Air conditioning system control device and air conditioning system control method
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing air-conditioning systems require dedicated operations for parameter identification, leading to prolonged times for determining control parameters, especially in systems with large heat capacities, which can impair comfort and fail to account for disturbances like outdoor air temperature.
Innovation Solution
A controller that determines feedback control parameters during normal operation, using data from the air-conditioning system's heat characteristic model, allowing continuous temperature control and accounting for disturbances without deviating from the set target temperature.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the output of the PID controller is changed stepwise to identify parameters, then the parameters can be determined, but it requires dedicated operations that deviate from normal operation and may impair comfort
Solution Approach 1:
The system performs parameter identification in advance during normal operation by utilizing naturally occurring temperature variations and control actions, so that when parameter update is needed, the identification has already been completed without requiring dedicated identification operations that would disrupt comfort
Solution Approach 2:
The system uses its own normal operating data (temperature measurements and control outputs during regular operation) to perform parameter identification, eliminating the need for external dedicated identification procedures. The air-conditioning system identifies its own parameters using data generated during normal temperature control operations
2Measurement precision
If the primary delay time constant is calculated by waiting for the control target to reach 63% of final increase amount, then accurate parameter identification is achieved, but in systems with large heat capacity, this takes a long time and may impair comfort
Solution Approach 1:
Instead of waiting for the complete 63% rise to calculate the time constant, the system uses partial data from the temperature variation curve during normal operation to estimate parameters. By utilizing incremental information as it becomes available during normal operation rather than waiting for full response, the system reduces determination time while maintaining acceptable accuracy
Solution Approach 2:
The system performs parameter identification in advance during normal operation using naturally occurring temperature variations, so that parameter determination is completed before it would be needed, avoiding delays during actual operation
3Device complexity
If the control target is approximated using only dead time and primary delay time constant, then the model is simple, but the identified parameters may be greatly different depending on the initial state
Solution Approach 1:
The system transitions from a static approximation model to a dynamic model that accounts for multiple states. By incorporating multiple time constants and considering the system's behavior across different operating conditions rather than relying on a single initial state approximation, the model becomes more adaptive and produces consistent parameters regardless of starting conditions
4Device complexity
If the model is identified from only the output of the PID controller and the output of the control target, then the identification process is simple, but the influence of disturbances such as outdoor air temperature cannot be modeled
Solution Approach 1:
The system enhances the identification process to simultaneously capture both the system's inherent dynamics and the effects of external disturbances. By designing the identification approach to extract multiple time constants that represent different physical processes including disturbance responses, the model becomes universally applicable to various operating conditions and disturbance scenarios
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
A controller of an air-conditioning system comprises a feedback control unit (4) configured to generate a state quantity command from a control parameter and a difference between a target value of a room temperature and a measured room temperature and control the room temperature to the target value on the basis of the state quantity command, a control target heat characteristic model calculation unit (1) configured to calculate a parameter of a model regarding a heat characteristic of a control target (5) from at least a state quantity regarding an amount of heat supplied to the control target (5) or the state quantity command, and the measured room temperature during normal operation, and a control parameter determination unit (2) configured to determine the control parameter by using the parameter and a formula derived from the model regarding the heat characteristic of the control target (5).