A control method, device and storage medium for adaptive adjustment of a magnetic levitation fluorine pump air conditioner

Abstract

The application discloses a kind of magnetic levitation fluorine pump air conditioner self-adapting regulation control method, equipment and storage medium, it is related to air conditioner control technical field, to solve the problem that existing VRF system load perception lags behind, ignores pipe network thermodynamics loss, the method comprises: real-time acquisition includes distributed heterogeneous load data, calculate the comprehensive load index of the system total demand;The index is matched to the preset performance mapping model, to determine an optimal operation baseline vector consisting of pump speed, valve opening degree and other multi-dimensional target parameters;Further, construct the current operating state feature vector, and apply Mahalanobis distance algorithm, calculate the operating state deviation between it and the optimal operation baseline vector;Finally, based on the deviation degree generates the coordinated regulation instruction of magnetic levitation fluorine pump and electronic expansion valve.The application realizes the leap from single-point temperature difference control to multi-dimensional optimal state coordination by the depth of load deconstruction and overall evaluation of system state.

Description

Technical Field

[0001] This invention relates to the field of air conditioning control technology, specifically to a control method, device, and storage medium for adaptive adjustment of a magnetic levitation fluorine pump air conditioner. Background Technology

[0002] As a core driving component of next-generation air conditioning systems, magnetic levitation fluorine pumps utilize electromagnetic force to achieve contactless levitation and high-speed rotation of the pump rotor. Compared to traditional mechanical bearing pumps, they offer significant advantages such as oil-free operation, low noise and vibration, and wide-range speed regulation. Theoretically, their broad speed range (e.g., from 500 RPM to 60,000 RPM) provides a physical basis for precisely matching dynamically changing cooling demands.

[0003] With the development of modern buildings towards high-rise and integrated structures, multi-split variable refrigerant flow (VRF) air conditioning systems have been widely used in large commercial buildings and office complexes due to their energy-saving, comfort, and flexible installation advantages. The core of this type of system lies in connecting multiple indoor units to a single outdoor unit and independently adjusting the amount of refrigerant delivered to each temperature-controlled area based on the actual needs of that area, thus achieving personalized temperature control for each zone. VRF air conditioning systems have long refrigerant piping with many branches, and the load differences between different parts of the system are significant. For example, in a commercial building, some temperature-controlled areas facing the sun require strong cooling, while others in shaded areas require weak cooling. In such scenarios, existing multi-split air conditioning control methods typically have the following shortcomings: Firstly, at the load sensing level, traditional control methods primarily rely on monitoring the return air temperature in each terminal temperature control zone. This temperature difference is compared to the user-set temperature, forming a single temperature difference signal as the basis for control. This approach is essentially a lagging, one-dimensional feedback adjustment. It cannot finely analyze the causes of the load; for example, it cannot distinguish between dynamic loads caused by a sudden increase in indoor occupants and static loads caused by increased solar radiation. This crude load sensing method leads to untimely system response, easily resulting in overshooting or undershooting of temperature control, affecting user comfort and causing unnecessary energy waste.

[0004] Secondly, for complex systems with long pipelines and numerous branches, existing control methods often neglect the thermodynamic losses of refrigerant during long-distance transport. These losses mainly include heat exchange losses between the refrigerant and the surrounding environment, as well as pressure losses caused by flow friction and local resistance (such as bends and branches). Traditional control logic typically treats all terminal areas as equal load demanders, failing to quantify the additional energy cost incurred to meet the needs of specific distant areas. This neglect leads to a lack of basis for energy allocation, potentially causing the entire system to consume disproportionate amounts of power to meet the needs of a small load area with significant pipeline losses, thus significantly reducing the system's overall energy efficiency ratio. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a control method, device, and storage medium for adaptive adjustment of a magnetic levitation fluorine pump air conditioner, thereby solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a control method for adaptive adjustment of a magnetic levitation refrigerant pump air conditioner, comprising the following steps: S1: Real-time acquisition of multi-source operational data characterizing the system's operating status. The multi-source operational data includes distributed heterogeneous load data acquired in multiple terminal temperature control zones defined by the multiple indoor units, as well as pump chamber internal status data. The distributed heterogeneous load data includes dynamic personnel heat load characterizing personnel activities, pipe network thermodynamic load characterizing pipeline pressure loss and heat exchange effect, and static terminal temperature control zone heat load characterizing heat transfer in the building envelope and heat dissipation in the equipment. S2: Based on the internal state data of the pump chamber, calculate the instantaneous operating efficiency index that quantifies the current two-phase flow state of the refrigerant and the operating efficiency; S3: Based on the distributed heterogeneous load data, calculate the dynamic heat source load index, corresponding pipeline heat loss pressure index, building envelope heat transfer index, and equipment heat dissipation index for each terminal temperature control zone, and obtain the comprehensive load index by correlation. S4: Based on the comprehensive load index, match the optimal operating baseline of the system. The optimal operating baseline is the total cooling capacity required by the comprehensive load index established in advance through physical model simulation or machine learning training on historical high-efficiency operating data. Obtain the current total cooling capacity, apply the Mahalanobis distance algorithm to calculate the distance between the current total cooling capacity and the total cooling capacity required by the optimal operating baseline, and use it as the operating deviation. Output the corresponding adjustment command.

[0007] Preferably, after collecting the internal state data of the pump cavity, the process further includes: estimating key two-phase flow parameters characterizing the vapor-liquid distribution state within the pump cavity in real time based on the pump current and magnetic signal inside the pump cavity, wherein the key two-phase flow parameters include the vapor-liquid ratio and the vapor-liquid distribution index of the pump cavity; and performing nonlinear fusion of the key two-phase flow parameters and the pump cavity vibration signal to obtain a pump cavity stability index that comprehensively characterizes the fluid stability inside the pump cavity.

[0008] Preferably, S2 specifically includes: Based on the pump current and magnetic signal in the internal state data of the pump chamber, the key parameters of the two-phase flow characterizing the vapor-liquid distribution state in the pump chamber are estimated in real time. The pump cavity vibration signal in the key parameters of the two-phase flow and the internal state data of the pump cavity is nonlinearly fused to obtain a pump cavity stability index that comprehensively characterizes the internal fluid stability of the pump cavity. The instantaneous operating efficiency index is obtained by weighting and fusing the thermodynamic cycle efficiency coefficient, which characterizes the performance of the refrigeration cycle, the energy consumption ratio, which characterizes the unit power consumption, and the pump cavity stability index.

[0009] Preferably, the dynamic heat load index of each terminal temperature control area is used to characterize the dynamic heat load generated by comprehensively evaluating the number of people and the intensity of their activities within the terminal temperature control area. Specifically, the difference between the real-time collected carbon dioxide concentration value in the terminal temperature control area and the preset environmental reference concentration value is used to obtain the personnel density, and the frequency or amplitude of motion events acquired by the infrared sensor or image analysis module per unit time is used to obtain the personnel activity level. The personnel density and personnel activity level are multiplied by their respective preset weights, and the sum obtained is the dynamic heat load index of the terminal temperature control area.

[0010] Preferably, the heat loss pressure index of the pipeline corresponding to each terminal temperature control zone is used to characterize a composite index formed by the linear superposition of heat exchange loss component and flow friction loss component, in order to comprehensively quantify the thermodynamic loss of refrigerant during transportation. Specifically, it involves obtaining the physical length of the pipeline connected to the terminal temperature control zone, the real-time temperature of the refrigerant inside the pipeline, the average temperature of the environment where the pipeline is located, and the refrigerant mass flow rate estimated based on the system operating status; multiplying the physical length of the pipeline by the absolute value of the temperature difference between the inside and outside of the pipeline and the square value of the refrigerant mass flow rate, and then multiplying the two products by their respective preset heat exchange loss coefficient and flow friction loss coefficient, and summing them up. The final sum is the heat loss pressure index of the pipeline corresponding to the terminal temperature control zone.

[0011] Preferably, the heat transfer index of the building envelope of each terminal temperature control area is used to characterize the stable heat load transmitted through the building shell, determined by quantifying the building's physical characteristics and external meteorological conditions. Specifically, this involves: obtaining the real-time indoor-outdoor temperature difference and the real-time solar radiation intensity measured by a light sensor; calculating the product of the indoor-outdoor temperature difference and a preset comprehensive heat transfer coefficient representing the overall heat transfer capacity of the walls and roof of the terminal temperature control area with the area to obtain the conductive heat component; calculating the real-time solar radiation intensity and the preset total window area and solar heat gain coefficient of the terminal temperature control area to obtain the solar radiation heat component; and weighting and summing the conductive heat component and the solar radiation heat component to obtain the final weighted sum value, which is the heat transfer index of the building envelope of the terminal temperature control area.

[0012] Preferably, the calculation of the device heat dissipation index is a process of summing up the current heat dissipation power of all registered electronic devices and lighting devices that generate heat within the area. For each individual device in the list, its instantaneous heat dissipation power is determined by multiplying its preset rated power, a heat dissipation coefficient characterizing its electrothermal conversion efficiency, and a real-time operating status factor determined by smart socket readings or preset work schedules. Finally, the instantaneous heat dissipation power calculated for all devices is summed up, and the sum is the device heat dissipation index of the terminal temperature control area.

[0013] Preferably, the comprehensive load index is obtained by linearly superimposing and summing the three indices representing the actual heat load source of the area—the dynamic heat source load index, the building envelope heat transfer index, and the equipment heat dissipation index—with the pipeline heat loss pressure index, which represents energy transmission loss, by multiplying each index by its respective preset weighting coefficients used to normalize the dimensions and adjust their influence priority. Based on the comprehensive load index, the optimal operating baseline of the matching system is established in advance through physical model simulation or machine learning training on historical high-efficiency operating data to determine the total cooling capacity required for the comprehensive load index. The current total cooling capacity is obtained, and the Mahalanobis distance algorithm is applied to calculate the distance between the current total cooling capacity and the total cooling capacity required for the optimal operating baseline, which is used as the operating deviation. Corresponding adjustment commands are output, including: calculating the difference between the current speed and the target speed of the magnetic levitation fluorine pump, and the difference between the current opening degree and the target opening degree of the electronic expansion valve, which are executed as adjustment commands.

[0014] Preferably, an apparatus includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the apparatus executes the computer program to implement the steps of a control method for adaptive adjustment of a magnetic levitation fluorine pump air conditioner.

[0015] Preferably, a storage medium is used to perform the steps of a control method for adaptive adjustment of a magnetic levitation fluorine pump air conditioner.

[0016] This invention provides a control method, device, and storage medium for adaptive adjustment of a magnetic levitation refrigerant pump air conditioner. It offers the following advantages: (1) At the load perception level, this method achieves a shift from "passive response" to "active prediction" and enables a deep deconstruction of the causes of load. By refining the total load into three dimensions—dynamic personnel load, pipeline thermodynamic load, and static regional load—this method can gain insight into the intrinsic composition of the load, rather than simply perceiving a general temperature change. In particular, by quantifying the pipeline thermodynamic load, the controller is able to "see" and assess for the first time the energy cost incurred to meet the needs of terminals at different distances and with different pipeline complexities, providing an unprecedented basis for subsequent energy optimization allocation.

[0017] (2) At the control strategy level, an advanced control paradigm based on multidimensional state vectors and Mahalanobis distance is introduced, achieving a leap from "single-point control" to "system-level optimal coordination". The core innovation of this method lies in the fact that it does not pursue the achievement of a single parameter (such as return air temperature), but rather the "statistical alignment" of the entire system's operating state vector with the theoretically optimal baseline vector in multidimensional space. The application of Mahalanobis distance can comprehensively consider the dimensional differences of various operating parameters (such as pump speed and valve opening) and their inherent physical correlations, and calculate a deviation index that can truly reflect the overall "health" of the system. This makes the adjustment command no longer an isolated adjustment of the pump or valve, but generates a set of coordinated actions to drive the entire system to meet the load demand with the highest efficiency, fundamentally avoiding the problem of local optima and global inefficiency, and significantly improving the overall energy efficiency ratio and control accuracy of the VRF system under complex and dynamic loads. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of three different end temperature control zones of the present invention; Figure 2 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0019] The technical solutions of 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 within the scope of protection of the present invention.

[0020] Example 1 Please see Figure 1-2This invention provides a control method, device, and storage medium for adaptive adjustment of a magnetic levitation refrigerant pump air conditioner. The method is applied to a VRF air conditioning system and includes: an outdoor unit containing a magnetic levitation refrigerant pump, multiple indoor units connected to the outdoor unit via refrigerant piping, and multiple electronic expansion valves respectively disposed at the refrigerant inlets of the multiple indoor units; the method includes the following steps: S1: Real-time acquisition of multi-source operational data characterizing the system's operating status. The multi-source operational data includes: distributed heterogeneous load data acquired in multiple terminal temperature control zones defined by the multiple indoor units, and pump chamber internal status data acquired in the magnetic levitation fluorine pump chamber. The distributed heterogeneous load data includes: dynamic personnel heat load characterizing personnel activities, pipe network thermodynamic load characterizing pipeline pressure loss and heat exchange effect, and static terminal temperature control zone heat load characterizing heat transfer of the building envelope and heat dissipation of equipment. S2: Based on the internal state data of the pump chamber, calculate the instantaneous operating efficiency index that quantifies the current two-phase flow state of the refrigerant and the operating efficiency; S3: Based on the distributed heterogeneous load data, calculate the dynamic heat source load index, corresponding pipeline heat loss pressure index, building envelope heat transfer index, and equipment heat dissipation index for each terminal temperature control zone, and obtain the comprehensive load index in relation to them; determine the optimal operating baseline of the system that matches the total cooling demand of all terminal temperature control zones based on the comprehensive load index; and calculate the operating state deviation degree, which characterizes the degree of deviation between the current operating state and the optimal operating baseline, in conjunction with the instantaneous operating efficiency index. S4: Based on the operating state deviation, generate an adaptive control strategy to correct the deviation. The adaptive control strategy includes a global operating command for the magnetic levitation fluorine pump and multiple independent adjustment commands for the multiple electronic expansion valves.

[0021] After collecting the internal state data of the pump chamber, the process further includes: estimating key two-phase flow parameters characterizing the vapor-liquid distribution state within the pump chamber in real time based on the pump current and magnetic signals inside the pump chamber, wherein the key two-phase flow parameters include the vapor-liquid ratio and the vapor-liquid distribution index of the pump chamber; and performing nonlinear fusion of the key two-phase flow parameters and the pump chamber vibration signal to obtain a pump chamber stability index that comprehensively characterizes the fluid stability inside the pump chamber.

[0022] Based on the pump current and magnetic signal in the internal state data of the pump chamber, the key parameters of the two-phase flow characterizing the vapor-liquid distribution state in the pump chamber are estimated in real time. The pump cavity vibration signal in the key parameters of the two-phase flow and the internal state data of the pump cavity is nonlinearly fused to obtain a pump cavity stability index that comprehensively characterizes the internal fluid stability of the pump cavity. The instantaneous operating efficiency index is obtained by weighting and fusing the thermodynamic cycle efficiency coefficient, which characterizes the performance of the refrigeration cycle, the energy consumption ratio, which characterizes the unit power consumption, and the pump cavity stability index.

[0023] The calculation of the thermodynamic cycle efficiency coefficient includes comparing and analyzing the real-time collected evaporation pressure and condensation pressure data with a preset reference refrigerant thermodynamic property database; taking the thermodynamic cycle efficiency coefficient, energy consumption ratio, and pump cavity stability index as multi-dimensional inputs, and performing nonlinear fusion through a pre-trained neural network model or fuzzy inference system to output the instantaneous operating efficiency index.

[0024] Nonlinear fusion is performed using a pre-trained neural network model, specifically: a feedforward multilayer perceptron (MLP) is used as the neural network model, which is pre-trained through supervised learning. The training dataset contains input vectors (thermodynamic cycle efficiency coefficient, EER, pump cavity stability index) collected from a large number of historical operating conditions, as well as "true" efficiency levels (as labels) calibrated by domain experts or obtained through offline comprehensive evaluation. The training objective is to minimize the mean squared error between the model's predicted output and the true label. Through this process, the neural network learns the complex, nonlinear intrinsic relationship between the three input parameters, especially learning to apply a much more linear penalty weight to the final output when the pump cavity stability index is low; the core output is the instantaneous operating efficiency index normalized by the sigmoid activation function, with a value range of [0,1]. When the instantaneous operating efficiency index (ηinst) approaches 1, it indicates that the system is in a highly efficient and stable "healthy" operating state. This not only means that the current thermodynamic cycle is close to ideal (… th High efficiency (high th), high conversion rate of electrical energy to cold energy (high EER), and more importantly, stable fluid state inside the core power component (magnetic levitation-floating fluorine pump) (high Spump), indicating that this high efficiency is sustainable and safe.

[0025] When the instantaneous operating efficiency index (ηinst) approaches 0, it indicates that the system is in an inefficient or dangerous "sub-healthy" or even "failed" state. This may be caused by one or more factors: for example, even if the EER is still acceptable, a sharp drop in Spum indicates an impending risk of liquid slugging or a sudden drop in efficiency, and the model will significantly lower ηinst in advance to provide a predictive risk warning.

[0026] Thermodynamic cycle efficiency coefficient ( th The energy efficiency ratio (EER) and the instantaneous operating efficiency index (ηinst) of the final output are both strongly positively correlated; the thermodynamic cycle efficiency coefficient (tc) is also strongly positively correlated. th Energy efficiency ratio (EER) and energy consumption ratio (HER) are recognized core indicators in the refrigeration field, characterizing system efficiency. The former reflects the theoretical performance limit, while the latter reflects actual performance. Any highly efficient system will exhibit excellent performance in both indicators. Therefore, through training, the neural network model will inevitably learn the positive mapping relationship between these two inputs and outputs, which fully conforms to the basic physical laws of refrigeration technology. The pump cavity stability index (Spump) and instantaneous operating efficiency index (ηinst) show a strong nonlinear positive correlation. Its peculiarity lies in the fact that when Spump decreases from a high level (e.g., 0.9) to a mid-to-low level (e.g., 0.6), the negative impact weight on ηinst increases dramatically, far exceeding its linear proportion; this is the core difference between this invention and traditional efficiency assessments. Spump quantifies the "health" of system operation. A healthy system is a prerequisite for efficient operation. Even if the current EER value is acceptable, an unstable pump (e.g., internal two-phase flow turbulence, increased vibration) cannot maintain efficient operation for a long time and is at risk of damage. Neural networks, by learning from training data that includes the process of "from stability to instability," can deeply understand this "cliff effect": once stability falls below a certain threshold, no matter how good other indicators are, the overall system's "true efficiency" (including sustainability and safety) has deteriorated sharply. This nonlinear design accurately incorporates the forward-looking and predictive dimension of mechanical health into the comprehensive judgment of current efficiency; The specific experimental data is shown in Table 1 below: Table 1. Experimental data for Example 1: This experiment aims to verify that the instantaneous operating efficiency index (ηinst) proposed in this invention has significant advantages over traditional efficiency evaluation methods that rely solely on the system energy efficiency ratio (EER), especially in its ability to diagnose and predict potential system risks. The experiment designed three operating conditions: Scenario A represents ideal, efficient, and stable operation; Scenario B simulates the "sub-healthy" state at the initial stage of liquid slugging risk, where the traditional index EER has not yet significantly deteriorated; Scenario C represents the inefficient and unstable state after liquid slugging has occurred. Comparison between Scenario A and Scenario B: "Early warning" capability for hidden risks; Data Comparison: From Scenario A (high efficiency and stability) to Scenario B (latent instability), the internal state of the system deteriorates significantly: the pump cavity vibration amplitude surges from 0.5 mm / s² to 3.5 mm / s², causing the pump cavity stability index to plummet from 0.98 to 0.50. However, at this stage, the system's energy efficiency ratio (EER), the core of traditional efficiency, only slightly decreases from 4.2 to 4.1, a reduction of only 2.4% (calculation process: subtract 4.1 from the baseline value of 4.2 to get an improvement of 0.1, then divide 0.1 by the baseline value of 4.2, and multiply the result by 100%, approximately 2.4%). Traditional methods, based on this minute change, are almost unable to determine that the system has entered a dangerous "sub-healthy" state.

[0027] Demonstration of Effectiveness: In stark contrast, the instantaneous operating efficiency index (ηinst) of this invention, due to the incorporation of the rapidly deteriorating stability index, plummeted from 0.95 to 0.45, a decrease of 52.6% (calculation process: subtract 0.45 from the baseline value of 0.95 to obtain an improvement of 0.5, then divide 0.5 by the baseline value of 0.95, and multiply the result by 100%, approximately 52.6%). This data irrefutably proves that the method of this invention can detect internal system risks that traditional indicators cannot perceive in advance, interpreting a seemingly insignificant change in EER as a significant, high-risk efficiency deterioration event, thus providing a valuable early warning window for the control system to take preventative measures (such as adjusting the opening of the electronic expansion valve and reducing the pump speed).

[0028] Comparison of Scenario B and Scenario C: Precise quantification of fault status; Data Comparison: From scenario B to scenario C (liquid slugging occurs), the system's EER finally "reacts," plummeting from 4.1 to 2.5, confirming that the system was indeed on the verge of failure in scenario B. The ηinst of this invention also further decreased from 0.45 to 0.22, consistent with the trend of EER changes.

[0029] Demonstration of Effectiveness: This set of comparisons demonstrates the full-range effectiveness of the present invention's indicators. It not only provides early warnings in the early stages of a fault but also accurately quantifies the severe state of the system when a fault occurs. More importantly, it reveals the "lagging" defect of traditional EER indicators—they can only detect problems after a fault has caused substantial, macroscopic performance loss. The present invention, by integrating internal states, achieves a fundamental technological leap from "post-event diagnosis" to "pre-event warning."

[0030] Example 2 The dynamic heat load index of each terminal temperature control zone is used to characterize the dynamic heat load generated by comprehensively evaluating the number of people and their activity intensity within the terminal temperature control zone. Specifically, it involves obtaining the personnel density by comparing the real-time collected carbon dioxide concentration value within the terminal temperature control zone with a preset environmental reference concentration value, and obtaining the personnel activity level by acquiring the frequency or amplitude of motion events per unit time using infrared sensors or image analysis modules. The personnel density and personnel activity level are then multiplied by their respective preset weights, and the sum of these values ​​is the dynamic heat load index of that terminal temperature control zone. ; in, This represents the dynamic heat source load index of the j-th terminal temperature control zone. Indicates the weight of personnel density. This represents the real-time carbon dioxide concentration value within the terminal temperature control area. This indicates the preset environmental baseline concentration value; Indicates the weight of personnel activity intensity; This represents the activity level of personnel in the j-th terminal temperature control zone; This is the CO2 concentration difference, measured in ppm. It represents the number of people. This represents the personnel density weight, which is not only a weight but also a conversion factor. Its implicit unit can be understood as [dynamic load unit / ppm]. Therefore, the first calculation process is [dynamic load unit / ppm] × [ppm] = [dynamic load unit]; It is the level of personnel activity. The weight representing the intensity of personnel activity is also a conversion factor, and its unit is [dynamic load unit]. The calculation result of the second item is [dynamic load unit] × [dimensionless number] = [dynamic load unit] × [dimensionless number] = [dynamic load unit].

[0031] The heat loss pressure index of the corresponding pipeline for each terminal temperature control zone is used to characterize a composite index formed by the linear superposition of heat exchange loss and flow friction loss components. This index comprehensively quantifies the thermodynamic losses of the refrigerant during transport. Specifically, it involves obtaining the physical length of the pipeline connected to the terminal temperature control zone, the real-time temperature of the refrigerant inside the pipeline, the average ambient temperature of the pipeline, and the refrigerant mass flow rate estimated based on the system operating status. The physical length of the pipeline is multiplied by the absolute value of the temperature difference between the inside and outside of the pipeline and the square of the refrigerant mass flow rate. These two products are then multiplied by their respective preset heat exchange loss coefficient and flow friction loss coefficient, and summed. The final sum is the heat loss pressure index of the corresponding pipeline for that terminal temperature control zone. ; in, This represents the heat loss pressure index of the pipe connected to the j-th terminal temperature control zone. This indicates the heat exchange loss conversion factor, and the implicit unit is [pipeline loss unit / (m·K)]; This represents the actual length of the refrigerant piping from the outdoor unit to the j-th indoor unit; This indicates the temperature of the refrigerant flowing into the j-th terminal temperature-controlled zone; This indicates the average temperature of the environment in which the pipeline is located; This represents the conversion factor for flow friction loss, and its implicit unit is [pipeline loss unit / (m·(kg / s)²)]; This represents the refrigerant mass flow rate to the j-th terminal temperature control zone. The first part quantifies heat leakage or intrusion (heat loss) caused by the temperature difference between the inside and outside of the pipe. The longer the pipe and the greater the temperature difference, the more severe the loss. The second part, based on fluid mechanics principles, quantifies the pressure loss caused by friction when the refrigerant flows in a long pipe; this loss is proportional to the square of the flow rate. The first term... The calculation result is: [pipeline loss unit / (m×K)]×[m×K]=[pipeline loss unit][pipeline loss unit / (m×K)]×[m×K]=[pipeline loss unit]; Second item The calculation result is: [pipeline loss unit / (m × (kg / s)2)] × [m × (kg / s)2] = [pipeline loss unit].

[0032] The heat transfer index of the building envelope for each terminal temperature-controlled area is used to characterize the stable heat load transmitted through the building shell, determined by quantifying the building's physical characteristics and external meteorological conditions. Specifically, it involves: acquiring the real-time indoor-outdoor temperature difference and the real-time solar radiation intensity measured by a light sensor; calculating the product of the indoor-outdoor temperature difference and a preset comprehensive heat transfer coefficient representing the overall heat transfer capacity of the walls and roof of the terminal temperature-controlled area, multiplied by the area, to obtain the conductive heat component; calculating the real-time solar radiation intensity and the preset total window area and solar heat gain coefficient of the terminal temperature-controlled area to obtain the solar radiation heat component; and finally, weighted summing the conductive heat component and the solar radiation heat component to obtain the weighted sum value, which is the heat transfer index of the building envelope for the terminal temperature-controlled area. ; in, This represents the heat transfer index of the building envelope in the j-th terminal temperature control zone. and These represent the weights of the conduction and radiation heat gain, respectively. and Let represent the total heat transfer coefficient and total area of ​​the enclosure structure of the j-th terminal temperature control zone, respectively; and Let these represent the outdoor ambient temperature and the indoor temperature of the j-th terminal temperature control zone, respectively. This represents the Solar-Heat-Gain-Coefficient of the window in the j-th terminal temperature control zone, pre-inputted from architectural design data. This represents the total area of ​​the windows in the j-th end-temperature control zone. Indicates real-time solar radiation intensity; Part 1 The heat conduction caused by the temperature difference between indoors and outdoors was calculated. Part Two The solar radiation heat entering through the windows was calculated. This formula accurately describes the heat load under the combined effects of the building's inherent characteristics (U-value, window area) and the external environment (outdoor temperature, solar radiation). The calculation of the device heat dissipation index involves summing the current heat dissipation power of all registered electronic devices and lighting fixtures within the area that generate heat. For each individual device in the list, its instantaneous heat dissipation power is determined by multiplying its preset rated power, a heat dissipation coefficient characterizing its electrothermal conversion efficiency, and a real-time operating status factor (a value between 0 and 1) determined through real-time monitoring (such as smart socket readings) or a preset work schedule. Finally, the instantaneous heat dissipation power calculated for all devices is summed, and the total sum is the device heat dissipation index for that terminal temperature control area. ; in, This represents the heat dissipation index of the equipment in the j-th terminal temperature control zone. This represents the rated power of the i-th type of equipment; This represents the operating status factor of the i-th type of device at time t, where 0 represents off, 1 represents on, and a range of 0-1 represents partial load operation. The heat dissipation coefficient of the i-th type of equipment represents the proportion of electrical energy that is ultimately converted into heat energy. The comprehensive load index is obtained by linearly superimposing and summing the three indices representing the actual heat load source of the area—the dynamic heat source load index, the building envelope heat transfer index, and the equipment heat dissipation index—with the pipeline heat loss pressure index, which represents energy transmission loss, by multiplying each index by its respective preset weighting coefficients used to normalize the dimensions and adjust their influence priority. Among them, the dynamic heat source load index, the corresponding pipeline heat loss pressure index, the building envelope heat transfer index, and the equipment heat dissipation index of each terminal temperature control zone are positively correlated with the comprehensive load index. 1. Among them, the dynamic heat source load index of the j-th terminal temperature control zone. The core advantage of this design lies in its predictability. Traditional thermostats only react after the heat generated by people has significantly raised the room temperature, exhibiting a serious lag. This invention, by monitoring the two "causes" of heat load—CO2 concentration and human activity—rather than the "effect" of temperature, can predict load increases before heat has fully accumulated. For example, in a meeting room, the CO2 concentration rises rapidly after people sit down, and the dynamic heat source load index... As the temperature increases, the system can immediately begin to increase cooling output, thus suppressing the rise in perceived temperature and greatly improving dynamic comfort.

[0033] 2. Heat loss pressure index of the pipeline connected to the j-th terminal temperature control zone It is positively correlated with the comprehensive load index. Longer pipes, larger temperature differences between inside and outside the pipes, or larger refrigerant flow rates all lead to... Monotonically increasing.

[0034] Reasoning Process: This design solves the problem of "ignored energy transfer losses" in traditional control systems. In large-scale VRF systems, different indoor units exhibit significant differences in energy loss due to variations in pipe length and path. To achieve the same room temperature, a room 50 meters from the outdoor unit with pipes passing through a high-temperature zone requires a much larger "factory-set" cooling capacity than a room 5 meters away. This invention quantifies... By incorporating this portion of the transmission cost into the total load, precise energy allocation is achieved for each end point on demand, avoiding the problem of "not cooling at the far end and overcooling at the near end" caused by unified control; 3. The heat transfer index of the building envelope of the j-th terminal temperature control zone and the heat dissipation index of the equipment in the j-th terminal temperature control zone The comprehensive load index is positively correlated with the outdoor temperature; hotter outdoor temperatures, stronger solar radiation, or more indoor devices being used all lead to a monotonically increasing index. The value of these two indices lies in their real-time nature and precision. Traditional methods typically use a fixed, empirical estimate of building load based on season and building orientation. This invention transforms this vague estimate into a dynamic, instantaneous, and precise calculation through real-time sensor monitoring (light, temperature) and equipment status monitoring. For example, if a cloud drifts by in the afternoon and blocks the sun... The temperature will drop immediately, and the system will reduce its cooling output accordingly; when multiple computers in the office are awakened from sleep mode, The temperature will rise immediately, and the system will increase the cooling capacity accordingly. This real-time calculation capability eliminates energy waste or insufficient cooling capacity caused by static estimation; The following numerical example demonstrates how to calculate the comprehensive load index for two typical terminal temperature-controlled zones (Zone A and Zone B) in a large commercial building: Scene setting: Time: 2:00 PM on a weekday afternoon during summer.

[0035] Area A: An open-plan office area located on the south side of the building. It is large, densely populated, has many computer devices, a large glass curtain wall, and a long distance from the outdoor unit piping.

[0036] Area B: The archive storage room located on the north side of the building. It is small, sparsely populated, has only basic lighting, receives no direct sunlight, and is relatively close to the outdoor unit's piping.

[0037] The real-time collected data is shown in Table 2: Calculation process for each item: Calculation of the dynamic heat source load index of the region; Area A (Office Area): 0.7 × (1000) 400) + 0.3 × 0.6 = 420 + 0.18 = 420.18 (dimensionless); Area B (Archives): 0.7 × (450) 400) + 0.3 × 0.1 = 35 + 0.03 = 35.03; (dimensionless) Calculation of heat loss pressure index; Area A (Office Area) 0.2 × 80 × |28 10 | + 50 × 80 × (0.05) 2 =288+10=298 (dimensionless); Area B (Archives): 0.2 × 15 × |27 11 | +50 × 15 × (0.01) 2 =48 + 0.075 = 48.075 (dimensionless); Calculation of heat transfer index of building envelope; Area A (Office Area): 50 × (35 24)+(0.4×50×800)=1650+16000=17650W; Area B (Archives): 40 × (35) 24)+(0.63×2×50)=440+63=503W; Calculation of equipment heat dissipation index; Area A (Office Area): 4000W; Area B (Archives): 100W; This value is a physical quantity obtained by direct measurement or accumulation, and the unit is watts. To convert the original calculated values ​​of different physical units into a unified, dimensionless exponent (range 0-100), a preset maximum design value (MaxValue) for each load in each region needs to be set, and then multiplied by 100 as the normalization benchmark, as shown in Table 3 below: (0.35×52.52)+(0.15×74.50)+(0.30×88.25)+(0.20×80.00)=72.04; (0.35×35.03)+(0.15×60.09)+(0.30×83.83)+(0.20×50.00)=56.42; The final results show that the overall load index for Area A (office area) is 72.04; and the overall load index for Area B (archives room) is 56.42. The system internally stores an "optimal operating baseline" database (Look-up Table) pre-established through extensive experimental data or physical model simulations. This database establishes a mapping relationship between the "comprehensive load index" and various optimal operating parameters of the system, as shown in Table 4 below: Substituting the comprehensive load index of area A (office area) as 72.04, the controller is determined that in order to fully meet the complex needs composed of four dimensions (personnel, piping, enclosure, and equipment), the system needs to accurately output a total cooling capacity of 12,510W. Substituting the comprehensive load index of area B (archive area) into 56.42, the controller is determined that in order to fully meet the complex needs composed of four dimensions (personnel, piping, enclosure, and equipment), the system needs to accurately output a total cooling capacity of 9591. Specifically, the difference between the current speed and the target speed of the magnetic levitation fluorine pump, and the difference between the current opening and the target opening of the electronic expansion valve are calculated and executed as adjustment commands to drive each actuator to move smoothly toward the target value.

[0038] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value, it is acceptable.

[0039] The above formulas are all derived from software simulation using a large amount of data and are selected to be close to the actual values. The coefficients in the formulas are set by those skilled in the art according to the actual situation. The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any equivalent substitutions or changes made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the protection scope of the present invention.

Claims

1. A control method for adaptive adjustment of a magnetic levitation refrigerant pump air conditioner, the method being applied to a VRF air conditioning system, comprising: An outdoor unit comprising a magnetic levitation fluorine pump, multiple indoor units connected to the outdoor unit via refrigerant piping, and multiple electronic expansion valves respectively disposed at the refrigerant inlets of the multiple indoor units; characterized in that it includes the following steps: S1: Real-time acquisition of multi-source operational data characterizing the system's operating status. The multi-source operational data includes distributed heterogeneous load data acquired in multiple terminal temperature control zones defined by the multiple indoor units, as well as pump chamber internal status data. The distributed heterogeneous load data includes dynamic personnel heat load characterizing personnel activities, pipe network thermodynamic load characterizing pipeline pressure loss and heat exchange effect, and static terminal temperature control zone heat load characterizing heat transfer in the building envelope and heat dissipation in the equipment. S2: Based on the internal state data of the pump chamber, calculate the instantaneous operating efficiency index that quantifies the current two-phase flow state of the refrigerant and the operating efficiency; S3: Based on the distributed heterogeneous load data, calculate the dynamic heat source load index, corresponding pipeline heat loss pressure index, building envelope heat transfer index, and equipment heat dissipation index for each terminal temperature control zone, and obtain the comprehensive load index by correlation. S4: Based on the comprehensive load index, match the optimal operating baseline of the system. This optimal operating baseline is established in advance through physical model simulation or machine learning training on historical high-efficiency operating data to obtain the corresponding relationship between the total cooling capacity required by the comprehensive load index and the current total cooling capacity. Apply the Mahalanobis distance algorithm to calculate the distance between the current total cooling capacity and the total cooling capacity required by the optimal operating baseline as the operating deviation, and output the corresponding adjustment command.

2. The adaptive control method for a magnetic levitation fluorine pump air conditioner according to claim 1, characterized in that, After collecting the internal state data of the pump chamber, the process further includes: estimating key two-phase flow parameters characterizing the vapor-liquid distribution state within the pump chamber in real time based on the pump current and magnetic signals inside the pump chamber, wherein the key two-phase flow parameters include the vapor-liquid ratio and the vapor-liquid distribution index of the pump chamber; and performing nonlinear fusion of the key two-phase flow parameters and the pump chamber vibration signal to obtain a pump chamber stability index that comprehensively characterizes the fluid stability inside the pump chamber.

3. The adaptive control method for a magnetic levitation fluorine pump air conditioner according to claim 1, characterized in that, S2 specifically includes: Based on the pump current and magnetic signal in the internal state data of the pump chamber, the key parameters of the two-phase flow characterizing the vapor-liquid distribution state in the pump chamber are estimated in real time. The pump cavity vibration signal in the key parameters of the two-phase flow and the internal state data of the pump cavity is nonlinearly fused to obtain a pump cavity stability index that comprehensively characterizes the stability of the fluid inside the pump cavity. The instantaneous operating efficiency index is obtained by weighting and fusing the thermodynamic cycle efficiency coefficient, which characterizes the performance of the refrigeration cycle, the energy consumption ratio, which characterizes the unit power consumption, and the pump cavity stability index.

4. The adaptive control method for a magnetic levitation fluorine pump air conditioner according to claim 1, characterized in that, The dynamic heat load index of each terminal temperature control area is used to quantify the dynamic heat load generated by comprehensively evaluating the number of people and the intensity of their activities within the terminal temperature control area. Specifically, the difference between the carbon dioxide concentration value in the terminal temperature control area and the preset environmental benchmark concentration value is collected in real time to obtain the personnel density, and the frequency or amplitude of motion events obtained by the infrared sensor or image analysis module per unit time is used to obtain the personnel activity level. The personnel density and personnel activity level are multiplied by their respective preset weights, and the sum of the results is the dynamic heat load index of the terminal temperature control area.

5. The adaptive control method for a magnetic levitation fluorine pump air conditioner according to claim 1, characterized in that, The heat loss pressure index of the pipeline corresponding to each terminal temperature control zone is used to characterize a composite index formed by the linear superposition of heat exchange loss and flow friction loss components. It is used to comprehensively quantify the thermodynamic loss of refrigerant during transportation. Specifically, it involves obtaining the physical length of the pipeline connected to the terminal temperature control zone, the real-time temperature of the refrigerant inside the pipeline, the average temperature of the environment where the pipeline is located, and the refrigerant mass flow rate estimated based on the system operating status. The physical length of the pipeline is multiplied by the absolute value of the temperature difference between the inside and outside of the pipeline and the square of the refrigerant mass flow rate. The two products are then multiplied by their respective preset heat exchange loss coefficient and flow friction loss coefficient, and the sum is obtained. The final sum is the heat loss pressure index of the pipeline corresponding to the terminal temperature control zone.

6. The adaptive control method for a magnetic levitation fluorine pump air conditioner according to claim 1, characterized in that, The heat transfer index of the building envelope of each terminal temperature control zone is used to characterize the stable heat load transmitted through the building shell, which is determined by quantifying the building's physical characteristics and external meteorological conditions. Specifically, it involves: obtaining the real-time indoor and outdoor temperature difference and the real-time solar radiation intensity measured by a light sensor; calculating the indoor and outdoor temperature difference by multiplying it by a preset comprehensive heat transfer coefficient that characterizes the overall heat transfer capacity of the walls and roof of the terminal temperature control zone by the area to obtain the conductive heat component; and calculating the real-time solar radiation intensity by multiplying it by the preset total window area of ​​the terminal temperature control zone and its solar heat gain coefficient to obtain the solar radiation heat component. The heat transfer index of the building envelope of the terminal temperature control area is obtained by weighted summation of the conductive heat component and the solar radiation heat component.

7. The adaptive control method for a magnetic levitation fluorine pump air conditioner according to claim 1, characterized in that, The calculation of the device heat dissipation index is a process of summing up the current heat dissipation power of all registered electronic devices and lighting devices that generate heat within the area. For each individual device in the list, its instantaneous heat dissipation power is determined by multiplying its preset rated power, a heat dissipation coefficient characterizing its electrothermal conversion efficiency, and a real-time operating status factor determined by smart socket readings or preset work schedules. Finally, the instantaneous heat dissipation power calculated for all devices is summed up, and the sum is the device heat dissipation index of the terminal temperature control area.

8. The adaptive control method for a magnetic levitation fluorine pump air conditioner according to claim 7, characterized in that, The comprehensive load index is obtained by linearly superimposing and summing the three indices representing the actual heat load source of the area—the dynamic heat source load index, the building envelope heat transfer index, and the equipment heat dissipation index—with the pipeline heat loss pressure index, which represents energy transmission loss, by multiplying each index by its respective preset weighting coefficients used to normalize the dimensions and adjust their influence priority. Based on the comprehensive load index, the optimal operating baseline of the matching system is established in advance through physical model simulation or machine learning training on historical high-efficiency operating data to determine the relationship between the total cooling capacity required by the comprehensive load index and obtain the current total cooling capacity. The Mahalanobis distance algorithm is applied to calculate the distance between the current total cooling capacity and the total cooling capacity required by the optimal operating baseline, which is used as the operating deviation. Corresponding adjustment commands are output, including: calculating the difference between the current speed and the target speed of the magnetic levitation fluorine pump, and the difference between the current opening degree and the target opening degree of the electronic expansion valve, which are executed as adjustment commands.

9. An apparatus comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the device executes the computer program, it implements the steps of the adaptive adjustment control method for a magnetic levitation fluorine pump air conditioner as described in any one of claims 1 to 8.

10. A storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the adaptive adjustment control method for a magnetic levitation fluorine pump air conditioner as described in any one of claims 1 to 8.

Images