Building management system with simulation and user action reinforcement machine learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Building automation systems face challenges in efficiently managing energy usage due to user-induced unscheduled environmental setpoint changes, which lead to energy inefficiencies and require frequent adjustments in operating conditions.
Innovation Solution
A method and system that simulate various operating values of building devices under different environmental conditions, assess penalties associated with these changes, and select the least penalized operating values based on future conditions to optimize energy usage and minimize user-induced inefficiencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If building automation systems operate based on fixed setpoints, then system operation is simple and reliable, but energy inefficiency occurs due to user-induced unscheduled setpoint changes
Solution Approach 1:
The system performs simulation of operating values before actual implementation. The building management system simulates various operating scenarios and selects optimal operating values in advance, preventing energy inefficiency caused by unscheduled setpoint changes before they occur.
Solution Approach 2:
The system uses reinforcement machine learning to automatically learn from user actions and environmental conditions, enabling the system to self-optimize operating values without requiring manual intervention or complex user input, thereby reducing energy waste while maintaining simplicity.
2Ease of operation
If the system allows frequent adjustments to accommodate user behavior, then user comfort is improved, but energy management efficiency deteriorates
Solution Approach 1:
The system incorporates feedback from user actions (penalties) into the reinforcement learning process. By monitoring unscheduled setpoint changes and using this feedback to adjust future operating decisions, the system balances user comfort needs with energy management efficiency.
Solution Approach 2:
The system dynamically changes operating parameters based on learned patterns from simulation and actual operation. By adjusting operating values according to environmental conditions and predicted user behavior, the system maintains comfort while improving energy management efficiency.
3Loss of energy
If the system implements simulation and penalty assessment, then energy efficiency is improved, but computational requirements and processing time increase
Solution Approach 1:
The system implements simulation selectively rather than continuously. By using reinforcement learning to identify key scenarios and operating values that require simulation, the system achieves energy efficiency improvements without requiring exhaustive simulation of all possible operating conditions, thus reducing processing time.
Data Source
AI summary
A method for controlling energy usage of one or more building devices associated with a building space including determining, by the one or more processing circuits based on a simulation, penalties associated with the one or more varied operating values of the one or more building devices, wherein the penalties indicate user behavior that causes energy inefficiency of the one or more building devices, and selecting, by the one or more processing circuits, one or more optimal operating values from the varied one or more operating values based on one or more future environmental conditions and a number of the penalties associated with each of the one or more varied operating values.