Circulating water pump adaptive constant temperature control method and system based on ambient temperature

Abstract

This invention discloses an adaptive constant temperature control method and system for circulating water pumps based on ambient temperature, relating to the field of adaptive constant temperature control. The method involves acquiring the spatial structural parameters of an air conditioning cooler, constructing a three-dimensional thermal and humid air motion model coupling exhaust and intake airflows, generating a potential energy field for the risk of exhaust hot and humid air mass recirculation based on the three-dimensional thermal and humid air motion model, and combining real-time ambient temperature, forecast temperature, and meteorological boundary conditions to perform recirculation self-excited oscillation detection. A corrected equivalent ambient temperature is output to eliminate recirculation self-interference deviation. Using the corrected equivalent ambient temperature as a feedforward variable, a target operating current setpoint for the water pump is generated through a compensation function that compensates for the inverse relationship between the pump operating current and ambient temperature. After smoothing and filtering, the setpoint is output to the variable frequency drive of the circulating water pump. The variable frequency drive adjusts the output frequency according to the current setpoint to ensure that the pump operating current tracks the setpoint, maintaining a constant cooling water supply temperature.

Description

Technical Field

[0001] This invention relates to the field of adaptive constant temperature control, specifically to an adaptive constant temperature control method and system for circulating water pumps based on ambient temperature. Background Technology

[0002] In cooling water circulation systems of industrial and civil buildings, variable frequency control of circulating water pumps is the core means to achieve energy saving and constant temperature. Traditional control schemes mainly rely on single-loop closed-loop feedback regulation based on supply or return water temperatures. This method adjusts the pump frequency by a PID controller based on the deviation between the cooling water temperature and the set value. However, in actual operation and maintenance, it has been found that this pure feedback control mode exhibits obvious local instability under specific meteorological conditions such as early mornings in winter and late evenings in early spring, as well as under specific meteorological conditions such as low temperature and low wind speed. Through in-depth physical modeling and on-site flow field observation and deduction, it was found that the root cause of the problem lies in the short-circuit heat recirculation of the air conditioning cooler under specific boundary conditions. Specifically, the saturated hot and humid air discharged from the top of the air conditioning cooler, when there is no horizontal wind diffusion, is driven by density difference to undergo dynamic sinking and re-enters the heat exchange area under the negative pressure suction of the air inlet at the bottom of the air conditioning cooler.

[0003] This backflow phenomenon exhibits a non-steady-state, self-excited oscillation characteristic: the backflow of hot air causes a sharp drop in heat exchange efficiency, leading to an increase in supply water temperature; the temperature rise triggers a feedback loop to significantly increase pump power consumption to enhance cooling, and subsequently, the backflowing air mass dissipates due to thermodynamic evolution. Because existing control systems only sense the water temperature and lack predictive awareness of the evolution of the microenvironmental thermodynamic field at the air conditioner cooler inlet, control commands consistently lag behind changes in heat exchange boundary conditions. Furthermore, even when conventional control systems reference the ambient temperature from a weather station, they often fail to capture the local microenvironmental self-interference fluctuations caused by the equipment's own exhaust air because the sensors are often located far from the physical cooling equipment. This lack of sensing dimension and the mismatch in spatial layout result in frequent and difficult-to-eliminate constant-amplitude oscillations in the pump current even under stable macroscopic air temperature conditions, affecting the system's operational stability and energy efficiency.

[0004] To address the aforementioned shortcomings, a technical solution is provided. Summary of the Invention

[0005] To address the technical problems mentioned in the background section, this invention is proposed. This invention provides an adaptive constant temperature control method and system for circulating water pumps based on ambient temperature.

[0006] This invention is achieved through the following technical solution:

[0007] An adaptive constant temperature control method for circulating water pumps based on ambient temperature, the method comprising the following steps:

[0008] Obtain the spatial structural parameters of the air conditioner cooler, construct a three-dimensional thermal and humid air motion model of exhaust-intake airflow coupling, and generate the risk potential energy field of exhaust thermal and humid air mass recirculation based on the three-dimensional thermal and humid air motion model.

[0009] Based on the risk potential energy field of the exhaust hot and humid air mass recirculation, combined with the real-time ambient temperature, the weather forecast temperature and the meteorological boundary conditions, the recirculation self-excited oscillation detection is performed, and the corrected equivalent ambient temperature is output to eliminate the recirculation self-interference deviation.

[0010] Using the corrected equivalent ambient temperature as a feedforward variable, a target operating current setting value for the water pump is generated through a compensation function that compensates for the inverse change between the water pump operating current and the ambient temperature. After smoothing and filtering, the setting value is output to the variable frequency drive of the circulating water pump. The variable frequency drive adjusts the output frequency according to the current setting value so that the water pump operating current tracks the setting value, thereby maintaining a constant cooling water supply temperature.

[0011] Furthermore, the construction steps of the three-dimensional thermal and humid air motion model are as follows:

[0012] Obtain the vertical height of the air conditioner cooler's exhaust vent from the roof, the exhaust vent's cross-sectional diameter, the height of the bottom edge of the air inlet louvers, and the range of their circumferential curvature; with the geometric center of the exhaust vent as the origin, vertically upwards... A three-dimensional cylindrical coordinate system is established, with the exhaust vent defined as the emission source surface of hot and humid air masses, and the circumferential space area covered by the air inlet louvers defined as the air mass capture and confluence surface.

[0013] Furthermore, the construction steps of the three-dimensional thermal and humid air motion model also include:

[0014] The hot and humid air discharged from the exhaust vent is discretized into a sequence of micro-elemental air masses. For each micro-elemental air mass, the buoyancy-driven component is calculated based on its temperature difference with the ambient air. The temperature decay path is corrected based on the release of latent heat of condensation of moisture content. The radial diffusion rate is calculated based on the turbulent diffusion coefficient. The spatial motion trajectory of each micro-elemental air mass in the three-dimensional cylindrical coordinate system is obtained by combining the buoyancy-driven component, the temperature decay path, and the radial diffusion rate. The combination of the buoyancy-driven component, the temperature decay path, and the radial diffusion rate constitutes the three-dimensional hot and humid air motion model.

[0015] Furthermore, the step of correcting the temperature decay path includes:

[0016] Within each calculation time step, it is determined whether the current moisture content of the volumetric micro-element air mass exceeds the saturation moisture content corresponding to the current temperature. If it does, the excess water vapor mass is multiplied by the latent heat of vaporization to obtain the heat released by condensation, which is converted into the temperature rise increment of the air mass and added to the sensible heat cooling amount. At the same time, the mass of condensed water vapor is deducted from the moisture content of the air mass. When the moisture content drops below the saturation value, condensation terminates, and the air mass cools down rapidly and enters the rapid sinking stage.

[0017] Furthermore, the steps for constructing the risk potential energy field of the exhaust hot and humid air mass recirculation are as follows:

[0018] Based on the spatial motion trajectory of all volumetric micro-element air masses output by the three-dimensional thermal and humid air motion model, a geometric intersection determination with the air mass capture confluence surface is performed. Air masses that cross the air mass capture confluence surface are marked as return air masses and their remaining temperature during the crossing is recorded. The spatial distribution of return heat flux density per unit time is statistically analyzed to generate the risk potential energy field of the exhaust thermal and humid air mass return flow.

[0019] Furthermore, the calculation steps for the reflux heat flux density are as follows:

[0020] The gas mass capture surface is discretized into a rectangular array of cells along the circumferential and vertical directions. For each recirculating gas mass, the assigned cell is determined according to the crossing position. The difference between its remaining temperature and the ambient temperature is multiplied by the volumetric flow rate and the specific heat capacity of air to obtain the single-cell recirculating heat power. The single-cell recirculating heat power of all assigned recirculating gas masses in the statistical period is accumulated for each cell and then divided by the cell area and the statistical duration to obtain the time-averaged recirculating heat flux density.

[0021] Furthermore, the output step for correcting the equivalent ambient temperature is as follows:

[0022] Real-time outdoor wind speed is obtained and weather forecast temperature is obtained through the meteorological service interface. The real-time outdoor dry-bulb temperature and the weather forecast temperature are weighted and fused to obtain a comprehensive ambient temperature reference value. When the real-time outdoor wind speed is lower than the critical recirculation wind speed threshold, it is determined to be a high-incidence condition of recirculation. The critical recirculation wind speed threshold is determined by the height difference between the exhaust vent and the air inlet louver and the cross-sectional diameter of the exhaust vent. It is the minimum wind speed at which the farthest recirculation air mass is horizontally blown away from the capture radius before sinking to the air mass capture confluence.

[0023] Under conditions of high recirculation, the measured temperature sequence of the air inlet temperature sensor at the air inlet louver of the air conditioner cooler is continuously collected. Short-time sliding window spectrum analysis is performed to extract the periodic fluctuation component within the preset oscillation period range. The peak-valley temperature difference is defined as the recirculation self-excited oscillation amplitude and the center period is defined as the recirculation self-excited oscillation period. The recirculation heat flux density at the corresponding position in the recirculation risk potential energy field is compared and verified.

[0024] After confirming the existence of backflow self-excited oscillation through comparison and verification, the average time of a sliding window with a width that is an integer multiple of the backflow self-excited oscillation period is taken from the measured temperature sequence of the air intake. The difference between this average value and the comprehensive ambient temperature reference value is defined as the backflow average temperature rise offset. The corrected equivalent ambient temperature is output by superimposing the comprehensive ambient temperature reference value with the backflow average temperature rise offset.

[0025] Furthermore, the reflux average temperature rise offset also includes peak value retention correction:

[0026] The maximum measured temperature of the incoming air is extracted within the sliding window as the peak temperature of the window. The difference between the peak temperature of the window and the mean temperature of the window is defined as the peak-to-average deviation. After multiplying by the safety margin coefficient, it is superimposed on the mean temperature of the window to obtain the corrected mean. The corrected mean is used to replace the original mean in the calculation of the return flow mean temperature rise offset.

[0027] Furthermore, the steps for generating and outputting the target operating current setpoint for the water pump are as follows:

[0028] Set the target water supply temperature, reference ambient temperature, and corresponding reference operating current for constant temperature control. Use the difference between the equivalent ambient temperature and the reference ambient temperature as input, and output the current change through the reverse change compensation function. Add the current change to the reference operating current to obtain the initial target operating current of the water pump.

[0029] When a backflow self-excited oscillation is detected, a low-pass filter is activated with a cutoff period of a preset multiple of the backflow self-excited oscillation period to attenuate residual fluctuations; the filter is turned off when the outdoor wind speed is higher than the critical backflow wind speed threshold.

[0030] The filtered target operating current setting value of the water pump is output to the frequency converter driver. The frequency converter driver adjusts the output frequency according to the current setting value so that the water pump operating current tracks the setting value. At the same time, the deviation between the measured value of the water supply temperature and the target value is used as a feedback correction and added to the current setting value to realize feedforward-feedback composite constant temperature control.

[0031] An adaptive constant temperature control system for circulating water pumps based on ambient temperature includes:

[0032] The potential energy modeling module obtains the spatial structural parameters of the air conditioner cooler, constructs a three-dimensional thermal and humid air motion model coupled with exhaust and intake airflow, and generates a risk potential energy field for the return flow of exhaust thermal and humid air masses based on the three-dimensional thermal and humid air motion model.

[0033] The backflow detection module, based on the risk potential energy field of the exhaust hot and humid air mass backflow, combined with the real-time ambient temperature, the weather forecast temperature and the meteorological boundary conditions, performs backflow self-excited oscillation detection and outputs the corrected equivalent ambient temperature to eliminate backflow self-interference deviation.

[0034] The constant temperature control module uses the corrected equivalent ambient temperature as a feedforward variable and generates a target operating current setpoint for the water pump through a compensation function that compensates for the inverse change between the water pump operating current and the ambient temperature. After smoothing and filtering, the setpoint is output to the variable frequency drive of the circulating water pump. The variable frequency drive adjusts the output frequency according to the current setpoint so that the water pump operating current tracks the setpoint and maintains a constant cooling water supply temperature.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] This invention effectively solves the spatial mismatch and perception lag problem between the macroscopic ambient temperature perception and the actual intake microenvironment thermal field of the air conditioner cooler by constructing a potential energy model of the exhaust hot and humid air mass and performing backflow detection and analysis. Compared with traditional feedforward or pure feedback regulation that relies on a single meteorological parameter, this scheme can accurately identify and dynamically offset the local thermal self-interference caused by the exhaust backflow of the air conditioner cooler itself; by introducing the corrected equivalent ambient temperature as a feedforward variable into the control model, it realizes nonlinear compensation for abrupt changes in the intake boundary conditions, fundamentally eliminating the frequent overshoot of the control system and the blind following of control commands caused by airflow short-circuiting.

[0037] To address the intermittent self-excited oscillations of hot and humid airflow unique to low-temperature, windless conditions, the feedforward-feedback composite architecture of this invention achieves scale decoupling of the control system in terms of response speed and steady-state accuracy. This method can significantly suppress the periodic constant-amplitude oscillations of the variable frequency current of the circulating water pump under specific meteorological boundaries. It not only avoids ineffective adjustments and mechanical fatigue of the actuator due to sensing deviations, but also ensures that the cooling system can maintain the absolute thermal steady-state output of the supply water temperature in a physically meaningful sense under complex conditions where external meteorological parameters are in a pseudo-steady state while local heat load fluctuates. This greatly improves the robustness of the process cooling system to extreme boundary conditions. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. The following drawings are not drawn to scale according to the actual size, but are intended to show the main idea of ​​the present invention.

[0039] Figure 1 The flowchart shows an adaptive constant temperature control method for circulating water pumps based on ambient temperature.

[0040] Figure 2 This is a block diagram of an adaptive constant temperature control system for circulating water pumps based on ambient temperature.

[0041] Figure 3 A schematic diagram illustrating the definition of a three-dimensional cylindrical coordinate system and the source / sink surface;

[0042] Figure 4 This is a schematic diagram of the discretization of the emission source surface of a hot and humid air mass. Detailed Implementation

[0043] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are also within the scope of protection of the present invention.

[0044] Example 1.

[0045] like Figure 1 As shown, this invention includes an adaptive constant temperature control method for a circulating water pump, applied to a central air conditioning cooling water circulation system. The basic components of this circulation system are: an air conditioning cooler, a cooling water pump, a chiller unit condenser, and cooling water pipes connecting these devices, forming a closed cooling water circulation loop. The air conditioning cooler described in this invention refers to an evaporative cooler, which utilizes the evaporation and heat absorption of sprayed water on the outer surface of the heat exchange coil to cool the circulating cooling water within the coil. Its main body has a cylindrical or nearly cylindrical outer contour, with a circular exhaust port driven by a fan at the top and air inlet louvers around the lower perimeter. Air enters the interior of the main body through the air inlet louvers, flows through the heat exchange coil area, and is discharged vertically upwards from the exhaust port. The discharged air is saturated or nearly saturated hot and humid air after evaporative humidification. After the cooling water flows out of the heat exchange coil of the air conditioner cooler, it is pressurized by the cooling water pump and sent to the condenser of the chiller unit. In the condenser, it absorbs the condensation heat discharged from the chiller unit's refrigeration cycle and its temperature rises. Then it returns to the heat exchange coil of the air conditioner cooler to dissipate heat and cool down. The cooled cooling water flows out of the heat exchange coil again, completing a complete cycle.

[0046] In the aforementioned circulation loop, the air conditioning cooler plays a crucial role in dissipating heat from the cooling water into the atmosphere. The cooling capacity of the air conditioning cooler is directly affected by the ambient temperature: when the ambient temperature is low, the temperature difference between the air conditioning cooler and the ambient air is large, resulting in sufficient heat dissipation driving force and a significant temperature drop in the cooling water; conversely, when the ambient temperature is high, the temperature difference decreases, the heat dissipation driving force is insufficient, and the temperature drop of the cooling water is reduced. The core idea of ​​this invention is to use the ambient temperature as a feedforward variable to establish a compensating function relationship between the pump operating current and the ambient temperature—that is, reducing the pump operating current when the ambient temperature rises and increasing the pump operating current when the ambient temperature falls, thereby achieving adaptive constant temperature control of the circulating water pump.

[0047] However, in actual engineering operations, a phenomenon was observed under specific meteorological conditions: during cold, windless periods such as early winter mornings or early spring evenings, the cooling water supply temperature frequently oscillated within a small range of ±0.5 to 1 degree Celsius around the target value, and the water pump operating current also fluctuated periodically, with a fluctuation period of approximately 5 to 10 minutes. At this time, the outdoor temperature was stable, and according to the logic of ambient temperature feedforward control, the setpoint of the water pump operating current should also be constant and should not fluctuate. Through on-site observation and theoretical analysis, it was found that the root cause of this phenomenon was the intermittent recirculation of the hot, humid air discharged from the air conditioning cooler under windless conditions. This recirculated hot, humid air mass was re-drawn into the air inlet of the air conditioning cooler, causing a periodic deviation between the local ambient temperature at the air conditioning cooler inlet and the outdoor temperature measured by a distant meteorological station or outdoor sensor. This deviation prevented the compensation function, which used the outdoor temperature as the feedforward input, from accurately reflecting the actual heat dissipation environment temperature of the air conditioning cooler, thus causing fluctuations in the supply water temperature and the water pump current.

[0048] To address the aforementioned technical problems, the adaptive constant temperature control method for circulating water pumps based on ambient temperature provided in this embodiment of the invention includes three main steps: S10, S20, and S30. Step S10 constructs a three-dimensional thermal and humid air motion model coupling exhaust-intake airflow based on the spatial structural parameters of the air conditioner cooler, and generates a potential energy field for the risk of exhaust thermal and humid air mass recirculation. This output serves as the input for step S20. Step S20 performs recirculation self-excited oscillation detection based on this potential energy field and real-time meteorological conditions, outputting an equivalent ambient temperature corrected for recirculation self-interference deviation. This output serves as the input for step S30. Step S30 uses this corrected equivalent ambient temperature as a feedforward variable, generates a target operating current setpoint for the water pump through a reverse change compensation function, and outputs it to the frequency converter after smoothing and filtering. The frequency converter adjusts the output frequency to track this setpoint, achieving constant temperature control of the cooling water supply temperature. The following provides a detailed description of each step.

[0049] Step S10: Obtain the spatial structure parameters of the air conditioner cooler, construct a three-dimensional thermal and humid air motion model of exhaust-intake airflow coupling, and generate the risk potential energy field of exhaust thermal and humid air mass recirculation based on the three-dimensional thermal and humid air motion model.

[0050] The overall objective of step S10 is to obtain the spatial structural parameters of the air conditioner cooler, and on this basis, to construct a mathematical and physical model that can describe the movement behavior of hot and humid air discharged from the air conditioner cooler exhaust port in three-dimensional space. The model is then used to predict the risk distribution of the exhaust hot and humid air mass returning to the air inlet under windless or light wind conditions. Finally, the prediction results are output in the form of a return risk potential energy field, providing a theoretical basis for the return detection in the subsequent step S20.

[0051] Step S11: Obtain the vertical height of the air conditioner cooler exhaust vent from the roof, the exhaust vent cross-sectional diameter, the bottom edge height of the air inlet louvers, and their circumferential arc range; with the geometric center of the exhaust vent as the origin, vertically upward as... A three-dimensional cylindrical coordinate system is established, with the exhaust vent defined as the emission source surface of hot and humid air masses, and the circumferential space area covered by the air inlet louvers defined as the air mass capture and confluence surface.

[0052] like Figure 3 The diagram shown is a schematic representation of the three-dimensional cylindrical coordinate system and the source / sink plane definition.

[0053] Reference Figure 3 The model establishes a three-dimensional cylindrical coordinate system with the geometric center of the air conditioner cooler exhaust vent as the origin and the vertically upward z-axis as the z-axis. The figure shows the coordinate components consisting of radial distance, circumferential angle, and vertical height. The circular area in the upper plane of the figure is defined as the emission source surface of the hot and humid air mass, located on a plane with a vertical height of zero, representing the initial departure position of all exhaust hot and humid air. The annular cylindrical area in the lower plane of the figure is defined as the air mass capturing confluence surface, representing the circumferential space area covered by the air inlet louvers. Its radial distance fits against the outer wall of the model, and its vertical height is entirely in the negative value region of the z-axis. This figure visually illustrates the relative spatial structure between the emission source surface and the capturing confluence surface, providing a precise geometric analytical boundary for subsequent determination of whether the air mass trajectory crosses the confluence surface, i.e., whether backflow occurs.

[0054] Specifically, the spatial structural parameters of the air conditioning cooler are obtained. The air conditioning cooler involved in this invention is a mechanically ventilated evaporative cooler with a cylindrical or near-cylindrical outer contour and a circular exhaust port. The spatial structural parameters include the following: the vertical height of the air conditioning cooler exhaust port from the roof, the diameter of the exhaust port cross-section, the height of the bottom edge of the air inlet louvers from the roof, and the circumferential arc range of the air inlet louvers. The vertical height of the air conditioning cooler exhaust port from the roof refers to the vertical distance from the upper surface of the building roof to the center of the air conditioning cooler exhaust port cross-section. In actual engineering, this height is determined by the structural height of the air conditioning cooler body and the height of the air conditioning cooler's installation foundation on the roof. For common mechanically ventilated air conditioning coolers, the vertical height of the air conditioning cooler exhaust port from the roof is usually between 3 meters and 8 meters. This parameter is obtained through on-site measurement or by referring to the equipment installation drawings of the air conditioning cooler. The diameter of the exhaust port cross-section refers to the inner diameter of the exhaust port at the top of the air conditioning cooler. The fan is installed inside or directly above the exhaust vent. This parameter is obtained through on-site measurement or by consulting the equipment's technical specifications. For air conditioning coolers with a single unit handling cooling capacity ranging from 100 to 500 tons of refrigeration, the exhaust vent cross-sectional diameter is typically between 1.2 meters and 3.5 meters. The height of the bottom edge of the air inlet louvers from the roof refers to the vertical distance from the lowest edge of the air inlet louvers to the upper surface of the roof. The air inlet louvers of the air conditioning cooler are located around the lower part of the unit, and their function is to allow air to freely enter the air conditioning cooler while blocking direct sunlight and rainwater from entering. The height of the bottom edge of the air inlet louvers from the roof is typically 0.2 meters to 0.5 meters. In addition, the air inlet louvers also have a top edge height, which is the vertical distance from the highest edge of the air inlet louvers to the upper surface of the roof. The top edge height of the air inlet louvers can be obtained by adding the height of the louvers themselves to the height of the bottom edge of the air inlet louvers from the roof. The height of the louvers is obtained from the equipment drawings or on-site measurements. The circumferential arc range of the air inlet louvers refers to the angular range covered by the air inlet louvers on the outer circumferential surface of the air conditioner cooler body. For the circular cross-section air conditioner cooler involved in this invention, the air inlet louvers are typically arranged around the entire circumference of the body, and their arc range is 360 degrees.

[0055] After obtaining the above spatial structure parameters, the geometric center of the exhaust vent is taken as the origin of the coordinate system, and the vertical upward direction is used as the coordinate axis. Establish a three-dimensional cylindrical coordinate system along the positive axis. Since the air conditioner cooler involved in this invention has a circular cross-section, a circular exhaust vent, and air inlet louvers distributed circumferentially around the cylindrical body, all of the above geometric elements have... Due to the axisymmetric nature of the axis, a three-dimensional cylindrical coordinate system is chosen as the modeling coordinate system. This naturally matches the geometric symmetry of the air conditioner cooler, simplifying the calculation of subsequent motion trajectories and backflow determination. In this three-dimensional cylindrical coordinate system, the position of any spatial point is uniquely determined by three coordinate components: radial distance, circumferential angle, and vertical height. The radial distance represents the distance from the point to the vertical axis. The horizontal distance of the axis, the circumferential angle represents the rotation angle of the projection of the point on the horizontal plane relative to the reference direction, and the vertical height represents the height position of the point relative to the center plane of the exhaust vent. A positive vertical height indicates that the point is above the exhaust vent, and a negative vertical height indicates that the point is below the exhaust vent.

[0056] In the established three-dimensional cylindrical coordinate system, the exhaust vent is defined as the emission source surface of hot and humid air masses. Geometrically, the emission source surface is a circular region centered at the origin, with a radius of half the diameter of the exhaust vent cross-section, located on a plane with a vertical height of zero. The diameter of the exhaust vent cross-section is used here to determine the radius and area of ​​the emission source surface, which is used in subsequent step S12 to calculate the initial volume of each volumetric micro-element air mass. All hot and humid air discharged from the air conditioning cooler is discharged vertically upward from this emission source surface with a certain initial velocity. The circumferential space area covered by the air inlet louvers is defined as the air mass capturing confluence surface. The geometry of the air mass capturing confluence surface is centered around... The specific spatial range of the annular cylindrical region of the axis is defined by the following three conditions. The first condition is the radial distance condition: the radial distance of the air mass capturing surface is equal to the outer radius of the air conditioner cooler body, meaning the capturing surface adheres to the outer wall of the body. The second condition is the vertical height condition: the vertical height range of the air mass capturing surface extends from the vertical coordinate value corresponding to the bottom edge of the air inlet louver to the vertical coordinate value corresponding to the top edge of the air inlet louver. Since the air inlet louver is located below the exhaust vent, in a coordinate system with the exhaust vent center as the origin, the vertical coordinates of the bottom and top edges of the air inlet louver are both negative. Specifically, the vertical coordinate value corresponding to the bottom edge of the air inlet louver is equal to the negative value of the difference between the vertical height of the air conditioner cooler exhaust vent from the roof and the height of the bottom edge of the air inlet louver from the roof; the vertical coordinate value corresponding to the top edge of the air inlet louver is also equal to the negative value of the difference between the vertical height of the air conditioner cooler exhaust vent from the roof and the height of the top edge of the air inlet louver from the roof. The vertical height of the air conditioner cooler exhaust vent from the roof is used here to convert the relative height of the air inlet louvers to the roof into a vertical coordinate value in a three-dimensional cylindrical coordinate system. The third condition is the circumferential angle condition: since the air inlet louvers of the circular cross-section air conditioner involved in this invention are arranged around the entire circumference, and the arc range of the air inlet louvers along the circumference of the body is 360 degrees, the circumferential angle range of the air mass capture confluence surface covers the entire 360 ​​degrees. That is, air masses arriving at the capture confluence surface from any circumferential angle direction can be captured and drawn in by the air conditioner cooler inlet. The spatial region satisfying the above three conditions constitutes the air mass capture confluence surface, and any micro-element air mass whose trajectory crosses this confluence surface is determined to be a return air mass. The physical significance of establishing a three-dimensional cylindrical coordinate system and defining the emission source surface and the capture confluence surface is that it abstracts the spatial geometric relationship of the air conditioner cooler exhaust-inlet into a mathematically calculable three-dimensional spatial framework, enabling the calculation of the trajectory of hot and humid air masses and the determination of return flow in subsequent steps to be performed with precise geometric analysis under a unified coordinate system. The emission source surface represents the origin and initial condition application surface of the hot and humid air mass, while the capture confluence surface represents the receiving surface of the re-entering air mass into the air conditioning cooler. The spatial geometric relationship between the two directly determines the probability and severity of re-entry.

[0057] In existing technologies, the performance analysis of air conditioning coolers typically focuses only on the heat and mass exchange process within the heat exchange zone, treating the air conditioning cooler as a functional heat dissipation device without analyzing the behavior of its exhaust air in the external space. This step, by precisely quantifying the spatial geometric parameters of the air conditioning cooler's exhaust and inlet ports and establishing a unified coordinate system, lays the geometric foundation for describing the three-dimensional movement of exhaust air within the equipment space in subsequent steps. This is an analytical dimension not addressed in existing technologies.

[0058] Step S12: Discretize the hot and humid air discharged from the exhaust vent into a sequence of volumetric micro-element air masses. For each volumetric micro-element air mass, calculate the buoyancy-driven component based on its temperature difference with the ambient air. Correct the temperature decay path based on the release of latent heat of condensation of moisture content. Calculate the radial diffusion rate based on the turbulent diffusion coefficient. Combine these to obtain the spatial motion trajectory of each volumetric micro-element air mass in the three-dimensional cylindrical coordinate system as a function of time. The combination of the buoyancy-driven component, the temperature decay path, and the radial diffusion rate constitutes the three-dimensional hot and humid air motion model.

[0059] like Figure 4 The figure shows a schematic diagram of the discretization of the emission source surface of the hot and humid air mass.

[0060] Reference Figure 4 This figure illustrates the specific geometric layout of a spatial emission point array on a circular emission source surface of a continuous hot and humid airflow when the airflow is discretized. The outermost solid circle represents the exhaust port cross-section, forming the physical boundary of the emission source surface. Inside, several concentric rings are drawn at equal intervals from the center outwards using dashed lines, and several radial rays are drawn at equal angular intervals using solid lines. The solid dots at the intersections of the concentric rings and radial rays, as well as the solid dot at the center of the outermost circle, together constitute the emission points of the micro-element airflow. This discretized grid determines the release position of the initial micro-element airflow, allowing each discrete airflow to be assigned a uniform initial volume and independently carry state variables into the three-dimensional spatial motion model. This lays the foundation for subsequent calculation of independent return trajectories using the Lagrange particle tracking method.

[0061] The correction of the temperature decay path by the release of latent heat of condensation in step S12 is as follows: In each calculation time step, it is determined whether the current moisture content of the volumetric micro-element air mass exceeds the saturated moisture content corresponding to the current air mass temperature; if it exceeds, the mass of the excess water vapor is multiplied by the latent heat of vaporization to obtain the latent heat of condensation released in that time step, and the latent heat of condensation is converted into the increase in air mass temperature, which is added to the cooling amount caused by sensible heat exchange to obtain the net temperature change of the air mass in that time step; at the same time, the mass of condensed water vapor is deducted from the moisture content of the air mass to update the remaining moisture content of the air mass; when the moisture content of the air mass drops below the saturated moisture content, the condensation process terminates, the air mass temperature decay is driven only by sensible heat exchange, the cooling rate accelerates, the buoyancy decay accelerates, and the air mass enters the rapid sinking stage.

[0062] The goal of this step is to construct a three-dimensional model of hot and humid air motion. The core idea of ​​this model is to consider the continuous hot and humid air flow discharged from the exhaust vent as a sequence of air masses composed of numerous discrete micro-elemental air masses. For each micro-elemental air mass, a motion equation is established in three-dimensional space, and the spatial trajectory of each air mass over time is obtained by solving the motion equation. The hot and humid air discharged from the exhaust vent is discretized into a sequence of micro-elemental air masses. The analysis of discretizing the continuous hot and humid air flow into a finite number of micro-elemental air masses is based on the fundamental idea of ​​the Lagrange particle tracking method: in fluid motion analysis, a tiny volumetric unit within the fluid can be selected as the analysis object, and the motion and state changes of this tiny volumetric unit in space over time can be tracked. The macroscopic characteristics of the overall flow field are reconstructed by statistically summarizing a large number of tiny volumetric units. For the hot and humid airflow of the air conditioner cooler exhaust, which has the characteristics of buoyancy drive and phase change heat transfer, the advantage of using the Lagrange particle tracking method is that each volumetric micro-element can independently carry its own state variables such as temperature and humidity, and independently update its state according to the local environmental conditions during the movement. There is no need to solve the continuous field equations of the entire space, which greatly reduces the computational complexity while maintaining the integrity of the physical process.

[0063] The discretization is performed as follows: A spatial emission point array is set on the emission source surface of the hot and humid air mass defined in step S11. The emission source surface of the hot and humid air mass is a circular region with the origin as the center and half the diameter of the exhaust port cross-section as the radius. Within this circular region, several layers of concentric rings are arranged at equal intervals from the center to the outer edge in the radial direction. The radius of each layer of concentric rings increases sequentially. The radius of the innermost concentric ring is equal to half the diameter of the exhaust port cross-section divided by the number of concentric rings, and the radius of the outermost concentric ring is equal to half the diameter of the exhaust port cross-section. Several radial rays originating from the center are arranged at equal angular intervals in the circumferential direction. The angle between two adjacent radial rays is equal to 360 degrees divided by the total number of radial rays. The intersection of each layer of concentric rings and each radial ray constitutes an emission point. In addition, an extra emission point is set at the center of the circular region. Therefore, the total number of emission points equals the number of concentric ring layers multiplied by the total number of radial rays, plus one emission point at the center. In a specific embodiment, for an air conditioner cooler with an exhaust vent diameter of 2 meters, 5 concentric ring layers and 12 radial rays can be set. The radii of the 5 concentric ring layers are 0.2 meters, 0.4 meters, 0.6 meters, 0.8 meters, and 1.0 meters, respectively, and the angular interval of the 12 radial rays is 30 degrees. The total number of emission points is 5 multiplied by 12 plus 1, which equals 61. The number of concentric ring layers and the number of radial rays are determined by those skilled in the art based on the required computational accuracy and computational resources. The larger the number of layers and rays, the denser the emission points and the higher the model resolution, but the computational load also increases accordingly. In the time dimension, a new volumetric micro-element gas cloud is released simultaneously from each emission point periodically according to a preset time step. The time step is typically between 0.5 seconds and 2 seconds. Each released volumetric micro-element gas cloud is given the same initial volume. The basis for assigning the same initial volume is that, on the exhaust vent cross-section, the airflow velocity distribution driven by the fan is approximately uniform, and the air volume flow rate per unit area is basically consistent. Therefore, after dividing the emission source surface evenly according to the emission points, the area of ​​the local region represented by each emission point is approximately equal, and the air volume passing through each local region within the same time step is also approximately equal. The initial volume of each micro-element air mass is equal to the total area of ​​the emission source surface divided by the total number of emission points, multiplied by the initial vertical exhaust velocity of the air at the exhaust vent, and then multiplied by the time step. Each micro-element air mass has the following initial state parameters at the initial moment of leaving the emission source surface: initial temperature, initial humidity, and initial vertical exhaust velocity. The initial temperature is equal to the exhaust air temperature at the exhaust vent of the air conditioner cooler, which is usually 30 to 38 degrees Celsius. This temperature can be obtained by actual measurement using a temperature sensor installed at the exhaust vent, or it can be calculated based on the operating parameters of the air conditioner cooler.The initial moisture content is equal to the moisture content of the air at the exhaust vent. Since the exhaust air comes into full contact with the spray water as it passes through the heat exchange coil area, the relative humidity of the exhaust air is usually close to 100%, i.e., approximately saturated. Therefore, the initial moisture content can be taken as the saturated moisture content at the initial temperature. The specific value of the initial moisture content is determined as follows: Refer to the saturated moisture content curve on the enthalpy-humidity chart based on the initial temperature, and read the corresponding saturated moisture content value for that temperature. The enthalpy-humidity chart is a standard tool in the field of HVAC engineering. Its horizontal axis represents moisture content, and its vertical axis represents enthalpy. The chart includes isotherms and saturation curves. On the saturation curve, any temperature value corresponds to a unique saturated moisture content value. For example, when the initial temperature is 35 degrees Celsius, the corresponding saturated moisture content found on the saturation curve is approximately 36.6 grams per kilogram of dry air. The initial vertical discharge velocity is equal to the air velocity at the exhaust vent, which is obtained by dividing the rated airflow of the air conditioner cooler fan by the cross-sectional area of ​​the exhaust vent, and is typically between 4 and 10 meters per second. The cross-sectional area of ​​the exhaust vent is calculated from the diameter of the exhaust vent, and is equal to pi multiplied by the square of half the diameter of the exhaust vent.

[0064] For each volumetric micro-element air mass, an equation of motion is established in a three-dimensional cylindrical coordinate system. This equation describes the evolution of the position coordinates and state variables of the volumetric micro-element air mass over time. The position coordinates include radial distance. Circumferential angle and vertical height Three spatial components, with state variables including air mass temperature. Moisture content of air mass Among them, the temperature of the air mass The air temperature of a micro-elemental air mass at a given moment, expressed in degrees Celsius; the moisture content of the air mass. This represents the mass of water vapor contained in a unit mass of dry air within a micro-elemental air mass at a given moment, expressed in grams per kilogram of dry air. The equation of motion consists of three components: the buoyancy-driven component, the temperature decay path, and the radial diffusion rate. The buoyancy-driven component refers to the net buoyancy force gained by the air mass due to its temperature being higher than the ambient temperature, and the resulting vertical motion. Specifically, it manifests as the changes in net buoyancy acceleration, vertical velocity, and vertical position over time. The temperature decay path refers to the dynamic process of the air mass temperature gradually decreasing from its initial temperature to the ambient temperature, including the combined effects of cooling due to sensible heat exchange and warming due to the release of latent heat of condensation. The radial diffusion rate refers to the random diffusion motion of the air mass in the horizontal radial direction due to atmospheric turbulence, specifically manifested as the random displacement increment of the air mass's radial position within each time step. The first part is the buoyancy-driven component. The buoyancy-driven component determines the vertical motion behavior of the micro-elemental air mass. When the temperature of an air mass is higher than that of the surrounding air, its density is lower than that of the surrounding air. According to Archimedes' principle of buoyancy, the less dense air mass experiences a net upward buoyant force in the denser surrounding air. The net buoyant acceleration in the vertical direction... The calculation is based on the Businesk approximation: for cases where the density difference caused by temperature difference is small, buoyancy acceleration can be approximated as gravitational acceleration multiplied by the relative density difference between the air mass and the environment. Under the ideal gas assumption, the relative density difference equals the temperature difference divided by the absolute temperature of the environment. Therefore, the net buoyancy acceleration in the vertical direction... The calculation method is as follows:

[0065]

[0066] in The acceleration due to gravity is taken as 9.81 meters per second squared. The temperature of the volumetric air mass at the current moment is expressed in degrees Celsius. Ambient air temperature, in degrees Celsius; To convert ambient air temperature from Celsius to Kelvin as an absolute temperature value. When the air mass temperature... Higher than ambient temperature At that time, net buoyancy acceleration A positive value means the acceleration is vertically upward, driving the air mass upward; when the air mass temperature is... Equal to ambient temperature hour, When the temperature is zero, the air mass experiences no net vertical force; when the air mass temperature... Below ambient temperature hour, A negative value indicates that the acceleration is vertically downward, driving the air mass to sink.

[0067] The velocity of a micro-elemental air mass in the vertical direction Updated step-by-step according to Newton's second law of motion. Indicates the time of air mass vertical velocity, This indicates that the air mass has elapsed for a certain time step. The next moment The vertical velocity. Among them... Indicates the current calculation time. This represents the next calculation time after advancing one time step from the current time. The update method is as follows:

[0068]

[0069] The formula means that the vertical velocity at the next moment is equal to the current vertical velocity plus the current net buoyancy acceleration multiplied by the time step, resulting in the velocity increment. The position of the volumetric micro-elemental air mass in the vertical direction. It is also updated step-by-step. Indicates the time of air mass The vertical height, Indicates the air mass at the next moment The vertical height. The update method is as follows:

[0070]

[0071] The formula means that the vertical height at the next moment is equal to the vertical height at the current moment plus the displacement obtained by multiplying the vertical velocity at the current moment by the time step, plus the additional displacement generated within one time step under the action of net buoyancy acceleration. The air mass at the initial moment, i.e., time step... Vertical velocity when equal to zero Equal to the initial vertical discharge velocity, initial vertical position A value equal to zero indicates that the emission source is located on the surface of the emission source.

[0072] The second part is the temperature decay path. The temperature decay path describes the air mass temperature. The temperature of the air mass gradually decreases from its initial temperature. The temperature change at each time step is determined by the combined effects of cooling due to sensible heat exchange and warming due to the release of latent heat of condensation. First, we calculate the cooling caused by sensible heat exchange. Sensible heat exchange is a process of convection and mixing heat transfer between a small volumetric air mass and the surrounding cold air, driven by the temperature difference. The warmer air mass releases heat to the cooler ambient air, causing the air mass temperature to decrease. The cooling caused by sensible heat exchange is equal to the sensible heat exchange rate coefficient multiplied by the temperature difference between the air mass and the environment, and then multiplied by the negative of the time step.

[0073]

[0074] in This represents the change in air mass temperature caused by sensible heat exchange within a time step; a negative value indicates cooling. This is the sensible heat exchange rate coefficient, measured in seconds. The physical meaning of sensible heat exchange rate coefficient is the proportion of the temperature decrease of an air mass due to sensible heat exchange per unit time relative to the current temperature difference. The magnitude of this coefficient depends on the equivalent diameter of the air mass and the intensity of convective mixing between the air mass and the surrounding air. The equivalent spherical diameter of the air mass is determined as follows: The equivalent spherical diameter is calculated based on the initial volume of the air mass, i.e., the volume of the air mass is equivalent to the volume of a sphere, and then the diameter is calculated in reverse. Then, based on the equivalent spherical diameter and atmospheric turbulence conditions, the average convective heat transfer coefficient of the air mass surface is calculated using the empirical correlation of forced convection heat transfer in still or slightly moving ambient air. Specifically, the Reynolds number is first calculated based on the equivalent spherical diameter and the relative velocity between the air mass and the ambient air. The Reynolds number equals the equivalent spherical diameter multiplied by the relative velocity and then divided by the kinematic viscosity of the air. Finally, the Prandtl number of the air is obtained based on the ambient air temperature. The Prandtl number is the product of the kinematic viscosity and thermal diffusivity of the air. The ratio of the dispersion coefficients is approximately 0.71 within the normal temperature range. The Nusselt number is then calculated using the Ranz-Marshall correlation, which is: the Nusselt number equals 2.0 plus 0.6 multiplied by the Reynolds number to the power of 0.5, then multiplied by the Prandtl number to the power of 0.33. Finally, according to the definition of the Nusselt number, multiplying the Nusselt number by the thermal conductivity of air and dividing by the equivalent spherical diameter yields the average convective heat transfer coefficient of the air mass surface, with dimensions in watts per square meter per degree Celsius. Multiplying the average convective heat transfer coefficient by the ratio of the air mass's surface area to its volume and dividing by the product of the air density and its specific heat capacity at constant pressure yields the sensible heat exchange rate coefficient. For micro-elemental air masses with an equivalent spherical diameter ranging from 0.1 to 0.5 meters, under stable atmospheric conditions in winter, The typical value range is 0.01 to 0.05 per second. Calculate the temperature rise caused by the release of latent heat of condensation. At each calculation time step, determine the current moisture content of the volumetric micro-element gas mass. Does it exceed the current air mass temperature? The corresponding saturated moisture content. The method for determining the saturated moisture content corresponding to the current air mass temperature is the same as the method for determining the initial moisture content: based on the current air mass temperature... Consult the saturated moisture content curve on the enthalpy-humidity diagram of moist air and read the saturated moisture content value corresponding to that temperature. Since the air mass temperature continuously decreases during each time step, and the saturated moisture content decreases with decreasing temperature, even if the air mass moisture content remains unchanged, the saturated moisture content corresponding to the current air mass temperature continues to decrease as the air mass temperature decreases, gradually increasing the possibility that the air mass moisture content exceeds the saturated moisture content. If the current moisture content of a micro-element air mass... Not exceeding the current air mass temperature The corresponding saturated moisture content indicates that condensation does not occur within that time step, and the condensation temperature rise is... The moisture content of the air mass is equal to zero. It remains constant within this time step. If the current moisture content of the volumetric micro-elemental air mass... Exceeding the current air mass temperature At the corresponding saturated moisture content, condensation occurs. Multiplying the mass of the excess water vapor by the latent heat of vaporization yields the latent heat of condensation released during that time step. Specifically, the excess water vapor mass is the condensate amount. Equal to the current moisture content of the air mass Subtract the current air mass temperature The corresponding saturated moisture content. Condensation amount. Multiplying this by the latent heat of vaporization of water yields the latent heat of condensation released during that time step. The latent heat of vaporization of water is taken as approximately 2500 kJ per kilogram. This latent heat of condensation is then converted into the increment of the gas mass temperature rise. The conversion method is to divide the latent heat of condensation by the isobaric specific heat capacity of moist air, which is approximately 1.005 kJ / kg / degreeC. That is:

[0075]

[0076] in The condensate is the mass of the excess water vapor. The latent heat of vaporization of water, This is the isobaric specific heat capacity of moist air. A positive value indicates that the heat released during condensation raises the temperature of the air mass. Simultaneously, the mass of condensed water vapor is subtracted from the air mass moisture content, updating the remaining moisture content of the air mass. The updated air mass moisture content... Equal to the current air mass temperature The corresponding saturation moisture content is the moisture content of the air mass that the condensation process reduces to the saturation value at the current temperature. Therefore, the moisture content of the air mass... It decreases within each time step in which condensation occurs, and the amount of decrease is equal to the amount of condensation at that time step. As time progresses, the moisture content of the air mass... Gradually decrease. The temperature rise increment of the air mass, calculated by converting the latent heat of condensation, is... Cooling caused by superposition of sensible heat exchange The above yields the net temperature change of the air mass within that time step. The air mass temperature is updated as follows:

[0077]

[0078] in Let t be the temperature of the air mass at the current calculation time. Let be the temperature of the air mass at the next moment after a time step Δt; during the stage where condensation is occurring, The positive part canceled out The negative value of makes the absolute value of the net temperature change of the air mass within that time step smaller than the absolute value of the cooling rate when only sensible heat exchange occurs. The net rate of temperature decrease of the air mass is significantly slower than the pure sensible heat cooling rate without condensation. Because the air mass temperature decreases slowly, the temperature difference between the air mass and the environment remains at a higher level for a longer period of time, resulting in a higher net buoyancy acceleration in the vertical direction calculated in Part 1. It also means that the air mass remains at a higher positive value for a longer period, extending the time it takes for the air mass to rise or slowly descend in the air, resulting in a longer period of high-altitude residence and a greater likelihood that it will eventually descend back to the inlet area, i.e., the air mass capture and confluence surface. When the air mass moisture content... After a gradual condensation process over multiple time steps, the vapor density drops below saturation moisture content, at which point the condensation process terminates. Thereafter, no further condensation occurs in any of the time steps, and the vapor mass temperature increases. When the temperature of the air mass remains constant at zero, the temperature decay is driven solely by sensible heat exchange, leading to a faster cooling rate, accelerated buoyancy decay, and the air mass entering a rapid settling phase. The characteristics of this rapid settling phase are: the air mass temperature rapidly approaches or even falls below the ambient temperature, and the net buoyancy acceleration in the vertical direction... The vertical velocity becomes zero or negative. When a positive value changes to a negative value, the air mass changes from rising to sinking, and its vertical position changes. It begins to decrease rapidly. Based on the above process, the moisture content of the air mass... The dynamic change pattern is as follows: at the initial moment This equals the initial moisture content, i.e., the saturated moisture content at the initial temperature. In each subsequent time step, this is first determined based on the updated air mass temperature. Find the new saturated moisture content value, if If the moisture content exceeds the saturation moisture content value, condensation will occur, and the amount of condensation will be deducted. If the moisture content drops to this saturation value, If the saturated moisture content is less than or equal to this value, then It remains unchanged. Because the air mass temperature generally decreases, the saturated moisture content also decreases accordingly, and each condensation will... The humidity level was adjusted precisely to the saturation point at the current temperature, therefore Throughout the entire process, the temperature decreases gradually in a step-like manner, with each step corresponding to a condensation subtraction at a given time step. When the decrease in air mass temperature within a certain time step is small enough that the new saturated moisture content is still greater than or equal to the current... At that time step, no condensation occurs. The temperature remains unchanged until the temperature decreases further in subsequent time steps, causing the saturated moisture content to fall below the threshold again. It takes time for condensation to resume. The aforementioned temperature decay path and its latent heat of condensation correction mechanism are crucial to the accuracy of the three-dimensional thermal and humid air motion model. The moisture content of air conditioning cooler exhaust air is much higher than that of hot air in the general atmospheric environment. The buoyancy maintenance effect of latent heat release on the exhaust air mass is a key factor determining the residence height and settling time of the air mass. If the latent heat of condensation correction is ignored and the temperature decay path is calculated only based on sensible heat exchange, the model will severely underestimate the residence time of the air mass, leading to an underestimation of the risk of backflow. In low-temperature winter environments, the ambient temperature is much lower than the exhaust air temperature. A large amount of water vapor in the exhaust air mass begins to condense shortly after being discharged, releasing a considerable amount of latent heat, which has a particularly significant effect on correcting the temperature decay path.

[0079] The third part is the radial diffusion rate. The radial diffusion rate describes the random diffusion motion of a volumetric air mass in the horizontal radial direction due to atmospheric turbulence, as well as its directional drift motion due to horizontal winds. During its ascent and descent, the air mass is gradually diffused horizontally due to random disturbances from atmospheric turbulence, and if horizontal winds are present, it will also undergo directional translation along the wind direction. The radial displacement increment of the air mass within each time step is also described. The calculation method is as follows:

[0080]

[0081] The first term is the directional drift component caused by horizontal wind. For horizontal wind speed, The wind direction angle is the circumferential angle of the horizontal wind in a three-dimensional cylindrical coordinate system. This represents the current circumferential angle of the air mass. The cosine of the angle between the wind direction and the radial direction of the air mass projects the horizontal wind speed onto the radial direction of the air mass. The time step is the horizontal wind speed. When the first term is zero (i.e., under windless conditions), the radial displacement increment is determined solely by the turbulent random diffusion component of the second term. The turbulent diffusion coefficient in the horizontal direction is expressed in square meters per second. The random sample value follows a standard normal distribution, meaning a value is randomly selected from a normal probability distribution with a mean of zero and a standard deviation of one. This value is independently and randomly generated for each air mass within each time step, used to simulate the random influence of atmospheric turbulence on the horizontal displacement of the air mass. This formula originates from the discretized form of the Wiener process in random walk theory, superimposed with deterministic advection terms, and is a standard mathematical method describing the coupling process of turbulent diffusion and advection transport. Simultaneously, horizontal winds also cause changes in the circumferential angle of the air mass. The update method for the circumferential angle of the air mass within each time step is based on the horizontal wind speed... Calculate the circumferential displacement increment using the tangential component at the current position of the air mass. The circumferential displacement increment is equal to... Multiply Multiply by Divide by the current radial distance of the air mass The circumferential angle of the air mass is updated at each time step to the current circumferential angle plus the circumferential displacement increment. When the horizontal wind speed... When the coefficient is zero, the circumferential displacement increment is zero, and the circumferential angle of the air mass remains unchanged. The turbulent diffusion coefficient in the horizontal direction... Atmospheric stability is determined by: using real-time meteorological data or historical statistical data provided by local meteorological stations to determine the atmospheric stability level for the current period. Atmospheric stability levels are classified according to the Pasquale-Gifford classification system. to There are six levels in total, among which The level is highly unstable. The level is neutral. The atmospheric boundary layer is stable. During the cold, windless periods of winter mornings or early spring evenings, as described in this invention, the atmospheric boundary layer is typically in a stable state. level or The stability stratification state is determined at level [number]. Based on the determined stability level, refer to the Pasquale-Gifford diffusion parameter table for the corresponding range of horizontal diffusion coefficient values. For [specific value]... Level and Level stability conditions, Typical values ​​for the turbulent diffusion coefficient range from 0.5 to 2.0 square meters per second. A smaller turbulent diffusion coefficient means that the horizontal diffusion rate of the air mass is slower, and the air mass can maintain its spatial concentration better during ascent and descent, making it less likely to be diffused by turbulence to areas far from the air conditioner cooler. This is one of the physical reasons why backflow phenomena are more likely to occur under stable atmospheric conditions. The radial position of the air mass is updated in each time step to the current radial distance plus the radial displacement increment. The circumferential angle of the air mass remains unchanged under windless conditions.

[0082] Combining the buoyancy-driven component, the temperature decay path, and the radial diffusion rate, we construct a complete set of equations of motion for each volumetric gas cloud in three-dimensional cylindrical coordinates. This set of equations describes five variables of the gas cloud, including its vertical position. Vertical velocity radial distance air mass temperature Moisture content of air mass How to update the values ​​for the next time step based on the current values ​​and environmental conditions within each time step? The calculation process of this system of equations of motion within each time step is as follows: First, based on the current air mass temperature... and ambient temperature Calculate the net buoyancy acceleration in the vertical direction Update vertical speed and vertical position The second step is to calculate the cooling rate caused by sensible heat exchange. Determine the current moisture content of the air mass. Does it exceed the saturation moisture content corresponding to the current air mass temperature? If so, calculate the condensation amount. and condensation temperature rise And update the air mass moisture content. ,Will and The net temperature change of the air mass is obtained by superposition and the air mass temperature is updated. The third step is to determine the turbulent diffusion coefficient in the horizontal direction. and standard normal distribution random sample values Calculate radial displacement increment Update radial distance The above three steps are executed sequentially within each time step, with the result of the previous step serving as the input for the next. This process proceeds step by step, representing the successive numerical integration solution. Starting from the initial time step (i.e., time...) The trajectory calculation for the air mass begins at zero and continues until one of the following termination conditions is met: air mass temperature. Drop to ambient temperature Below and vertical velocity A negative value means the air mass has lost buoyancy and is sinking, subsequently reaching a vertical position. The air mass falls below the vertical coordinate value corresponding to the bottom edge of the air inlet louvers; or the radial distance of the air mass... If the air mass exceeds the preset upper limit of the computational domain radius, it means that the air mass has diffused to a region far away from the air conditioner cooler and will not flow back. The preset upper limit of the computational domain radius is usually taken as 5 to 10 times the outer wall radius of the air conditioner cooler. The sequence of position coordinates recorded for each volumetric micro-element air mass from release to termination in all time steps constitutes the spatial motion trajectory of that air mass. The set of spatial motion trajectories of all air masses released from all emission points in all time steps is the calculation output of the three-dimensional thermal and humid air motion model.

[0083] The combination of the aforementioned buoyancy-driven component, temperature decay path, and radial diffusion rate constitutes the three-dimensional thermal and humid air motion model described in this invention. The physical significance of this model lies in its transformation of the motion behavior of air conditioning cooler exhaust air in external three-dimensional space from an invisible and unquantifiable natural process into a calculable and predictable mathematical-physical model based on the spatial structural parameters of the air conditioning cooler and meteorological condition parameters. Through this model, the trajectory distribution of the exhaust air mass can be predicted under different meteorological boundary conditions, thereby determining whether backflow is possible and its severity.

[0084] In existing technologies, the diffusion analysis of air conditioning cooler exhaust air is usually simplified by referring to the Gaussian diffusion model of industrial chimney exhaust. This type of model generally does not consider the latent heat release effect of phase change in the exhaust gas because the moisture content of industrial flue gas is much lower than that of air conditioning cooler exhaust air. This step introduces moisture content judgment and latent heat of condensation correction in the time-step iteration, and uses the Lagrange particle tracking method to independently solve the coupled motion equations containing temperature, moisture content and three-dimensional position for each gas cloud. This enables the model to accurately describe the unique motion characteristics of air conditioning cooler exhaust air, which has a high moisture content, in a low-temperature environment, filling the gap between the analysis of external exhaust air behavior of air conditioning coolers and water pump control strategies.

[0085] Step S13: Based on the spatial motion trajectory of all volumetric micro-element air masses output by the three-dimensional thermal and humid air motion model, perform geometric intersection determination with the air mass capture confluence surface, mark the air masses that cross the air mass capture confluence surface as return air masses and record their remaining temperature when crossing, statistically analyze the spatial distribution of return heat flux density per unit time, and generate the risk potential energy field of the exhaust thermal and humid air mass return.

[0086] The calculation of the recirculation heat flux density in step S13 is as follows: the gas mass capture sink is discretized into a rectangular array of elements along the circumferential arc direction and the height direction; for each recirculation gas mass that crosses the gas mass capture sink, the target element to which it belongs is determined according to the circumferential arc and height coordinates of its crossing position; the recirculation temperature rise contribution value is obtained by subtracting the ambient air temperature from the remaining temperature of the recirculation gas mass at the crossing time, and multiplied by the volumetric flow rate and the specific heat capacity of air at constant pressure of the recirculation gas mass to obtain the single-unit recirculation heat power carried by the recirculation gas mass; for each element, the single-unit recirculation heat power of all the recirculation gas masses belonging to it is accumulated within the statistical period, and divided by the geometric area of ​​the element and the duration of the statistical period to obtain the hourly average recirculation heat flux density value of the element.

[0087] Specifically, in step S12, the spatial trajectories of all volumetric micro-element air masses evolving over time in a three-dimensional cylindrical coordinate system have been obtained. The task of step S13 is to analyze these spatial trajectories, determine which air masses will flow back to the air inlet area of ​​the air conditioner cooler, and quantify the severity of the backflow in the form of spatial distribution. Geometric intersection determination with the air mass capture surface is performed on the spatial trajectories of all volumetric micro-element air masses one by one. The air mass capture surface has been defined in step S11 as a spatial region that simultaneously satisfies the radial distance condition, the vertical height condition, and the circumferential angle condition. The specific method for determining geometric intersection is as follows: For the coordinates of two adjacent time steps on the trajectory of each air mass, check if the following conditions exist: in the previous time step, the radial distance of the air mass is greater than the outer wall radius of the air conditioner cooler body, i.e., it is located outside the air mass capture confluence; in the next time step, the radial distance of the air mass is less than or equal to the outer wall radius of the air conditioner cooler body, i.e., it reaches or crosses the inner side of the air mass capture confluence. Simultaneously, within this time interval, the vertical height of the air mass is between the vertical coordinate values ​​corresponding to the bottom and top edges of the air inlet louvers. Since the air inlet louvers of the circular cross-section air conditioner cooler involved in this invention are arranged around the entire circumference, the circumferential angle condition is automatically satisfied, requiring no additional judgment. If the above radial distance and vertical height conditions are simultaneously satisfied, it is determined that the trajectory of the air mass has crossed the air mass capture confluence. For air masses determined to have crossed the air mass capture confluence, they are marked as return air masses. Two key pieces of information about the air mass when crossing the air mass capture confluence are also recorded: the remaining temperature at the time of crossing and the crossing position. The residual temperature at crossover refers to the temperature of the backflowing air mass at the time step when it crosses the air mass capture confluence. This temperature value reflects the level of heat retained by the returning air mass after heat dissipation and condensation during its movement through the air. The crossing position is represented by the circumferential angle and vertical height of the returning air mass at the moment of crossing. The circumferential angle at the crossing position is the circumferential radian coordinate, and the vertical height at the crossing position is the height coordinate. Then, the spatial distribution of the returning heat flux density of all returning air masses at the capture surface of the crossing air mass is statistically analyzed per unit time.

[0088] The specific implementation method for calculating the recirculation heat flux density is as follows: The air mass capturing surface is discretized into a rectangular array of elements along the circumferential arc direction and the height direction. Along the circumferential arc direction, the circumferential arc range (360 degrees) of the air mass capturing surface is divided into several segments at preset angular intervals. Along the height direction, the height range of the air mass capturing surface (the interval from the bottom vertical coordinate of the air inlet louver to the top vertical coordinate) is divided into several layers at preset height intervals, thus forming a two-dimensional rectangular array of elements. The geometric area of ​​each rectangular element is equal to the outer wall radius of the air conditioner cooler body multiplied by the circumferential arc covered by the element, and then multiplied by the height covered by the element. In a specific embodiment, the circumference can be divided into 12 segments at 30 degrees, and the height range of the air inlet louver can be divided into several layers at 0.1-meter layers, thus forming an array of elements containing dozens to hundreds of rectangular elements. For each recirculation air mass crossing the air mass capturing surface, its target element is determined based on the circumferential arc and height coordinates of its crossing position. The determination method is as follows: The circumferential radian coordinates of the crossing position of the recirculating air mass are compared with the coverage area of ​​each surface element in the circumferential direction. The height coordinates of the crossing position are also compared with the coverage area of ​​each surface element in the height direction. If the crossing position falls within both the circumferential and height ranges of a surface element, the recirculating air mass is determined to belong to that target surface element. The recirculation temperature rise contribution value is obtained by subtracting the ambient air temperature from the remaining temperature of the recirculating air mass at the moment of crossing. The physical meaning of the recirculation temperature rise contribution value is: the additional temperature increment brought in by the recirculating air mass after it is re-inhaled by the air inlet, relative to the normal ambient cold air. The recirculation temperature rise contribution value is multiplied by the volumetric flow rate of the recirculating air mass and the specific heat capacity of air at constant pressure to obtain the single-unit recirculation heat power carried by the recirculating air mass. The volumetric flow rate of the recirculating air mass is equal to the initial volume of the air mass divided by the time step. The specific heat capacity of air at constant pressure is taken as approximately 1.005 kJ / kg / degree Celsius. The physical meaning of single-unit recirculation heat power is: the additional heat input into the air conditioning cooler through the air inlet per unit time by the recirculating air mass. For each surface element, the single-unit recirculation heat power of all recirculating air masses is accumulated within the statistical period to obtain the total recirculation heat power received by that surface element within the statistical period. Dividing this by the geometric area of ​​the surface element and the duration of the statistical period yields the hourly average recirculation heat flux density value of that surface element, with dimensions in watts per square meter. The length of the statistical period is typically taken as the entire calculation time span from the release to the termination of the volumetric micro-element air mass in step S12.

[0089] The time-averaged recirculation heat flux density values ​​of all rectangular surface elements are arranged according to their positions on the air mass capture surface, forming a two-dimensional distribution field with circumferential arc and height as coordinates and recirculation heat flux density as a scalar value. This two-dimensional distribution field is the exhaust air hot and humid air mass recirculation risk potential energy field output in this step. The physical meaning of the exhaust air hot and humid air mass recirculation risk potential energy field is that it describes, in a quantified spatial distribution form, the degree to which each area of ​​the air conditioner cooler inlet may be affected by exhaust air recirculation under specific conditions. Areas with higher time-averaged recirculation heat flux density values ​​in the potential energy field indicate that the air inlet at that location is more severely affected by the recirculation hot and humid air mass, and the temperature of the drawn-in air deviates more from the ambient temperature; areas with lower or zero time-averaged recirculation heat flux density values ​​indicate that the location is less affected by recirculation or is unaffected. The potential energy field of the exhaust hot and humid air mass backflow risk provides a spatial reference benchmark for the backflow detection in the subsequent step S20, so that the inlet air temperature fluctuation detected in step S22 can be compared and verified with the backflow location predicted by the model, confirming that the fluctuation is indeed caused by exhaust air backflow rather than other environmental disturbance factors.

[0090] In existing technologies, there is no practice to quantify the spatial distribution characteristics of exhaust air recirculation from air conditioning coolers into a potential energy field and apply it to a water pump control system. Even when the existence of recirculation problems is recognized during the operation and maintenance of air conditioning coolers, physical measures such as installing air guides or increasing the spacing between air conditioning coolers are typically taken to mitigate the impact of recirculation, rather than using control strategies to adapt to and compensate for its effects. The potential energy field for the risk of exhaust hot and humid air mass recirculation generated in this step correlates physical spatial information with the control system input, providing an information dimension for temperature correction in subsequent steps that is not seen in existing technologies.

[0091] Step S20: Based on the potential energy field of the risk of backflow of the exhaust hot and humid air mass, combined with the real-time ambient temperature, the weather forecast temperature and the meteorological boundary conditions, perform backflow self-excited oscillation detection, and output the corrected equivalent ambient temperature to eliminate the backflow self-interference deviation.

[0092] The overall objective of step S20 is to use the potential energy field of the risk of exhaust hot and humid air mass recirculation generated in step S10 as a theoretical benchmark, and combine it with the real-time collected ambient temperature data, the meteorological forecast temperature data obtained through the meteorological service interface and the meteorological boundary condition data to determine whether exhaust air recirculation is currently occurring, and to quantitatively evaluate the local temperature deviation of the air inlet caused by the recirculation, and finally output the corrected equivalent ambient temperature for the subsequent pump current calculation in step S30.

[0093] Step S21: Obtain the real-time outdoor wind speed and the weather forecast temperature through the meteorological service interface. The real-time outdoor dry-bulb temperature and the weather forecast temperature are weighted and fused to obtain a comprehensive ambient temperature reference value. When the real-time outdoor wind speed is lower than the critical recirculation wind speed threshold, it is determined to be a high-incidence condition of recirculation. The critical recirculation wind speed threshold is determined by the height difference between the exhaust vent and the air inlet louver and the cross-sectional diameter of the exhaust vent. It is the minimum wind speed at which the farthest recirculation air mass is horizontally blown away from the capture radius before sinking to the air mass capture confluence.

[0094] Specifically, real-time outdoor wind speed data needs to be obtained. This can be achieved by directly measuring the wind speed using a wind speed sensor installed on the roof near the air conditioner cooler, or by obtaining real-time wind speed data from a local meteorological station or meteorological service platform via a meteorological data interface. If an on-site wind speed sensor is used, it should be installed in a location unaffected by the air conditioner cooler's exhaust or building obstructions, and its installation height should preferably be higher than the air conditioner cooler's exhaust vent to ensure that the measured wind speed represents the actual atmospheric wind speed level at the location of the air conditioner cooler.

[0095] Simultaneously, weather forecast temperature data for the current period and the preset future period are obtained from the meteorological service platform via a meteorological service interface. This forecast temperature data provides information on the trend of ambient temperature changes, compensating for the inadequacy of on-site temperature sensors, which only reflect instantaneous values ​​and lack predictive capabilities. The real-time outdoor dry-bulb temperature measured by the on-site outdoor temperature sensor is weighted and fused with the weather forecast temperature to obtain a comprehensive ambient temperature reference value. The specific weighting method is as follows: the comprehensive ambient temperature reference value equals the real-time outdoor dry-bulb temperature multiplied by the measured weighting coefficient, plus the weather forecast temperature multiplied by the forecast weighting coefficient, with the sum of the measured and forecast weighting coefficients equal to 1. The measured weighting coefficient typically ranges from 0.6 to 0.8, and the forecast weighting coefficient typically ranges from 0.2 to 0.4. When the weather forecast temperature indicates a significant change in ambient temperature within the preset future period, the forecast weighting coefficient can be appropriately increased to enhance the forward-looking response capability of the control system.

[0096] After obtaining the real-time outdoor wind speed, it is input into the three-dimensional thermal and humid air motion model constructed in step S12 as a horizontal advection constraint term. The horizontal advection constraint term works by using the real-time outdoor wind speed... and wind direction angle Substituting the directional drift component into the radial displacement increment formula in part three of step S12, each volumetric micro-element air mass is driven by directional drift of horizontal wind in each time step, in addition to being subject to turbulent random diffusion. The horizontal advection constraint term causes the air mass to drift horizontally simultaneously during its ascent and descent, thus deviating from its position directly above the air conditioner cooler. The higher the wind speed, the farther the air mass drifts horizontally within the same ascent or descent time, and the lower the probability that the air mass will fall back into the air inlet area of ​​the air conditioner cooler after descent.

[0097] After adding a horizontal advection constraint term to the three-dimensional thermal and humid air motion model, the spatial trajectory of the volumetric micro-element air mass is recalculated, and the geometric intersection determination with the air mass capture confluence is performed according to the method described in step S13. When the horizontal wind speed is high enough, the spatial trajectory of all air masses no longer crosses the air mass capture confluence, i.e., no backflow occurs. The minimum wind speed value that allows the farthest backflow air mass to be horizontally blown away from the capture radius of the air conditioner cooler before sinking to the air mass capture confluence is defined as the critical backflow wind speed threshold. The critical backflow wind speed threshold is determined as follows: In the three-dimensional thermal and humid air motion model, firstly, under windless conditions, i.e., with the horizontal wind speed set to zero, the spatial trajectory of all volumetric micro-element air masses is calculated according to the complete process described in steps S12 and S13, and the geometric intersection determination is performed. Among all volumetric micro-element air masses marked as backflow air masses, the air mass with the longest total movement time from the exhaust vent, rising to the highest point, sinking, and crossing the air mass capture confluence is identified, and this air mass is called the key air mass. The critical air mass represents the extreme recirculation scenario with the longest recirculation path and the longest residence time in the air under windless conditions. The total descent time is defined as the total time taken for the critical air mass to rise from the exhaust vent to the highest point of its spatial trajectory and then descend to the top edge of the air mass capture surface. The total descent time depends primarily on the rate of temperature decay calculated in step S12, particularly the effect of latent heat release on the duration of air mass buoyancy maintenance. The more latent heat released, the longer the air mass remains at high altitude, and the longer the total descent time. Simultaneously, the horizontal distance from the outer edge of the air inlet louvers of the air conditioner cooler to the central axis of the air conditioner cooler body (i.e., the z-axis of the three-dimensional cylindrical coordinate system) is measured; this distance is the capture radius. For a circular cross-section air conditioner cooler, the capture radius is equal to the outer wall radius of the air conditioner cooler body. The critical recirculation velocity threshold is equal to the capture radius divided by the total descent time. The physical meaning of the above calculation process is as follows: If the horizontal wind speed is exactly equal to the critical recirculation wind speed threshold, the total distance that the key air mass drifts horizontally during its entire total sinking time is exactly equal to the capture radius, that is, the key air mass is blown just above the outer edge of the air inlet louver without being sucked in; if the horizontal wind speed is greater than the critical recirculation wind speed threshold, the key air mass will be blown further away and will not be sucked into the air inlet, and recirculation will not occur; if the horizontal wind speed is less than the critical recirculation wind speed threshold, the horizontal drift distance of the key air mass is insufficient to make it leave the capture range, and recirculation may occur.

[0098] In a specific embodiment, for a typical air conditioning cooler with an exhaust vent 5 meters vertically above the roof, an exhaust vent cross-sectional diameter of 2 meters, and the top edge of the air inlet louvers 1 meter above the roof (i.e., a height difference of 4 meters between the exhaust vent and the top edge of the air inlet louvers), under winter ambient temperatures of 3 degrees Celsius, the total settling time of the critical air mass is approximately 25 to 40 seconds, and the capture radius is approximately 1.5 to 2.0 meters. The calculated critical return velocity threshold is approximately 0.5 to 1.0 meters per second. Those skilled in the art will understand that this threshold varies with the spatial structural parameters of the air conditioning cooler and the ambient temperature, and needs to be calculated and determined separately for each specific project using the above method based on the air conditioning cooler parameters.

[0099] When the real-time outdoor wind speed is below the critical return wind speed threshold, it is determined that the current condition is prone to return flow, and the return flow self-excited oscillation detection function in subsequent step S22 is activated. When the real-time outdoor wind speed is higher than or equal to the critical return wind speed threshold, it is determined that return flow will not occur, the return flow detection function in step S22 is in standby mode, and the corrected equivalent ambient temperature is directly output as the comprehensive ambient temperature reference value. The physical significance of step S21 is that it compares the real-time outdoor wind speed with the critical return wind speed threshold calculated based on the spatial structure parameters of the air conditioner cooler, thereby realizing the real-time determination of the return flow occurrence condition. This determination avoids performing unnecessary detection and correction calculations under windy conditions where return flow correction is not required, while ensuring that the detection mechanism is activated in a timely manner under windless or light wind conditions where return flow may occur.

[0100] In existing technologies, the operation control of air conditioning coolers typically does not consider the dynamic impact of wind speed on the heat dissipation efficiency of the air conditioning cooler, nor does it involve the real-time determination of recirculation conditions. This step establishes a critical wind speed threshold with clear physical meaning by combining the spatial structural parameters of the air conditioning cooler with meteorological boundary conditions, providing a criterion for on-demand activation in subsequent steps.

[0101] Step S22: Under the condition of high recirculation, continuously collect the measured temperature sequence of the air inlet temperature sensor at the air inlet louver of the air conditioner cooler, perform short-time sliding window spectrum analysis, extract the periodic fluctuation component within the preset oscillation period range, define the peak-valley temperature difference as the recirculation self-excited oscillation amplitude and the center period as the recirculation self-excited oscillation period, and compare and verify it with the recirculation heat flux density at the corresponding position in the recirculation risk potential energy field.

[0102] After step S21 determines that the current operating condition is prone to backflow, backflow self-excited oscillation detection is initiated. Specifically, the measured inlet air temperature data output by the inlet air temperature sensor installed at the air inlet louvers of the air conditioner cooler is continuously collected to form an inlet air measured temperature sequence. The inlet air temperature sensor should be installed on the inner or outer side of the air inlet louvers of the air conditioner cooler, adjacent to the louver blades, and the installation height should preferably be in the middle of the louver height range. The sensor installation location should be selected with reference to the spatial region corresponding to the rectangular surface element with a high average backflow heat flux density value in the potential energy field of the exhaust hot and humid air mass backflow risk generated in step S13. The sensor should be installed at the corresponding air inlet louver position in this region to ensure that the sensor can sensitively capture the temperature changes brought about by the backflow air mass. The sampling frequency of the inlet air temperature sensor should not be less than once per second to ensure that waveform details in the oscillation signal with a period of 5 to 10 minutes can be captured. The collected temperature data are arranged in chronological order to form an inlet air measured temperature sequence.

[0103] A short-time sliding window spectral analysis is performed on the measured inlet air temperature sequence. The implementation method for short-time sliding window spectral analysis is as follows: a sliding window of fixed length is set, and the window length should be at least three times greater than the upper limit of the preset oscillation period range to ensure that the window contains a sufficient number of complete oscillation periods for spectral analysis identification. The preset oscillation period range is determined based on the prediction results of the recirculating air mass movement cycle from the three-dimensional thermal and humid air motion model in step S10. The period of the recirculating self-excited oscillation mainly depends on the total movement time of the volumetric micro-element air mass after it is discharged from the exhaust port, rises to its highest point, and then sinks back to the air mass capture surface, i.e., the total sinking time mentioned in step S21, and the system response time required for the disturbance to cooling efficiency caused by the recirculating air mass after being drawn into the air inlet to be transmitted to the exhaust temperature change and re-excite the next recirculation. The sum of these two constitutes a complete oscillation period. In a specific embodiment, the preset oscillation period range is 3 to 15 minutes, i.e., 180 seconds to 900 seconds, then the length of the sliding window is three times the upper limit of the oscillation period, i.e., 2700 seconds, approximately 45 minutes. A Fast Fourier Transform (FFT) is performed on the measured intake air temperature sequence within a sliding window to convert the time-domain temperature signal into a frequency-domain spectral signal. Within the spectral signal, a significant energy peak is searched for within a frequency range corresponding to a preset oscillation period. This preset oscillation period range is from approximately 0.0011 Hz (900-second period) to approximately 0.0056 Hz (180-second period). If a significant energy peak is detected within this frequency range, and the signal-to-noise ratio (SNR) of the peak exceeds a preset SNR threshold, a periodic fluctuation component is determined to exist in the measured intake air temperature sequence. The SNR threshold is typically set to 3 to 5 times the average background noise level within this frequency range to be considered a significant peak. This threshold value is determined during system debugging based on the background noise level of the intake air temperature sensor under normal operating conditions without backflow.

[0104] Two key characteristic parameters are extracted from the detected periodic fluctuation components: the amplitude of the reflux self-excited oscillation and the period of the reflux self-excited oscillation. The amplitude of the reflux self-excited oscillation is defined as the absolute value of the temperature difference between the peak and trough values ​​of the periodic fluctuation component in the time domain. The peak value is the highest temperature reached by the periodic fluctuation component within one oscillation cycle, and the trough value is the lowest temperature reached within the same oscillation cycle. The period of the reflux self-excited oscillation is defined as the central period of the periodic fluctuation component, which is the reciprocal of the frequency corresponding to the energy peak in the spectral analysis.

[0105] After extracting the amplitude and period of the recirculation self-excited oscillation, a comparison and verification step is performed. The purpose of the comparison and verification is to confirm that the detected periodic fluctuation component is indeed caused by exhaust air recirculation, rather than by other environmental disturbances. The implementation method of the comparison and verification is as follows: Based on the installation position of the inlet air temperature sensor on the air mass capture surface, the circumferential arc coordinates and height coordinates corresponding to the installation position are determined. Then, in the exhaust air hot and humid air mass recirculation risk potential energy field generated in step S13, the rectangular element that the circumferential arc coordinates and height coordinates fall into is found, and the time-averaged recirculation heat flux density value of the rectangular element is read. If the time-averaged recirculation heat flux density value at this position is greater than zero, it indicates that the three-dimensional hot and humid air motion model in step S10 predicts that there is indeed a possibility of a recirculation air mass arriving at this position. In this case, the detected temperature fluctuation is consistent with the model prediction, and the verification is successful. If the average hourly return heat flux density at that location is zero or close to zero, it indicates that the model predicts no return air mass will arrive at that location. Therefore, the detected temperature fluctuations may be caused by other reasons, and the verification fails; they are not treated as return oscillations. The physical significance of the comparison and verification step lies in its use of the prediction results of the three-dimensional thermal and humid air motion model established in step S10 as prior knowledge to screen and differentiate the measured signals. Temperature fluctuations near the air inlet of the air conditioner cooler may be caused by various factors, such as hot airflow emitted from the building, heat exhaust from nearby air conditioner outdoor units, or localized temperature unevenness caused by solar radiation reflected off the building surface. Detecting periodic fluctuations solely through spectral analysis cannot determine their source. By comparing the spatial location information with the potential energy field of the risk of return air mass return, temperature fluctuations caused by return can be effectively distinguished from temperature fluctuations from other sources, improving the accuracy of return detection.

[0106] Temperature fluctuations at the air inlet of an air conditioning cooler are typically not monitored. Even when monitored, they are only used as alarm signals for abnormal cooler performance, rather than for analyzing the source of the fluctuations or for adjusting control strategies. By cross-validating the periodic fluctuation characteristics detected by spectral analysis with the spatial predictions from a three-dimensional physical model, accurate identification of the specific disturbance source—exhaust air recirculation—is achieved, providing a validated and reliable input for the equivalent ambient temperature correction in subsequent step S23.

[0107] Step S23: After confirming the existence of backflow self-excited oscillation through comparison and verification, the average time of a sliding window with a width that is an integer multiple of the backflow self-excited oscillation period is taken from the measured temperature sequence of the air intake. The difference between this average value and the comprehensive ambient temperature reference value is defined as the backflow average temperature rise offset. The corrected equivalent ambient temperature is output by superimposing the comprehensive ambient temperature reference value with the backflow average temperature rise offset.

[0108] After step S22 completes the detection of recirculation self-excited oscillation and passes the comparison verification, the corrected equivalent ambient temperature is calculated using the detected recirculation self-excited oscillation amplitude and period. The calculation method is as follows: the sliding window time average is taken from the measured inlet air temperature sequence. The width of the sliding window is set to a preset integer multiple of the recirculation self-excited oscillation period. The preset integer multiple is typically 2 to 5 times. Setting it to an integer multiple ensures that the sliding window covers exactly an integer number of complete oscillation periods, thus allowing the periodic fluctuation components within the window to be fully canceled out during time averaging, and the final average value reflects the overall drift level of the inlet air temperature after eliminating oscillation fluctuations. A larger preset integer multiple results in better smoothing and elimination of oscillation fluctuations, but the average value becomes less responsive to changes in recirculation intensity; a smaller preset integer multiple results in more sensitive tracking of recirculation intensity changes, but the elimination of oscillation fluctuations may be insufficient. In engineering practice, a value of 3 times achieves a good balance between smoothing effect and response speed. In a specific embodiment, if the reflux self-excited oscillation period detected in step S22 is 8 minutes, or 480 seconds, and the preset integer multiple is 3, then the sliding window width is 1440 seconds, or 24 minutes.

[0109] At each calculation moment, all measured air intake temperature data within a window width prior to the current moment are taken, and their arithmetic mean is calculated to obtain the sliding window time mean. This mean eliminates oscillations and fluctuations while retaining the overall temperature rise trend caused by recirculation. Then, the sliding window time mean is compared with the comprehensive ambient temperature reference value obtained in step S21. The comprehensive ambient temperature reference value is obtained by weighted fusion of real-time outdoor dry-bulb temperature and weather forecast temperature, and its specific acquisition and calculation method has been detailed in step S21. The difference between the sliding window time mean and the comprehensive ambient temperature reference value is defined as the recirculation mean temperature rise offset. The physical meaning of the recirculation mean temperature rise offset is: the increase in the average temperature of the air actually drawn into the air inlet of the air conditioner cooler due to exhaust recirculation compared to the pure ambient air temperature under conditions without recirculation. The magnitude of the recirculation mean temperature rise offset depends on the number of recirculating air masses and the residual temperature of the recirculating air masses as they cross the air mass capture surface, i.e., on the magnitude of the time-averaged recirculation heat flux density in the exhaust hot and humid air mass recirculation risk potential energy field calculated in step S13. The more severe the recirculation, the more recirculating air masses cross the air mass capture surface, the higher the residual temperature, and the higher the average temperature of the mixed air drawn in through the inlet, resulting in a larger recirculation mean temperature rise offset. In a typical winter low-temperature, windless condition, the recirculation mean temperature rise offset is usually between 1 and 5 degrees Celsius, with the specific value depending on the severity of the recirculation.

[0110] Finally, based on the comprehensive ambient temperature reference value, the return flow mean temperature rise offset is superimposed to obtain the corrected equivalent ambient temperature and output it. The calculated result of the corrected equivalent ambient temperature is higher than the comprehensive ambient temperature reference value; the difference between the two is the return flow mean temperature rise offset. The physical meaning of the corrected equivalent ambient temperature is that it is no longer the nominal temperature value measured by a distant weather station or a location unaffected by return flow, but rather the effective heat dissipation ambient temperature actually felt by the air conditioning cooler inlet after integrating real-time ambient temperature and weather forecast temperature trend information and eliminating return flow self-interference bias. Under the condition of return flow, the air drawn into the air conditioning cooler inlet contains a portion of high-temperature air masses from the exhaust air, making the heat dissipation effect of the air conditioning cooler equivalent to working in a higher-temperature environment. The corrected equivalent ambient temperature is a quantitative description of this actual heat dissipation environment.

[0111] The corrected equivalent ambient temperature is output to the subsequent step S30 as the feedforward input variable of the inverse change compensation function. This allows the calculation of the water pump operating current to be based on the actual heat dissipation environment conditions faced by the air conditioner cooler, rather than the nominal air temperature, thereby eliminating the impact of ambient temperature self-interference deviation caused by exhaust air recirculation on control accuracy. The above method of taking the sliding window time average has advantages in control logic: the sliding window time average output is a stable signal that changes slowly over time, without containing the periodic fluctuation component of recirculation oscillation. Therefore, the input signal received by the inverse change compensation function in the subsequent step S30 is stable, and the set value of the water pump target operating current is also stable, without repeatedly adjusting the water pump speed due to recirculation oscillation. This avoids mechanical wear and water hammer impact caused by frequent water pump speed changes, and also reduces the processing burden of the subsequent filtering stage.

[0112] Step S23 also includes peak value retention correction: extract the maximum measured temperature of the incoming air within the sliding window as the peak temperature of the window, define the difference between the peak temperature of the window and the mean temperature of the window as the peak-to-average deviation, multiply it by the safety margin coefficient and then add it to the mean temperature of the window to obtain the corrected mean temperature, and use the corrected mean temperature to replace the original mean temperature in the calculation of the return flow mean temperature rise offset.

[0113] The specific implementation of the peak hold correction step is as follows: The maximum measured intake air temperature within the sliding window is extracted as the window peak temperature. That is, within the time range of each sliding window, in addition to calculating the arithmetic mean of the measured intake air temperature sequence as the sliding window time mean, all measured intake air temperature data points within that window period are simultaneously traversed to find the maximum value, which is defined as the window peak temperature. The window peak temperature corresponds to the extreme intake air temperature at the moment of most severe backflow intrusion within that window period. The difference between the window peak temperature and the window mean is defined as the peak-to-average deviation. That is, the peak temperature is subtracted from the sliding window time mean to obtain the peak-to-average deviation. The peak-to-average deviation reflects the degree of one-sided deviation of the temperature peak relative to the mean during backflow oscillation. The larger the peak-to-average deviation, the higher the intake air temperature at the peak of backflow intrusion, and the more severe the heat dissipation environment experienced by the air conditioning cooler at that instant. The peak-to-average deviation is multiplied by a safety margin coefficient and then superimposed onto the window mean to obtain the corrected mean. The safety margin coefficient ranges from 0 to 1, typically between 0.3 and 0.5. A safety margin coefficient of 0 indicates no peak hold correction and output based entirely on the sliding window time average; a coefficient of 1 indicates output based entirely on the window peak temperature, equivalent to setting the pump current according to the worst-case operating condition. In engineering practice, a coefficient of 0.3 to 0.5 achieves a good balance between supply water temperature stability and energy saving. The specific value of the safety margin coefficient is determined during the system commissioning phase based on measured data of the supply water temperature fluctuation range. The peak hold correction is obtained by multiplying the peak-to-average deviation by the safety margin coefficient. This peak hold correction is then added to the sliding window time average to obtain the corrected sliding window average. The corrected average is used instead of the original average in the calculation of the return flow average temperature rise offset. Specifically, the corrected return flow average temperature rise offset is obtained by subtracting the comprehensive ambient temperature reference value from the corrected sliding window average. The corrected equivalent ambient temperature is then output by adding the corrected return flow average temperature rise offset to the comprehensive ambient temperature reference value. Because the corrected mean is higher than the original sliding window time mean, the final output corrected equivalent ambient temperature is also correspondingly higher. This causes the target operating current allocated to the water pump by the reverse change compensation function in subsequent step S30 to be slightly lower than the current value calculated using the pure mean. According to the reverse change characteristics of the reverse change compensation function in step S31, a higher corrected equivalent ambient temperature means that the compensation function outputs a larger negative current change, reducing the target operating current of the water pump, decreasing the water pump speed, and reducing the cooling water flow rate. However, under harsh operating conditions with high return inrush peak temperatures, this adjustment moderately increases the residence time of the cooling water in the air conditioning cooler, which is beneficial for obtaining more sufficient cooling in a single pass and provides a safety margin for constant temperature control of the water supply. The physical significance of peak hold correction is that although the pure sliding window time mean eliminates oscillation fluctuations, it also loses information about extreme operating conditions during the oscillation process.When the amplitude of the backflow self-excited oscillation is large, the heat dissipation capacity of the air conditioner cooler is significantly lower than the average level at the peak of the backflow influx. If the water pump current is set exactly according to the operating condition corresponding to the average, a brief overshoot of the supply water temperature may occur at the peak. Peak hold correction, by appropriately biasing the correction average towards the peak, keeps the water pump in an operating state slightly higher than the average requirement throughout the entire oscillation cycle, which can cover the heat dissipation demand at the peak and thus reduce the risk of instantaneous overshoot of the supply water temperature.

[0114] Step S30: Using the corrected equivalent ambient temperature as a feedforward variable, a target operating current setting value for the water pump is generated through a compensation function for the inverse change of the water pump operating current and the ambient temperature. After smoothing and filtering, the value is output to the variable frequency drive of the circulating water pump. The variable frequency drive adjusts the output frequency according to the current setting value so that the water pump operating current tracks the setting value and maintains a constant cooling water supply temperature.

[0115] Step S31: Set the target water supply temperature, reference ambient temperature and corresponding reference operating current for constant temperature control. Using the difference between the equivalent ambient temperature and the reference ambient temperature as input, output the current change through the reverse change compensation function, and add it to the reference operating current to obtain the initial target operating current of the water pump.

[0116] Specifically, three basic control parameters are set: the target supply water temperature for constant temperature control, the reference ambient temperature, and the reference operating current at the reference ambient temperature. The target supply water temperature for constant temperature control refers to the desired value of the cooling water supply temperature; that is, the temperature at which the control system needs to maintain the cooling water flowing out of the air conditioner cooler outlet near this value. This temperature setting needs to comprehensively consider the chiller's condensing temperature requirements and operating energy efficiency. In a specific embodiment, the target supply water temperature for constant temperature control is set to 32 degrees Celsius. The reference ambient temperature refers to the reference ambient temperature point selected when establishing the inverse change compensation function. This temperature is usually taken as the outdoor calculation temperature corresponding to the cooling system's design operating conditions, or it can be taken as the summer outdoor calculation dry-bulb temperature of the air conditioner in the project area. In a specific embodiment, the reference ambient temperature is set to 35 degrees Celsius. The reference operating current refers to the operating current value required for the water pump to maintain the cooling water supply temperature equal to the target supply water temperature for constant temperature control under the reference ambient temperature conditions. This value can be determined as follows: During the system commissioning phase, when the outdoor temperature is close to the reference ambient temperature, manually adjust the output frequency of the water pump inverter, observe the change in the cooling water supply temperature, find the inverter output frequency that stabilizes the supply water temperature at the target value, and record the water pump's operating current at this time as the reference operating current. In a specific embodiment, the reference operating current is 25 amperes.

[0117] After setting the above three parameters, a reverse change compensation function is established. This function takes the difference between the corrected equivalent ambient temperature and the reference ambient temperature as the input independent variable and the change in the water pump operating current as the output dependent variable. The core feature of the reverse change compensation function is that the input and output are inversely related. That is, when the corrected equivalent ambient temperature is higher than the reference ambient temperature, the input is positive and the output is a negative change in current, meaning that the water pump operating current should be lower than the reference operating current; when the corrected equivalent ambient temperature is lower than the reference ambient temperature, the input is negative and the output is a positive change in current, meaning that the water pump operating current should be higher than the reference operating current. The physical logic of the reverse change is as follows: When the ambient temperature rises, the temperature difference between the air intake air and the cooling water in the air conditioning cooler decreases, and the temperature drop that a unit flow rate of cooling water can obtain in the air conditioning cooler decreases. However, at the same time, due to the increase in ambient temperature, the cooling load on the condenser side increases, and the supply and return water temperature difference can be appropriately amplified. Under a larger temperature difference condition, the cooling water circulation flow rate required to maintain the same total cooling power decreases, and the water pump operating current can be reduced. Conversely, when the ambient temperature decreases, the air conditioner's cooling capacity increases, but the heat loss from pipes and equipment also increases. This necessitates increasing the circulation flow to compensate for the heat loss and prevent the cooling water from becoming too cold, which could trigger low-pressure protection of the chiller unit. Consequently, the pump operating current should be increased. The inverse compensation function can be a linear function, where the current change equals the compensation slope multiplied by the negative value of the difference between the corrected equivalent ambient temperature and the reference ambient temperature. The compensation slope is a positive constant, reflecting the sensitivity of the pump operating current to changes in ambient temperature. The compensation slope can be determined by recording the pump current required to maintain the target supply water temperature under different ambient temperatures during system commissioning and then performing linear regression fitting on these data. In a specific embodiment, the compensation slope is 0.5 amperes per degree Celsius, meaning that for every 1 degree Celsius increase in ambient temperature, the pump operating current decreases by 0.5 amperes.

[0118] The initial target operating current of the water pump is obtained by adding the reference operating current and the current change. In the above embodiment, when the corrected equivalent ambient temperature is 30 degrees Celsius, the difference from the reference ambient temperature of 35 degrees Celsius is -5 degrees Celsius, the current change is +2.5 amperes, and the initial target operating current of the water pump is 25 plus 2.5 equals 27.5 amperes. When the corrected equivalent ambient temperature is 38 degrees Celsius, the difference is +3 degrees Celsius, the current change is -1.5 amperes, and the initial target operating current of the water pump is 25 minus 1.5 equals 23.5 amperes.

[0119] Step S31 uses the corrected equivalent ambient temperature as input instead of the simple outdoor air temperature, ensuring that the ambient temperature information upon which the compensation function is based already includes the correction for the local temperature rise deviation caused by exhaust recirculation. When recirculation occurs, the corrected equivalent ambient temperature is higher than the outdoor air temperature, and the current change output by the compensation function tends to decrease, resulting in a lower target current for the water pump. This aligns with the physical reality that recirculation reduces the heat dissipation efficiency and actual cooling capacity of the air conditioner cooler. Under recirculation conditions, the air conditioner cooler is essentially operating in a higher temperature environment, and the water pump control strategy should be adjusted accordingly.

[0120] Step S32: When a backflow self-excited oscillation is detected, a low-pass filter is started with a cutoff period of a preset multiple of the backflow self-excited oscillation period to attenuate the residual fluctuations; the filter is turned off when the outdoor wind speed is higher than the critical backflow wind speed threshold.

[0121] Although step S23 eliminates the periodic fluctuations of the backflow oscillation from the corrected equivalent ambient temperature by using a sliding window time averaging method, making the initial target operating current of the water pump output in step S31 stable, in actual systems, feedback correction loop and other sensor noise may still introduce residual short-period fluctuations into the initial target operating current of the water pump. The task of step S32 is to further attenuate and suppress these residual fluctuations. When backflow self-excited oscillation is detected in step S22, that is, the system is currently in a high-incidence backflow condition and backflow oscillation has been confirmed, a low-pass filter is activated. The cutoff period of the low-pass filter is set to a preset multiple of the backflow self-excited oscillation period, which is usually 1.5 to 3 times. The physical meaning of the cutoff period is: signal fluctuation components with a period shorter than the cutoff period will be attenuated and suppressed by the filter, while signal gradual changes with a period longer than the cutoff period will be retained. In a specific embodiment, if the backflow self-excited oscillation period is 8 minutes and the preset multiple is 2, then the cutoff period is 16 minutes. This means that short-period fluctuations within 16 minutes will be smoothed by the filter, while slow trends exceeding 16 minutes will be preserved. The low-pass filter can be implemented using a first-order exponentially weighted moving average filter. The calculation method for a first-order exponentially weighted moving average filter is: the current filtered output value equals the previous filtered output value multiplied by the smoothing coefficient, plus the current original input value multiplied by one minus the smoothing coefficient. The magnitude of the smoothing coefficient is determined by both the cutoff period and the sampling period; the longer the cutoff period, the closer the smoothing coefficient is to 1, and the stronger the filtering effect. Specifically, the smoothing coefficient... The calculation formula is ,in The sampling period of the control system, Let be the time constant of the filter, and let the time constant be . equals the deadline divided by .

[0122] When step S21 determines that the real-time outdoor wind speed is higher than the critical return wind speed threshold, indicating that the system is not currently in a high-risk return wind condition, the low-pass filter is turned off, and the initial target operating current of the water pump is directly used as the filtered output value. The reason for turning off the filter is that under windy conditions, there is no exhaust return wind, the intake air temperature does not oscillate due to return wind, the corrected equivalent ambient temperature is directly equal to the outdoor air temperature, and the initial target operating current of the water pump is itself stable, requiring no additional filtering. Keeping the filter always on would introduce unnecessary control response delays under normal operating conditions, affecting the system's ability to follow normal changes in ambient temperature. Step S32, as the final smoothing process before the control signal output, ensures that the current command signal finally delivered to the frequency converter driver is time-stable and does not contain short-period fluctuation components that could cause frequent sudden changes in water pump speed. Sudden changes in water pump motor speed can cause water hammer and flow shocks in the cooling water pipes, causing mechanical damage to the pipes and equipment, while also increasing the electrical stress and energy consumption of the motor. A smooth and stable current command enables the water pump to operate smoothly, which is beneficial for extending equipment life and reducing operating energy consumption.

[0123] Step S33: The filtered target operating current setting value of the water pump is output to the frequency converter driver. The frequency converter driver adjusts the output frequency according to the current setting value so that the water pump operating current tracks the setting value. At the same time, the deviation between the measured value of the water supply temperature and the target value is used as a feedback correction amount and added to the current setting value to realize feedforward-feedback composite constant temperature control.

[0124] The target operating current setpoint for the water pump, after smoothing and filtering in step S32, is output to the variable frequency drive (VFD) of the circulating water pump. The VFD uses this current setpoint as a tracking target and automatically adjusts its output frequency to make the actual operating current of the water pump approach and stabilize at the setpoint, thereby changing the speed of the water pump motor and ultimately changing the circulation flow rate of the cooling water. Under a given system load condition, there is an approximately proportional relationship between the water pump operating current and the VFD output frequency, as well as an approximately proportional relationship between the water pump speed and the cooling water circulation flow rate. Therefore, sending the current setpoint to the VFD and having it adjust the frequency for tracking is equivalent to controlling the cooling water circulation flow rate. Based on feedforward control, step S33 also introduces a feedback correction loop. The feedback correction is implemented by installing a cooling water supply temperature sensor on the air conditioner cooler outlet pipe to continuously collect the measured value of the cooling water supply temperature. The measured supply temperature is compared with the constant temperature control target supply temperature set in step S31, and the deviation between the two is calculated. The deviation value is multiplied by a preset feedback gain coefficient to obtain the feedback correction amount. This feedback correction amount is then added to the filtered target operating current setpoint of the water pump to obtain the final current setpoint, which is output to the frequency converter for frequency adjustment and tracking. The direction of the feedback correction is as follows: when the measured water supply temperature is higher than the target water supply temperature, the deviation is positive, and the feedback correction increases the water pump current, increasing the circulation flow to enhance the cooling effect and causing the water supply temperature to decrease towards the target value; when the measured water supply temperature is lower than the target water supply temperature, the deviation is negative, and the feedback correction decreases the water pump current, reducing the circulation flow to weaken the cooling effect and causing the water supply temperature to rise towards the target value. The value of the feedback gain coefficient needs to comprehensively consider the response speed and stability of the control system. An excessively large feedback gain coefficient will cause the control system to respond too sensitively, easily leading to oscillations in the water supply temperature; an excessively small feedback gain coefficient will result in a slow response, and the deviation in the water supply temperature cannot be corrected in a timely manner. The feedback gain coefficient can be gradually adjusted and determined during the trial operation phase of the system commissioning. In one specific embodiment, the feedback gain coefficient is set to 0.8 amperes per degree Celsius, meaning that for every 1 degree Celsius increase in the supply water temperature, the pump current increases by 0.8 amperes. The physical significance of the feedforward-feedback composite thermostatic control lies in the following: the feedforward stage predicts changes in the cooling environment of the air conditioner cooler by correcting the equivalent ambient temperature, and adjusts the pump operating current in advance to achieve a proactive and rapid response to changes in ambient temperature; the feedback stage performs closed-loop correction by correcting the deviation between the measured supply water temperature and the target value, compensating for the control residual caused by model simplification or parameter deviations in the feedforward stage, ensuring that the supply water temperature remains stable near the target value over a long period. Working together, the feedforward stage provides rapid response capability, while the feedback stage provides steady-state accuracy assurance, jointly achieving high-precision thermostatic control of the cooling water supply temperature.

[0125] The control strategy for circulating water pumps typically employs only feedback control, adjusting the pump's frequency converter output based on the temperature difference between the supply and return water, or the deviation in supply water temperature. Pure feedback control inherently suffers from response lag: a temperature deviation must first occur and be detected by the sensor before the controller can take adjustment action. The effect of this adjustment action then requires the cooling water to circulate through the pipeline once before it is reflected in the supply water temperature sensor reading. The entire feedback chain delay can reach several minutes. Under conditions of rapid ambient temperature changes or exhaust recirculation, pure feedback control struggles to keep up with environmental disturbances, leading to significant fluctuations in the supply water temperature. This step utilizes a feedforward-feedback composite control structure. By introducing a corrected equivalent ambient temperature as a feedforward variable, it performs pre-adjustment actions before a temperature deviation occurs, effectively shortening the equivalent response time of the control system and improving its ability to suppress environmental disturbances.

[0126] Example 2.

[0127] like Figure 2 As shown, the adaptive constant temperature control system for circulating water pump based on ambient temperature includes a potential energy modeling module, which acquires the spatial structure parameters of the air conditioner cooler, constructs a three-dimensional thermal and humid air motion model of exhaust-intake airflow coupling, and generates a risk potential energy field for exhaust thermal and humid air mass recirculation based on the three-dimensional thermal and humid air motion model.

[0128] The backflow detection module, based on the risk potential energy field of the exhaust hot and humid air mass backflow, combined with the real-time ambient temperature, the weather forecast temperature and the meteorological boundary conditions, performs backflow self-excited oscillation detection and outputs the corrected equivalent ambient temperature to eliminate backflow self-interference deviation.

[0129] The constant temperature control module uses the corrected equivalent ambient temperature as a feedforward variable and generates a target operating current setpoint for the water pump through a compensation function that compensates for the inverse change between the water pump operating current and the ambient temperature. After smoothing and filtering, the setpoint is output to the variable frequency drive of the circulating water pump. The variable frequency drive adjusts the output frequency according to the current setpoint so that the water pump operating current tracks the setpoint and maintains a constant cooling water supply temperature.

[0130] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0131] The foregoing description is illustrative of the invention and should not be construed as limiting it. Although several exemplary embodiments of the invention have been described, those skilled in the art will readily understand that many modifications can be made to the exemplary embodiments without departing from the novel teachings and advantages of the invention. Therefore, all such modifications are intended to be included within the scope of the invention as defined in the claims. It should be understood that the foregoing description is illustrative of the invention and should not be construed as limiting it to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The invention is defined by the claims and their equivalents.

Claims

1. An adaptive constant temperature control method for circulating water pumps based on ambient temperature, characterized in that, The method includes the following steps: Obtain the spatial structural parameters of the air conditioner cooler, construct a three-dimensional thermal and humid air motion model of exhaust-intake airflow coupling, and generate the risk potential energy field of exhaust thermal and humid air mass recirculation based on the three-dimensional thermal and humid air motion model. Based on the risk potential energy field of the exhaust hot and humid air mass recirculation, combined with the real-time ambient temperature, the weather forecast temperature and the meteorological boundary conditions, the recirculation self-excited oscillation detection is performed, and the corrected equivalent ambient temperature is output to eliminate the recirculation self-interference deviation. Using the corrected equivalent ambient temperature as a feedforward variable, a target operating current setting value for the water pump is generated through a compensation function that compensates for the inverse change between the water pump operating current and the ambient temperature. After smoothing and filtering, the setting value is output to the variable frequency drive of the circulating water pump. The variable frequency drive adjusts the output frequency according to the current setting value so that the water pump operating current tracks the setting value, thereby maintaining a constant cooling water supply temperature.

2. The adaptive constant temperature control method for circulating water pumps based on ambient temperature according to claim 1, characterized in that, The steps for constructing the three-dimensional thermal and humid air motion model are as follows: Obtain the vertical height of the air conditioner cooler's exhaust vent from the roof, the exhaust vent's cross-sectional diameter, the height of the bottom edge of the air inlet louvers, and the range of their circumferential curvature; with the geometric center of the exhaust vent as the origin, vertically upwards... A three-dimensional cylindrical coordinate system is established, with the exhaust vent defined as the emission source surface of hot and humid air masses, and the circumferential space area covered by the air inlet louvers defined as the air mass capture and confluence surface.

3. The adaptive constant temperature control method for circulating water pumps based on ambient temperature according to claim 2, characterized in that, The steps for constructing the three-dimensional thermal and humid air motion model also include: The hot and humid air discharged from the exhaust vent is discretized into a sequence of micro-elemental air masses. For each micro-elemental air mass, the buoyancy-driven component is calculated based on its temperature difference with the ambient air. The temperature decay path is corrected based on the release of latent heat of condensation of moisture content. The radial diffusion rate is calculated based on the turbulent diffusion coefficient. The spatial motion trajectory of each micro-elemental air mass in the three-dimensional cylindrical coordinate system is obtained by combining the buoyancy-driven component, the temperature decay path, and the radial diffusion rate. The combination of the buoyancy-driven component, the temperature decay path, and the radial diffusion rate constitutes the three-dimensional hot and humid air motion model.

4. The adaptive constant temperature control method for circulating water pumps based on ambient temperature according to claim 3, characterized in that, The steps for correcting the temperature decay path include: Within each calculation time step, it is determined whether the current moisture content of the volumetric micro-element air mass exceeds the saturation moisture content corresponding to the current temperature. If it does, the excess water vapor mass is multiplied by the latent heat of vaporization to obtain the heat released by condensation, which is converted into the temperature rise increment of the air mass and added to the sensible heat cooling amount. At the same time, the mass of condensed water vapor is deducted from the moisture content of the air mass. When the moisture content drops below the saturation value, condensation terminates, and the air mass cools down rapidly and enters the rapid sinking stage.

5. The adaptive constant temperature control method for circulating water pumps based on ambient temperature according to claim 1, characterized in that, The steps for constructing the risk potential energy field of the exhaust hot and humid air mass recirculation are as follows: Based on the spatial motion trajectory of all volumetric micro-element air masses output by the three-dimensional thermal and humid air motion model, a geometric intersection determination with the air mass capture confluence surface is performed. Air masses that cross the air mass capture confluence surface are marked as return air masses and their remaining temperature during the crossing is recorded. The spatial distribution of return heat flux density per unit time is statistically analyzed to generate the risk potential energy field of the exhaust thermal and humid air mass return flow.

6. The adaptive constant temperature control method for circulating water pumps based on ambient temperature according to claim 5, characterized in that, The calculation steps for the reflux heat flux density are as follows: The air mass capture surface is discretized into a rectangular array of elements along the circumferential and vertical directions; for each recirculating air mass, the assigned element is determined according to the crossing position, and the difference between its remaining temperature and the ambient temperature is multiplied by the volumetric flow rate and the specific heat capacity of air to obtain the single-unit recirculating heat power. The average hourly recirculation heat flux density is obtained by summing the individual recirculation heat power of all recirculation air masses belonging to each surface element within the statistical period and dividing by the surface element area and the statistical duration.

7. The adaptive constant temperature control method for circulating water pumps based on ambient temperature according to claim 1, characterized in that, The steps for outputting the corrected equivalent ambient temperature are as follows: Real-time outdoor wind speed is obtained and weather forecast temperature is obtained through meteorological service interface. The real-time outdoor dry-bulb temperature and the weather forecast temperature are weighted and fused to obtain a comprehensive ambient temperature reference value. When the real-time outdoor wind speed is lower than the critical return wind speed threshold, it is determined to be a high-incidence condition of return flow. The critical recirculation wind speed threshold is determined by the height difference between the exhaust port and the air inlet louver and the cross-sectional diameter of the exhaust port. It is the minimum wind speed at which the farthest recirculation air mass is horizontally blown away from the capture radius before sinking to the air mass capture confluence. Under conditions of high recirculation, the measured temperature sequence of the air inlet is continuously collected from the air inlet temperature sensor at the air inlet louvers of the air conditioner cooler. Short-time sliding window spectrum analysis is performed to extract the periodic fluctuation component within the preset oscillation period range. The peak-valley temperature difference is defined as the recirculation self-excited oscillation amplitude and the center period is defined as the recirculation self-excited oscillation period. The recirculation heat flux density at the corresponding position in the risk potential energy field of the exhaust hot and humid air mass recirculation is compared and verified. After confirming the existence of backflow self-excited oscillation through comparison and verification, the average time of a sliding window with a width that is an integer multiple of the backflow self-excited oscillation period is taken from the measured temperature sequence of the air intake. The difference between this average value and the comprehensive ambient temperature reference value is defined as the backflow average temperature rise offset. The corrected equivalent ambient temperature is output by superimposing the comprehensive ambient temperature reference value with the backflow average temperature rise offset.

8. The adaptive constant temperature control method for circulating water pumps based on ambient temperature according to claim 7, characterized in that, The reflux mean temperature rise offset also includes peak value retention correction: The maximum measured temperature of the incoming air is extracted within the sliding window as the peak temperature of the window. The difference between the peak temperature of the window and the mean temperature of the window is defined as the peak-to-average deviation. After multiplying by the safety margin coefficient, it is superimposed on the mean temperature of the window to obtain the corrected mean. The corrected mean is used to replace the original mean in the calculation of the return flow mean temperature rise offset.

9. The adaptive constant temperature control method for circulating water pumps based on ambient temperature according to claim 1, characterized in that, The steps for generating and outputting the target operating current setpoint for the water pump are as follows: Set the target water supply temperature, reference ambient temperature, and corresponding reference operating current for constant temperature control. Use the difference between the equivalent ambient temperature and the reference ambient temperature as input, and output the current change through the reverse change compensation function. Add the current change to the reference operating current to obtain the initial target operating current of the water pump. When backflow self-excited oscillation is detected, a low-pass filter is activated with a cutoff period of a preset multiple of the backflow self-excited oscillation period to attenuate residual fluctuations. The filter is turned off when the outdoor wind speed is higher than the critical return wind speed threshold. The filtered target operating current setting value of the water pump is output to the frequency converter driver. The frequency converter driver adjusts the output frequency according to the current setting value so that the water pump operating current tracks the setting value. At the same time, the deviation between the measured value of the water supply temperature and the target value is used as a feedback correction and added to the current setting value to realize feedforward-feedback composite constant temperature control.

10. An adaptive constant temperature control system for a circulating water pump based on ambient temperature, used to implement the adaptive constant temperature control method for a circulating water pump based on ambient temperature as described in any one of claims 1-9, characterized in that, include: The potential energy modeling module obtains the spatial structural parameters of the air conditioner cooler, constructs a three-dimensional thermal and humid air motion model coupled with exhaust and intake airflow, and generates a risk potential energy field for the return flow of exhaust thermal and humid air masses based on the three-dimensional thermal and humid air motion model. The backflow detection module, based on the risk potential energy field of the exhaust hot and humid air mass backflow, combined with the real-time ambient temperature, the weather forecast temperature and the meteorological boundary conditions, performs backflow self-excited oscillation detection and outputs the corrected equivalent ambient temperature to eliminate backflow self-interference deviation. The constant temperature control module uses the corrected equivalent ambient temperature as a feedforward variable and generates a target operating current setpoint for the water pump through a compensation function that compensates for the inverse change between the water pump operating current and the ambient temperature. After smoothing and filtering, the setpoint is output to the variable frequency drive of the circulating water pump. The variable frequency drive adjusts the output frequency according to the current setpoint so that the water pump operating current tracks the setpoint and maintains a constant cooling water supply temperature.

Images