A constant air volume control method and system for an air conditioning fan system

Abstract

The present invention belongs to the field of system control, and provides a constant air volume control method and system for an air conditioner fan system. A first machine learning model is used to predict environmental parameters, and the environmental parameters include temperature, humidity, and duct pressure; PID controller parameters are determined based on the predicted environmental parameters; based on the current system error and the predicted PID controller parameters, the air conditioner control parameters are dynamically adjusted; the system state is monitored in real time, and a feedback mechanism adjusts the control signal to maintain the set air volume target. The above solution can efficiently perform constant air volume control of the air conditioner fan system.

Description

Technical Field

[0001] The present invention belongs to the field of system control, and particularly relates to a constant air volume control method and system for an air conditioner fan system. Background Art

[0002] In an air conditioner fan system, constant air volume control is a key technology for maintaining the stability of the air supply volume within a set value range. The main goal of constant air volume control is to ensure that the air conditioner system can always supply air according to the predetermined air volume under various environmental conditions, so as to meet the requirements of indoor temperature, humidity, and air quality.

[0003] The operation of the air conditioner system is directly affected by external environmental conditions, especially the changes in factors such as temperature, humidity, and duct pressure. The fluctuations of these environmental parameters will cause changes in the load of the air conditioner system, thereby affecting the air supply volume. Due to its fixed control parameters, the traditional PID control method is difficult to maintain the stability of the system under these complex and changeable environments. When the temperature rises, the air conditioner system needs to increase the air volume for cooling; the change in humidity will change the dehumidification demand; the fluctuation of the duct pressure affects the resistance of air flow, increasing the requirement for the adjustment of the fan speed.

[0004] The air conditioner fan system needs to have good adaptive adjustment ability under different loads and environments. However, due to the slow response to environmental changes, the traditional control method usually requires manual intervention or excessive parameter adjustment during design to cope with different working conditions, which makes the adjustment of the system not flexible enough and inefficient. In scenarios with frequent load changes and complex environments, the traditional control method is difficult to meet the high-precision and high-efficiency air volume control requirements. Summary of the Invention

[0005] To solve the problems in the prior art, the present invention provides a constant air volume control method for an air conditioner fan system, and the method includes the following steps:

[0006] Use a first machine learning model to predict environmental parameters, where the environmental parameters include temperature, humidity, and duct pressure;

[0007] Determine the PID controller parameters based on the predicted environmental parameters;

[0008] Dynamically adjust the air conditioner control parameters based on the current system error and the predicted PID controller parameters;

[0009] Real-time monitor the system state, and use a feedback mechanism to adjust the control signal to maintain the set air volume target.

[0010] On the other hand, the present invention also provides a constant air volume control system for an air conditioner fan system, and the system includes the following modules:

[0011] A prediction module for predicting environmental parameters using a first machine learning model, where the environmental parameters include temperature, humidity, and air duct pressure;

[0012] A first calculation module for determining PID controller parameters based on the predicted environmental parameters;

[0013] A second calculation module for dynamically adjusting the air conditioner control parameters based on the current system error and the predicted PID controller parameters;

[0014] A loop module for real-time monitoring of the system state, adjusting the control signal through a feedback mechanism, and maintaining the set air volume target.

[0015] The constant air volume control method of the air conditioner fan system of the present invention has the following beneficial effects by adaptively adjusting the PID controller parameters and combining the machine learning model to predict environmental parameters:

[0016] The present invention can automatically adjust the parameters of the PID controller according to the real-time changes of temperature, humidity, air duct pressure, etc., so that the system always maintains the set air volume target, reduces the air volume fluctuation caused by environmental changes, and significantly improves the control accuracy of the system.

[0017] By introducing a machine learning model to predict environmental parameters (such as temperature, humidity, air duct pressure), the system can predict external environmental changes in advance and make corresponding adjustments, enhancing the adaptive ability of the system and enabling it to operate stably in a complex and changeable environment.

[0018] The adaptive PID control effectively avoids the oscillation or overshoot phenomenon in the traditional system, enables the air conditioner fan system to operate smoothly, and ensures the stability of the system under different load and environmental conditions.

[0019] Through the above beneficial effects, while maintaining the constant air volume of the air conditioner fan system, the present invention significantly improves the energy efficiency and reliability of the system, and is applicable to the air conditioner fan control in a variety of application scenarios. Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0021] Figure 1 It is a flowchart of the method of the present invention. Detailed Embodiments

[0022] Next, the present invention will be preferably described in combination with the drawings and specific embodiments.

[0023] This embodiment solves the above problems through the following steps:

[0024] In one embodiment, referring to Figure 1 , the present invention provides a constant air volume control method for an air-conditioning fan system, which aims to keep the air supply volume stable within a set value range by controlling key parameters in the air-conditioning fan system.

[0025] The air-conditioning fan system refers to a fan control system applied to air-conditioning equipment, mainly used to regulate the flow and distribution of air. The fan system controls the air flow by adjusting the speed of the fan or the opening of the air valve, and then adjusts the indoor temperature, humidity and air quality. This system usually consists of a fan, a motor, a controller, sensors, air ducts, etc., and is widely used in central air-conditioning systems, split air-conditioning systems, variable air volume air-conditioning systems, etc.

[0026] Constant air volume means that the air-conditioning fan system maintains a constant air supply volume during operation, avoiding fluctuations in air volume caused by changes in the external environment or system load fluctuations. Constant air volume control technology is used to ensure that the air flow output by the fan can continuously remain within the set target range, thereby ensuring the stability of the indoor environment.

[0027] Use a first machine learning model to predict environmental parameters, where the environmental parameters include temperature, humidity, and air duct pressure.

[0028] In the constant air volume control method of the present invention, a first machine learning model is used to predict environmental parameters, which include but are not limited to temperature, humidity, and air duct pressure. Specifically, the first machine learning model can, through the analysis and learning of historical data and real-time collected data of the air-conditioning system, predict in advance the key environmental parameters affecting the operation of the air-conditioning fan system, thereby providing accurate parameter input for subsequent control strategies.

[0029] The first machine learning model can adopt one or a combination of multiple machine learning algorithms to improve the accuracy of environmental parameter prediction.

[0030] The first machine learning model can be a regression model or a time series prediction model based on supervised learning or unsupervised learning. Preferably, this model can be a support vector machine regression (SVR), a random forest regression (Random Forest Regression), or a deep learning model based on a long short-term memory network (LSTM).

[0031] The input data of the first machine learning model includes real-time collected data from sensors and historical operation data, specifically including:

[0032] External temperature data (such as the measurement values of outdoor or indoor temperature sensors), which are used to capture the temperature change trend in the current and future periods;

[0033] Air humidity data, which are used to monitor the humidity level of the air and affect the dehumidification and air volume adjustment requirements of the air conditioning system;

[0034] Duct pressure data, which are collected by differential pressure sensors installed in the duct, reflect the resistance of air flow in the duct, and thus affect the load of the air conditioning system.

[0035] The machine learning model is trained with pre-collected historical environmental parameter data. The historical data include the changes in temperature, humidity, and duct pressure at different times, supplemented by relevant data such as the actual load and running time of the air conditioning system. In the model training stage, the model parameters are optimized through supervised learning to enable it to provide high-precision parameter prediction values in future prediction processes.

[0036] The machine learning model predicts environmental parameters within a certain future period based on the input real-time data and historical data, including temperature, humidity, and duct pressure at a certain future moment or within a certain future period. The prediction results can be provided to the control module of the air conditioning fan system for adjusting the operating parameters of the fan in advance, such as fan speed, damper opening, etc.

[0037] In a preferred embodiment, the first machine learning model can be implemented based on the long short-term memory network (LSTM) architecture. This model captures the time series correlations hidden in the input data to predict temperature, humidity, and duct pressure for a future period. The specific steps are as follows:

[0038] Collect temperature, humidity, and duct pressure data for the past 30 days;

[0039] Use this data to train the LSTM model so that the model can identify the patterns and trends of parameter changes;

[0040] Obtain the latest temperature, humidity, and duct pressure data in real time and input them into the trained LSTM model;

[0041] Output the predicted temperature, humidity, and duct pressure values within the next 24 hours.

[0042] By using the first machine learning model to predict the environmental parameters of the air conditioning system, the response speed and control accuracy to environmental changes can be improved, enabling the air conditioning fan system to continuously maintain a constant air volume under complex and changeable environmental conditions, and enhancing the stability and energy efficiency of the system.

[0043] Determine the PID controller parameters based on the predicted environmental parameters.

[0044] This system first predicts key environmental parameters through a machine learning model. The environmental parameters include, but are not limited to, temperature T ext , temperature H ext , and duct pressure P duct . These parameters affect the operating load of the air conditioning system and the air supply volume demand of the fan. Suppose the environmental parameters at future time t + k are predicted through the machine learning model at time t, and we get:

[0045] Predicted temperature represents the predicted value of the environmental temperature at future time t + k;

[0046] Predicted humidity represents the predicted value of the humidity at future time t + k;

[0047] Predicted duct pressure represents the predicted value of the duct pressure at future time t + k.

[0048] The PID controller (Proportional-Integral-Derivative Controller) is a feedback control system widely used in the industrial control field. It generates a control signal by performing three operations on the error of the system (the difference between the set value and the actual output value) - proportional (Proportional, P), integral (Integral, I), and derivative (Derivative, D) - to adjust the behavior of the system and make the output of the system as close as possible to the set target value.

[0049] The PID controller makes corresponding adjustments according to the current error of the system and the change of the error over time. Its control formula is:

[0050]

[0051] Where:

[0052] u(t) is the output signal of the controller, usually used to adjust the actuator of the system (such as motors, heaters, etc.);

[0053] e(t) = r(t) - y(t) is the error of the system, representing the difference between the set value r(t) and the actual output y(t);

[0054] K p is the proportional coefficient, indicating the intensity of proportional control; proportional control adjusts according to the magnitude of the current error. The larger the error, the larger the output control signal.

[0055] K iis the integral coefficient, representing the intensity of integral control; integral control adjusts according to the accumulated value of the error. It can eliminate the steady-state error in the system, that is, the situation where the system cannot reach the set value after long-term operation.

[0056] K d is the derivative coefficient, representing the intensity of derivative control. Derivative control adjusts according to the rate of change of the error. It can predict the trend of the error, reduce the oscillation of the system, and improve the stability of the system.

[0057] In the present invention, according to the changes in temperature, humidity, and duct pressure, the respective parameters of the PID controller will be dynamically adjusted to ensure constant air volume control of the air conditioning system under different environmental conditions.

[0058] External temperature T ext affects the load demand of the air conditioning system. A higher temperature means that the system needs to increase the air volume for cooling. Therefore, when a higher temperature is predicted in the future, the proportionality coefficient K p needs to be increased to accelerate the response speed of the air volume adjustment. The dynamic adjustment of the proportionality coefficient can be expressed by the following formula:

[0059]

[0060] Where:

[0061] K p0 is the initial proportionality coefficient of the system at normal temperature;

[0062] is the predicted value of the environmental temperature at the future time t + k

[0063] T ref is the reference temperature (usually the ideal temperature during system design, which may be the median value within the comfortable temperature range);

[0064] α T is the sensitivity coefficient of the temperature change to the proportionality coefficient K p of.

[0065] When is greater than T ref , the proportionality coefficient K p increases to accelerate the response speed; on the contrary, when the temperature is lower than the reference temperature, K p decreases to avoid overreaction.

[0066] Humidity H ext The increase in will increase the dehumidification load of the air conditioning system, and the change in humidity generally leads to a long-term steady-state error in the system. If the system operates under high humidity conditions for a long time, the steady-state error will accumulate. Therefore, the integral coefficient K iIt needs to be increased to eliminate this error more quickly. The adjustment formula for the integral coefficient can be expressed as:

[0067]

[0068] Where:

[0069] K i0 is the initial integral coefficient of the system under normal humidity;

[0070] is the predicted future humidity at time t + k;

[0071] H ref is the reference humidity;

[0072] α H is the sensitivity coefficient of humidity change to the integral coefficient K i ;

[0073] When the humidity is higher than H ref , K i increases to help eliminate the system error more quickly; when the humidity is low, K i can be appropriately reduced to prevent the integral effect from being too strong and causing system instability.

[0074] The air duct pressure P duct is an important parameter reflecting the system resistance. When the air duct pressure rises, it indicates that the air flow is blocked and the load on the fan increases. To avoid oscillations caused by pressure fluctuations, the differential coefficient K d needs to be more sensitive, thus smoothing the system response and reducing fluctuations caused by pressure changes. The adjustment formula for the differential coefficient can be expressed as:

[0075]

[0076] Where:

[0077] K d0 is the initial differential coefficient of the system under normal air duct pressure;

[0078] is the predicted future air duct pressure at time t + k;

[0079] P ref is the reference pressure of the air duct;

[0080] α P is the sensitivity coefficient of air duct pressure change to the differential coefficient K d ;

[0081] When is higher than P ref , the system needs to increase Kd , thereby improving the sensitivity of the response and reducing the oscillation caused by pressure fluctuations; when the pressure is low, the system can reduce K d to ensure the stable operation of the system.

[0082] Through the above formula, the system can adaptively adjust the parameters K p 、K i 、K d of the PID controller when environmental parameters such as temperature, humidity, and air duct pressure change, so as to maintain the constant air volume control of the air-conditioning fan system under complex environmental conditions and ensure the stability and accuracy of the system.

[0083] Dynamically adjust the air-conditioning control parameters based on the current system error and the predicted PID controller parameters. The dynamic adjustment process of the control parameters specifically includes the following steps:

[0084] Real-time collect the error between the set air volume and the actual air volume during the operation of the system

[0085] e(t) = Q s (t) - Q m (t)

[0086] e(t) is the current system error;

[0087] Q s (t) is the target air volume set by the system;

[0088] Q m (t) is the actual air volume of the system at the current moment t.

[0089] Based on the current error e(t) and the predicted PID controller parameters K p (t + k), K i (t + k) and K d (t + k), calculate the control signal u(t) in real time and dynamically adjust the control parameters (fan speed) of the air-conditioning system.

[0090] The generation formula of the control signal u(t) is:

[0091]

[0092] Real-time monitor the system status, and the feedback mechanism adjusts the control signal to maintain the set air volume target.

[0093] The system real-time collects key parameters related to air volume control through multiple sensors, including:

[0094] Air volume sensor data: used to measure the actual air volume of the air-conditioning fan system at time t;

[0095] Air duct pressure sensor data: used to monitor the air pressure inside the air duct, reflecting the flow resistance and air volume change in the air duct;

[0096] Ambient parameter data: real-time monitoring of the external environmental temperature and humidity to ensure that the system can accurately adjust the air volume under different environmental conditions.

[0097] The system uses a closed-loop feedback control mechanism to compare the monitored actual air volume with the set target air volume and generate an error signal. This error signal is used to guide the controller to adjust the operating parameters of the air conditioning system to reduce the error and ensure that the air volume is stable near the set value.

[0098] Based on the error, a control signal is generated through a PID controller. The control signal is used to adjust the execution unit of the system, such as the fan speed. That is, it returns to the previous step again, and the control parameters are determined cyclically for dynamic control, that is, based on the real-time error and the predicted PID controller parameters, the air conditioning control parameters are dynamically adjusted.

[0099] On the other hand, the present invention also provides a constant air volume control system for an air conditioning fan system, including:

[0100] A prediction module for predicting ambient parameters using a first machine learning model, where the ambient parameters include temperature, humidity, and air duct pressure;

[0101] A first calculation module for determining the PID controller parameters based on the predicted ambient parameters;

[0102] A second calculation module for dynamically adjusting the air conditioning control parameters based on the current system error and the predicted PID controller parameters;

[0103] A loop module for real-time monitoring of the system state and adjusting the control signal through a feedback mechanism to maintain the set air volume target.

[0104] For the part of the module structure not specifically defined in the present invention, it shall be subject to the content recorded in the prior art. The prior art mentioned in the foregoing background art part and the specific embodiment part of the present invention can be used as a part of the present invention to understand the meaning of some technical features or parameters.

Claims

1. A constant air volume control method for an air conditioning fan system, characterized in that: The method comprises the following steps: Predicting environmental parameters using a first machine learning model, the environmental parameters including temperature, humidity, and air duct pressure; Determining PID controller parameters based on predicted environmental parameters; Dynamically adjust air conditioning control parameters based on current system errors and predicted PID controller parameters; Real-time monitoring of system status, feedback mechanism adjusts control signals to maintain set air volume targets; Determining the PID controller parameters based on the predicted environmental parameters includes: Adjusting the proportional coefficient of the PID controller by predicting the temperature; adjusting the integral coefficient of the PID controller by predicting humidity; Adjusting the differential coefficient of the PID controller by predicting the air duct pressure; The dynamically adjusting the air conditioning control parameters based on the current system error and the predicted PID controller parameters includes: Substituting the system error and the proportional coefficient, integral coefficient and differential coefficient into the PID control equation to obtain the air conditioning control parameter; The real-time monitoring of the system status and the feedback mechanism to adjust the control signal to maintain the set air volume target include: Calculating the real-time error of the system in real time; dynamically adjusting the air conditioning control parameters based on the real-time error and the predicted PID controller parameters; The air conditioning control parameter is the fan speed; The generation formula of the control signal u(t) is: in, K p (t+k) represents the dynamic adjustment amount of the proportional coefficient; K p0 is the initial proportionality coefficient of the system at normal temperature; is the predicted value of the ambient temperature at the future time t+k; T ref is the reference temperature; α T is the temperature change proportionality coefficient K p The sensitivity coefficient of e(t)=Q s (t)-Q m (t) e(t) is the current system error; Q s (t) is the target air volume set for the system; Q m (t) is the actual air volume of the system at the current time t; K i (t+k) is the adjustment amount of the integral coefficient; K i0 is the initial integration coefficient of the system under normal humidity; is the predicted future humidity at time t+k; H ref is the base humidity; α H is the integral coefficient K of humidity change i The sensitivity coefficient of K d (t+k) is the adjustment amount of the differential coefficient; K d0 is the initial differential coefficient of the system at normal duct pressure; is the future duct pressure predicted at time t+k; P ref is the base pressure of the air duct; α P is the differential coefficient K of the duct pressure change d The sensitivity coefficient of .

2. A constant air volume control method for an air conditioning fan system according to claim 1, characterized in that: The machine learning model is a long short-term memory network model.

3. A constant air volume control system for an air conditioning fan system, characterized in that: The system includes the following modules: A prediction module, configured to predict environmental parameters using a first machine learning model, wherein the environmental parameters include temperature, humidity, and air duct pressure; A first calculation module, used to determine PID controller parameters based on predicted environmental parameters; A second calculation module is used to dynamically adjust the air conditioning control parameters based on the current system error and the predicted PID controller parameters; The circulation module is used to monitor the system status in real time, and the feedback mechanism adjusts the control signal to maintain the set air volume target; Determining the PID controller parameters based on the predicted environmental parameters includes: Adjusting the proportional coefficient of the PID controller by predicting the temperature; adjusting the integral coefficient of the PID controller by predicting humidity; Adjusting the differential coefficient of the PID controller by predicting the air duct pressure; The dynamically adjusting the air conditioning control parameters based on the current system error and the predicted PID controller parameters includes: Substituting the system error and the proportional coefficient, integral coefficient and differential coefficient into the PID control equation to obtain the air conditioning control parameter; The real-time monitoring of the system status and the feedback mechanism to adjust the control signal to maintain the set air volume target include: Calculating the real-time error of the system in real time; dynamically adjusting the air conditioning control parameters based on the real-time error and the predicted PID controller parameters; The air conditioning control parameter is the fan speed; The generation formula of the control signal u(t) is: in, K p (t+k) represents the dynamic adjustment amount of the proportional coefficient; K p0 is the initial proportionality coefficient of the system at normal temperature; is the predicted value of the ambient temperature at the future time t+k; T ref is the reference temperature; α T is the temperature change proportionality coefficient K p The sensitivity coefficient of e(t)=Q s (t)-Q m (t) e(t) is the current system error; Q s (t) is the target air volume set for the system; Q m (t) is the actual air volume of the system at the current time t; K i (t+k) is the adjustment amount of the integral coefficient; K i0 is the initial integration coefficient of the system under normal humidity; is the predicted future humidity at time t+k; H ref is the base humidity; α H is the integral coefficient K of humidity change i The sensitivity coefficient of K d (t+k) is the adjustment amount of the differential coefficient; K d0 is the initial differential coefficient of the system at normal duct pressure; is the future duct pressure predicted at time t+k; P ref is the base pressure of the air duct; α P is the differential coefficient K of the duct pressure change d The sensitivity coefficient of .

4. A constant air volume control system for an air conditioning fan system according to claim 3, characterized in that: The machine learning model is a long short-term memory network model.

Images