Q-learning based model-free control method for indoor thermal environment of aged care building
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing aged care buildings lack flexible indoor temperature control mechanisms that consider the impact of indoor temperature on cardiovascular health, particularly for the elderly, leading to increased cardiovascular disease risk.
Innovation Solution
A Q-learning based model-free control method utilizing monitored indoor temperatures and cardiovascular parameters like heart rate and systolic pressure to optimize heating, ventilation, and air conditioning systems, employing a Q-learning algorithm to achieve a dynamic thermal environment suitable for the elderly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If traditional fixed temperature control is used in aged care buildings, then energy consumption is reduced, but thermal comfort and cardiovascular health of the elderly deteriorate
Solution Approach 1:
The patent implements dynamic temperature control by using reinforcement learning algorithms to continuously adjust indoor temperature based on real-time physiological parameters (heart rate, blood pressure) of elderly residents. The system transitions from static fixed-temperature control to dynamic adaptive control, allowing the thermal environment to respond to changing physiological states and external conditions, thereby improving both health outcomes and energy efficiency through intelligent optimization.
Solution Approach 2:
The system establishes a closed-loop feedback mechanism by continuously monitoring physiological parameters (heart rate, blood pressure) and environmental temperature, then using this feedback to adjust HVAC control decisions. The reinforcement learning agent learns from feedback signals including health status changes and energy consumption patterns to optimize temperature control strategies, resolving the contradiction between energy savings and health comfort.
2Reliability
If reinforcement learning based control is implemented, then thermal comfort and health monitoring are improved, but system complexity increases
Solution Approach 1:
The patent introduces a reinforcement learning agent as an intermediary between the HVAC system and physiological monitoring devices. This intelligent mediator processes complex multi-parameter data (temperature, heart rate, blood pressure) and translates it into optimized control decisions, reducing the complexity burden on individual components while improving overall system reliability through integrated intelligent processing.
Solution Approach 2:
The reinforcement learning system implements self-service by autonomously learning optimal control strategies from historical data and real-time feedback without requiring manual programming or complex configuration. The system automatically adapts to individual elderly residents' physiological characteristics and preferences, reducing the need for manual system management and simplifying operational complexity.
3Measurement precision
If real-time physiological parameter monitoring is performed, then cardiovascular health prediction is improved, but data processing requirements increase
Solution Approach 1:
The system performs preliminary action by pre-processing and storing physiological parameter data as it becomes available, rather than waiting for batch processing. The reinforcement learning agent continuously updates its knowledge base with incoming data, enabling real-time health status assessment and predictive analytics without significant processing delays, thus maintaining measurement precision while minimizing time loss.
Data Source
AI summary
The present disclosure provides a Q-learning based model-free control method for an indoor thermal environment of an aged care building and belongs to the technical field of building environment control. According to the present disclosure, the monitored indoor temperatures of individual users and the heart rate and systolic pressure data of the aged are used as input data to a constructed Q-learning model, thus outputting a running control policy for a heating, ventilation and air conditioning system in the corresponding building. As a result, the control efficiency of the indoor temperature and the energy efficiency of the heating, ventilation and air conditioning system are improved. Compared with a traditional control model, the reinforced learning method based on the Q-learning theory can realize more accurate prediction on the cardiovascular health risk of the aged and can create a dynamic indoor thermal environment more suitable for the physical health of the aged.