Predictive Facility Control via Learned User Preferences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing environmental control systems for facilities often rely on prescriptive controls based on objective factors, neglecting individual user preferences, which can lead to discomfort for occupants.
Innovation Solution
A method using a machine learning model that learns user preferences through input and predicts future device states, allowing for personalized control of facilities by considering user input, historical data, and preferences of similar users or facilities, and implementing these predictions through a network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If prescriptive control based on objective factors is used, then energy efficiency is improved, but occupant comfort deteriorates due to neglecting individual user preferences
Solution Approach 1:
The system implements feedback loops where user preferences and behavioral data are continuously collected, processed, and used to adjust environmental control settings. This allows the system to learn from user responses and refine its predictions, balancing energy efficiency with personalized comfort requirements through iterative improvement based on actual user feedback
Solution Approach 2:
The system performs preliminary actions by predicting future user preferences and pre-adjusting environmental settings before users actually need them. The machine learning model analyzes historical data and patterns to anticipate when users will want specific environmental conditions, allowing the system to proactively create comfortable environments while optimizing energy usage
2Ease of operation
If personalized control based on user preferences is implemented, then occupant comfort is improved, but device complexity increases due to machine learning models and data processing requirements
Solution Approach 1:
The system provides self-service by automatically learning user preferences and making control decisions without requiring complex user interfaces or manual programming. The machine learning model autonomously processes user feedback and environmental data to generate optimized control settings, reducing the perceived complexity for users while maintaining sophisticated personalized control capabilities
Solution Approach 2:
The system merges multiple functions including data collection, machine learning inference, prediction generation, and device control into an integrated architecture. By combining these previously separate components into a unified system, the patent reduces overall complexity while maintaining the ability to provide personalized comfort control through coordinated operation of merged functional elements
Data Source
AI summary
In various embodiments, methods, systems, software, and apparatuses for controlling devices of a facility are provided, e.g., based on user input and/or user preferences.