Predictive control for domestic heating system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Domestic gas boilers lack efficient control strategies for balancing thermal comfort and energy consumption, as existing thermostats often operate in switching modes due to oversizing and insufficient data for model predictive control (MPC) deployment, which is computationally intensive and uncommon in domestic use.
Innovation Solution
A method that utilizes gas consumption and temperature data to estimate parameters of a dynamic heat loss model for MPC, employing a two-level MPC scheme to manage the fast dynamics of the heating system and efficiency curve, with cloud computing aiding in model estimation and reducing computational demands on embedded devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If Model Predictive Control (MPC) is deployed for domestic heating systems, then control efficiency and thermal comfort are improved, but device complexity and computational requirements increase significantly
Solution Approach 1:
The patent replaces complex real-time computational mechanics with a pre-computed lookup table approach. Instead of performing intensive MPC calculations in real-time on domestic controllers, the system pre-calculates optimal control strategies offline and stores them in lookup tables, which are then accessed during operation. This substitution reduces real-time computational burden while maintaining MPC performance benefits.
Solution Approach 2:
The patent performs model identification and control parameter optimization in advance before actual heating operations. System parameters are identified offline using historical data, and MPC control tables are pre-computed based on these parameters. This preliminary action transfers computational complexity from real-time operation to offline setup, making MPC feasible for domestic applications with limited processing power.
2Loss of energy
If advanced control strategies are implemented, then energy savings and thermal comfort are improved, but ease of operation and deployment become more difficult
Solution Approach 1:
The patent implements self-service through automatic model identification using readily available historical data from standard domestic heating systems. The system automatically identifies thermal model parameters from past temperature and control data without requiring manual site surveys or expert intervention. This self-configuration capability significantly simplifies deployment while enabling advanced energy-saving control strategies.
Solution Approach 2:
The patent creates a universal deployment approach that works across different domestic heating systems by using generic thermal models that can be adapted to specific systems through automatic parameter identification. The lookup table structure and model identification algorithm are designed to be system-agnostic, allowing the same control strategy to be deployed across diverse domestic applications without custom engineering for each installation.
3Device complexity
If domestic boilers use standard thermostat switching control, then device complexity is reduced, but energy efficiency and modulation capability deteriorate
Solution Approach 1:
The patent implements periodic sampling of temperature data at fixed intervals to build the thermal model and generate control decisions. Instead of continuous complex calculations, the system uses regularly spaced measurements to update its state and select appropriate pre-computed control actions from lookup tables. This periodic approach maintains simplicity while improving efficiency over traditional on/off switching.
Solution Approach 2:
The patent changes the control parameter from simple binary on/off states to continuous modulation based on predicted thermal behavior. By using identified thermal model parameters (heat capacity, loss coefficients) to predict future temperature trends, the system adjusts heater output continuously to maintain optimal efficiency points, transforming the control from discrete switching to continuous parameter adjustment.
Data Source
AI summary
One or more systems, devices, and techniques for heating a structure using boiler water stored by a boiler are described herein. For example, a method includes obtaining gas consumption data for a heat source that is heating a structure and obtaining zone and ambient temperature data while the structure is being heated. Additionally, the method includes estimating unknown parameters of elements of a dynamic heat loss model as part of a model predictive control (MPC) model for the structure using inverse modeling as a function of the obtained gas consumption data and zone and ambient data.