Air-Conditioning Airflow Control Using Reduced-Order Observer Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Air-conditioning systems face challenges in real-time control due to the complexity of airflow models, which are often of infinite dimension and difficult to use in real-time applications, and changes in system conditions such as refrigerant leaks or corrosion, leading to inefficiencies and inaccurate heat load management.
Innovation Solution
A reduced-order model is used to simplify airflow dynamics, transforming partial differential equations into ordinary differential equations, with a Lyapunov approach to handle uncertainties in physical parameters, allowing for real-time adaptation and control of air-conditioning systems by projecting the physical model onto a finite-dimensional space and incorporating a term that reduces observation errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a physical model of airflow is used to describe the system, then measurement precision is improved, but device complexity increases because the model is of infinite dimension and too complex for real-time control
Solution Approach 1:
The patent segments the infinite-dimensional physical model into a finite-dimensional reduced-order model that captures the dominant airflow dynamics. This segmentation allows real-time control while maintaining adequate measurement precision by retaining only the most significant model components.
Solution Approach 2:
The patent extracts the essential dynamics from the complex physical model by identifying and retaining only the dominant modes of airflow behavior. This extraction process creates a simplified model that is computationally tractable for real-time control while preserving the critical measurement characteristics.
2Use of energy by moving object
If model-based methods are used to predict input combinations, then energy efficiency is improved, but adaptability worsens because the models do not account for installation-specific characteristics and system changes over time
Solution Approach 1:
The patent implements feedback mechanisms that continuously update the reduced-order model parameters based on actual system performance and observed airflow patterns. This feedback loop enables the system to adapt to installation-specific characteristics and changes over time (such as refrigerant leaks or corrosion) while maintaining energy-efficient operation.
Solution Approach 2:
The patent allows model parameters to change dynamically based on operating conditions and system degradation. By adjusting parameters in real-time, the system maintains accuracy and energy efficiency despite changes in system characteristics or installation-specific variations.
3Ease of operation
If conventional control methods are used, then ease of operation is maintained, but productivity decreases because the system cannot optimize energy consumption in real-time
Solution Approach 1:
The patent creates a simplified computational copy of the physical system through the reduced-order model. This copy can be manipulated easily for real-time optimization without the complexity of the full physical model, thereby maintaining ease of operation while enabling advanced energy optimization and improving overall productivity.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A system for controlling an operation of an air-conditioning system generating airflow in a conditioned environment, includes a set of sensors to produce measurements of the airflow in a set of points in the conditioned environment, a memory to store a model of the airflow dynamics including a combination of a first term transitioning a previous state of the airflow to a current state of the airflow and a second term assisting the transitioning, an observer to estimate the current state of the airflow in the conditioned environment by transitioning the previous state of the airflow forward in time according to the model of airflow dynamics to reduce the observation error in the current state of the airflow, and a controller to control the air-conditioning system based on the current state of the airflow. The first term in the model includes a projection of a physical model of the airflow on a finite-dimensional space. The physical model of the airflow includes physical parameters of the conditioned environment and the projection preserves the physical parameters of the conditioned environment in the first term. The second term includes a function of the range of the bounded uncertainty of the physical parameter, a negative gain, and an observation error between the measurements of the airflow in the set of points and estimations of the airflow in the set of points according to the model of the airflow.