Building management autonomous HVAC control using reinforcement learning with occupant feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
HVAC systems in commercial buildings face inefficiencies due to unreliable occupancy sensors, leading to unnecessary heating or cooling of unoccupied spaces, resulting in energy waste.
Innovation Solution
Implementing voice assist devices to control temperature by analyzing occupant feedback, using natural language statements to adjust HVAC settings based on sentiment and historical data, thereby optimizing energy usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional occupancy sensors are used to control HVAC systems, then automated temperature adjustment is achieved, but false-positives occur causing unnecessary heating or cooling of unoccupied spaces
Solution Approach 1:
The patent combines multiple occupancy detection methods including traditional sensors, voice assist device presence detection, and sentiment analysis to create a more reliable occupancy determination system. This multi-layered approach cross-validates occupancy status, reducing false-positives while maintaining automated control.
Solution Approach 2:
The system uses sentiment analysis from voice assist devices as feedback to verify occupancy status. When sensors indicate occupancy but sentiment analysis shows no human presence or negative sentiment about temperature, the system adjusts or overrides the temperature setting, creating a feedback loop that improves reliability.
2Reliability
If HVAC systems continuously heat or cool spaces based on sensor data, then temperature comfort is maintained, but energy waste occurs in unoccupied zones
Solution Approach 1:
The system dynamically adjusts temperature setpoints based on real-time occupancy determination. When occupancy is confirmed through multiple validation methods, comfortable temperatures are maintained. When occupancy is doubtful or negative sentiment is detected, the system dynamically overrides to energy-saving temperatures, balancing comfort and energy efficiency.
Solution Approach 2:
The patent changes the temperature parameter dynamically based on occupancy confidence levels. High confidence occupancy triggers comfortable temperature ranges, while low confidence or detected unoccupancy triggers energy-saving temperature adjustments, optimizing the balance between comfort and energy consumption.
3Loss of energy
If programmable thermostats are used with pre-set schedules, then energy conservation is achieved during non-working hours, but flexibility is lost when occupancy patterns change
Solution Approach 1:
The system uses voice assist devices already present in occupied spaces as self-service occupancy detectors. These devices naturally detect presence and sentiment without additional infrastructure, allowing the HVAC system to automatically adapt to changing occupancy patterns while maintaining energy efficiency.
Solution Approach 2:
The system performs preliminary occupancy validation using voice assist devices before committing to temperature adjustments. By pre-checking occupancy status and sentiment, the system avoids unnecessary heating or cooling while maintaining flexibility to respond to actual occupancy patterns.
Data Source
AI summary
A building management system includes one or more processors, and one or more computer-readable storage media communicably coupled to the one or more processors and having instructions stored thereon that cause the one or more processors to: define a state of a zone or space within a building; control an HVAC system to adjust a temperature of the zone or space corresponding to a first action; receive utterance data from a voice assist device located in the zone or space; analyze the utterance data to identify a sentiment relating to the temperature of the zone or space; calculate a reward based on the state, the first action, and the sentiment; determine a second action to adjust the temperature of the zone or space based on the reward; and control the HVAC system to adjust the temperature of the zone or space corresponding to the second action.