Comfort Model Extrapolation for HVAC Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing comfort management systems in buildings face challenges in effectively and efficiently initializing comfort models for incoming occupant profiles and new spatial elements, due to limited adaptability in complex operational environments with large numbers of occupant profiles, spatial elements, and changing architectural compositions.
Innovation Solution
The implementation of cross-model extrapolation, cross-profile extrapolation, and cross-space extrapolation methods using a comfort management computing device to generate unknown comfort models by aggregating and extrapolating from known comfort models associated with primary and secondary occupant profiles and spatial elements, enabling efficient initialization of comfort models for new scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional comfort model initialization methods are used, then the system structure remains simple, but the adaptability to complex operational environments with large numbers of occupant profiles and spatial elements deteriorates
Solution Approach 1:
The comfort model generation process is segmented into three distinct extrapolation methods: cross-model extrapolation, cross-profile extrapolation, and cross-space extrapolation. Each method handles specific scenarios (new models, new occupants, new spaces) separately, allowing the system to adapt to complex environments by selecting appropriate segmentation strategies without overwhelming system complexity.
Solution Approach 2:
The system performs preliminary actions by pre-establishing comfort models for secondary occupant profiles and known spatial elements before encountering new scenarios. When a new comfort model is needed, the system extrapolates from these pre-existing models rather than initializing from scratch, enabling rapid adaptation to complex operational environments.
2Productivity
If comfort models are manually initialized for each occupant and spatial element, then model accuracy is high, but the time and resources required increase significantly
Solution Approach 1:
The system creates copies of existing comfort models through extrapolation. Instead of manually initializing each new comfort model, the system copies and adapts patterns from known comfort models associated with similar occupant profiles and spatial elements, maintaining model reliability while dramatically improving initialization efficiency for large numbers of occupants and spaces.
Solution Approach 2:
The extrapolation methods adjust model parameters based on similarities between primary and secondary occupant profiles, or between known and unknown spatial elements. By changing parameters systematically rather than requiring complete manual initialization, the system maintains reliability while improving productivity in scaling to complex environments.
3Adaptability or versatility
If the system stores comfort data for all possible occupant profiles and spatial elements, then complete coverage is achieved, but the data storage requirements and system complexity increase
Solution Approach 1:
The system achieves universal coverage through extrapolation rather than explicit storage. A single set of known comfort models can serve multiple purposes: directly for known occupants/spaces and indirectly (through extrapolation) for new occupants and spaces. This multi-functionality allows the system to cover all scenarios without storing separate data for each possible case.
Solution Approach 2:
The extrapolation algorithms act as intermediaries between stored known comfort models and required unknown comfort models. Rather than storing all possible comfort data, the system stores compact known models and uses extrapolation as a mediator to generate additional models on-demand, reducing data storage requirements while maintaining complete coverage capability.
Data Source
AI summary
Method, apparatus and computer program product for comfort model extrapolation. For example, the apparatus includes at least one processor and at least one non-transitory memory including program code. The at least one non-transitory memory and the program code are configured to, with the at least one processor, obtain a plurality of known comfort models including: one or more cross-space comfort models each associated with a primary occupant profile and a known spatial element of one or more known spatial elements, one or more cross-profile comfort models each associated with a secondary occupant profile and an unknown spatial element, and one or more cross-context comfort models each associated with a secondary occupant profile and a known spatial element; and generate a unknown comfort model for the primary occupant profile and the unknown spatial element based on the plurality of known comfort models.