View source for 20240019156. IDENTIFYING SUITABLE MODELS FOR ADAPTIVE MODEL PREDICTIVE CONTROL OF BUILDING HVAC USING MOVING HORIZON ESTIMATION simplified abstract (Palo Alto Research Center Incorporated)

=IDENTIFYING SUITABLE MODELS FOR ADAPTIVE MODEL PREDICTIVE CONTROL OF BUILDING HVAC USING MOVING HORIZON ESTIMATION=


==Organization Name==

[[:Category:Palo Alto Research Center Incorporated|Palo Alto Research Center Incorporated]]


[[Category:Palo Alto Research Center Incorporated]]

==Inventor(s)==

[[:Category:Saman Mostafavi of San Francisco CA (US)|Saman Mostafavi of San Francisco CA (US)]][[Category:Saman Mostafavi of San Francisco CA (US)]]

[[:Category:Harish Doddi of Palo Alto CA (US)|Harish Doddi of Palo Alto CA (US)]][[Category:Harish Doddi of Palo Alto CA (US)]]

[[:Category:Yu-Wen Lin of Palo Alto CA (US)|Yu-Wen Lin of Palo Alto CA (US)]][[Category:Yu-Wen Lin of Palo Alto CA (US)]]

[[:Category:David Schwartz of Palo Alto CA (US)|David Schwartz of Palo Alto CA (US)]][[Category:David Schwartz of Palo Alto CA (US)]]

==IDENTIFYING SUITABLE MODELS FOR ADAPTIVE MODEL PREDICTIVE CONTROL OF BUILDING HVAC USING MOVING HORIZON ESTIMATION - A simplified explanation of the abstract==


This abstract first appeared for US patent application 20240019156 titled 'IDENTIFYING SUITABLE MODELS FOR ADAPTIVE MODEL PREDICTIVE CONTROL OF BUILDING HVAC USING MOVING HORIZON ESTIMATION

==Simplified Explanation==

The patent application describes a model that combines a differentiable physics model of a building with a data-driven model of a heating, ventilation, and air-conditioning (HVAC) system. The model learns the behaviors of the HVAC system components based on past performance data and inputs. The model updates the parameters of both the physics model and the data-driven model using moving horizon estimation. It also determines the current input to the controlled components using model predictive control.

* The model combines a differentiable physics model of a building with a data-driven model of an HVAC system.
* The model learns the behaviors of the HVAC system components based on past performance data and inputs.
* The parameters of the physics model and the data-driven model are updated jointly using moving horizon estimation.
* The model determines the current input to the controlled components using model predictive control.

Potential applications of this technology:

* Energy management systems for buildings
* HVAC system optimization
* Smart buildings with adaptive control

Problems solved by this technology:

* Inefficient energy usage in buildings
* Lack of optimization in HVAC systems
* Difficulty in adapting HVAC systems to changing conditions

Benefits of this technology:

* Improved energy efficiency in buildings
* Cost savings through optimized HVAC system operation
* Enhanced comfort and indoor air quality in buildings


==Original Abstract Submitted==

a differentiable physics model of a building is used that defines thermodynamic relationships between zones of the building and a heating, ventilation, and air-conditioning (hvac) system. a physics-constrained, data driven model learns behaviors of controlled components of the hvac system. for each of a series of times during online operation of the hvac system, past state values are recorded representing a performance of the hvac system in the building and past inputs to the hvac system to maintain the states. the past state values and the past inputs are input into the differentiable physics model and the data driven model to: jointly update first parameters of the differentiable physics model and second parameters of the data driven model, e.g., using moving horizon estimation; and determine a current input to the controlled components, e.g., using model predictive control.



[[Category:F24F11/64]]
[[Category:F24F11/46]]
[[Category:G05B13/04]]