System and method for controlling operations of air-conditioning system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Air-conditioning systems face challenges in real-time control due to complex, infinite-dimensional physical models of airflow dynamics, which are difficult to use in real-time applications and fail to adapt to changing conditions such as refrigerant leaks or corrosion, and ignore installation-specific characteristics.
Innovation Solution
A reduced-order model is used, transforming partial differential equations (PDEs) into ordinary differential equations (ODEs) with a combination of a projection term preserving physical parameters and an energy-based term to handle uncertainties, allowing for real-time control and adaptation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a physical model of airflow dynamics is used to control the air-conditioning system, then the accuracy of heat load prediction is improved, but the device complexity increases due to the infinite dimension and complexity of the model
Solution Approach 1:
The patent creates a simplified copy of the complex physical airflow model that retains the essential dynamics while reducing mathematical complexity. Instead of using the full infinite-dimensional physical model, a reduced-order model is constructed that copies the key behavioral characteristics, enabling real-time control applications while maintaining prediction accuracy.
Solution Approach 2:
The patent transforms the complex physical model by changing its parameter representation. The infinite-dimensional physical model is converted into a finite-dimensional reduced-order model with a manageable number of parameters, making it suitable for real-time control while preserving the essential thermal dynamics for accurate heat load prediction.
2Device complexity
If a fixed model is used for air-conditioning control, then the device complexity is reduced, but the adaptability worsens when system conditions change over time
Solution Approach 1:
The patent implements a dynamic model that adapts to changing system conditions. The reduced-order model includes time-varying parameters that can adjust to reflect system degradation, refrigerant leaks, or changing installation characteristics, enabling the control system to maintain accuracy over time without requiring complex reconfiguration.
Solution Approach 2:
The patent incorporates feedback mechanisms that allow the model to learn from actual system behavior. By comparing model predictions with actual measurements and adjusting parameters accordingly, the system adapts to changing conditions while maintaining a relatively simple model structure suitable for real-time control.
3Loss of energy
If conventional model-based energy efficiency methods are used, then the energy optimization is improved, but the ease of operation worsens due to the complexity of implementing and maintaining accurate physical models
Solution Approach 1:
The patent creates a simplified copy of the complex physical airflow model that retains the essential dynamics while reducing mathematical complexity. Instead of using the full infinite-dimensional physical model, a reduced-order model is constructed that copies the key behavioral characteristics, enabling real-time control applications while maintaining prediction accuracy.
Solution Approach 2:
The patent transforms the complex physical model by changing its parameter representation. The infinite-dimensional physical model is converted into a finite-dimensional reduced-order model with a manageable number of parameters, making it suitable for real-time control while preserving the essential thermal dynamics for accurate heat load prediction.
Data Source
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.