Neural daylight control system
Horatiu Stefan Grif

TL;DR
This paper presents a neural control system for automatic daylight management that uses an online-trained neural network to identify the inverse model of the process, ensuring consistent illuminance despite variable daylight conditions.
Contribution
It introduces a neural controller that employs an online neural network to identify the inverse process model for improved daylight control.
Findings
Neural controller maintains constant illuminance levels.
Online neural network effectively models inverse process.
System adapts to variable daylight conditions.
Abstract
The paper describes the design, the implementation of a neural controller used in an automatic daylight control system. The automatic lighting control system (ALCS) attempt to maintain constant the illuminance at the desired level on working plane even if the daylight contribution is variable. Therefore, the daylight will represent the perturbation signal for the ALCS. The mathematical model of process is unknown. The applied structure of control need the inverse model of process. For this purpose it was used other artificial neural network (ANN) which identify the inverse model of process in an on-line manner. In fact, this ANN identify the inverse model of process + the perturbation signal. In this way the learning signal for neural controller has a better accuracy for the present application.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsBuilding Energy and Comfort Optimization
