring locations at each same time. The mean value of the cosine of the wind direction angle corresponding to each monitoring location at each time is used as the first characteristic coefficient, and the negative correlation coefficient of the mean value of the temperature non-uniformity coefficient of all monitoring locations is used as the second characteristic coefficient; the product of the first characteristic coefficient and the second characteristic coefficient is the degree of consistency of the airflow organization.

5. The intelligent control method for an air conditioning system for temperature control in a server room according to claim 3, characterized in that, The thermal sensitivity is determined by considering the difference between the rate of change of temperature data and the rate of change of load power data at each monitoring location within the time monitoring window, combined with the difference between the magnitude of change of temperature data and the magnitude of change of load power data at adjacent time points. Specifically, this includes: Based on the difference in temperature data at adjacent times within the time monitoring window at the selected monitoring location, a first slope value and a first difference value are determined for each two adjacent times. Based on the difference in load power data at adjacent times within the time monitoring window at the selected monitoring location, a second slope value and a second difference value are determined for each two adjacent times. The degree of thermal sensitivity is determined based on the difference between the first slope value and the second slope value corresponding to the same time sequence at the selected monitoring location, as well as the difference between the first difference value and the second difference value.

6. The intelligent control method for an air conditioning system for temperature control in a server room according to claim 1, characterized in that, The step of obtaining the temperature abrupt change moment for each channel location based on the rate of change of temperature data within the time monitoring window specifically includes: Obtain the slope value corresponding to the temperature data of every two adjacent moments within the time monitoring window for the target channel location. Arrange all slope values ​​in a set order to obtain a slope value sequence. Obtain the first-order difference value of the slope value sequence. Take the slope value corresponding to the maximum value of the first-order difference value as the dividing point. Take all moments corresponding to the slope value that is greater than or equal to the dividing point as the temperature change moment of the target channel location.

7. The intelligent control method for an air conditioning system for temperature control in a server room according to claim 1, characterized in that, The step of obtaining the first distribution coefficient of the target channel position based on the relative distribution of the temperature abrupt change times corresponding to the first adjacent position and the target channel position, and the relative distribution of the temperature abrupt change times corresponding to the second adjacent position and the target channel position, specifically includes: Obtain the first ratio of the temperature change time corresponding to the first adjacent position and the target channel position, obtain the second ratio of the temperature change time corresponding to the second adjacent position and the target channel position, and sum the first ratio and the second ratio of all times as the first distribution coefficient of the target channel position.

8. The intelligent control method for an air conditioning system for temperature control in a server room according to claim 1, characterized in that, The process of obtaining the second distribution coefficient for the target channel position based on the relative distribution of the fluctuation levels of the total load data corresponding to the first adjacent position and the target channel position, and the relative distribution of the fluctuation levels of the total load data corresponding to the second adjacent position and the target channel position, specifically includes: The variance of all total load data for each channel location within the time monitoring window is used as the fluctuation coefficient. Obtain the third ratio of the fluctuation coefficient between the target channel position and the first adjacent position, obtain the fourth ratio of the fluctuation coefficient between the target channel position and the second adjacent position, and use the sum of the third and fourth ratios as the second distribution coefficient of the target channel position.

9. The intelligent control method for an air conditioning system for temperature control in a server room according to claim 1, characterized in that, The step of regulating the air conditioner temperature based on the heat regulation demand index and heat radiation characteristic index corresponding to each air conditioner specifically includes: The normalized coefficient of the product of the thermal regulation demand index and the thermal radiation characteristic index corresponding to each air conditioner is used as the intelligent regulation coefficient of each air conditioner. The difference between the current target temperature value set for each air conditioner and the mean value of all temperature data at all monitoring locations corresponding to the air conditioner within the time monitoring window is obtained as the temperature deviation of each air conditioner. The product of the intelligent control coefficient and the temperature deviation is added to the target temperature value to obtain the controlled target temperature value for each air conditioner. This controlled target temperature value is used to adjust the set temperature of the air conditioner.

Images