Building control system with zone grouping based on predictive models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of building systems, such as HVAC systems, leads to inefficiencies in processing multiple predictive models, resulting in delays and increased costs due to excessive computational complexity when performing model predictive control (MPC) across multiple zones.
Innovation Solution
A controller that performs clustering analysis to group zones with similar model parameters, generating zone group models to reduce computational complexity by allowing a single MPC calculation to be applied across all zones in a group, thereby optimizing processing time and energy efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple predictive models are generated for different zones, then measurement precision and control accuracy are improved, but device complexity and processing time increase
Solution Approach 1:
The patent combines multiple zone-specific predictive models into a single building-level predictive model by aggregating thermal responses and model parameters across zones. This merging reduces computational complexity while maintaining adequate control accuracy for building-wide HVAC optimization.
Solution Approach 2:
The patent creates a universal building-level predictive model that serves multiple zones simultaneously, replacing the need for separate zone-specific models. This multi-functional approach allows a single model to provide control decisions for the entire building, reducing processing requirements.
2Measurement precision
If multiple predictive models are generated for different zones, then control accuracy is improved, but productivity and processing speed decrease
Solution Approach 1:
The patent merges multiple zone models into a single building-level model, significantly reducing the number of computational operations required for predictive control calculations while maintaining the ability to accurately predict and optimize building-wide thermal conditions.
Solution Approach 2:
The patent segments the building into thermal zones for model aggregation purposes, allowing the predictive model to account for zone-specific thermal characteristics through parameter aggregation while using a unified computational framework that improves processing speed.
3Measurement precision
If multiple predictive models are generated for different zones, then model accuracy is improved, but loss of time due to computational delays increases
Solution Approach 1:
The patent combines multiple zone-specific predictive models into a single building-level model, reducing computational complexity and processing time for generating control decisions while maintaining the accuracy needed for effective predictive control of building HVAC systems.
Data Source
AI summary
A controller for operating building equipment of a building. The controller includes one or more processors. The controller includes one or more non-transitory computer-readable media storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations. The operations include comparing one or more model parameters of predictive models describing zones of the building to determine one or more zone groups for the building. The operations include generating one or more zone group models corresponding to the one or more zone groups. The operations include operating the building equipment using the one or more zone group models to affect a variable state or condition of the building.