A DIFFERENTIAL DISCRETE PARTICLE SWARM OPTIMIZATION APPROACH TO MODEL PREDICTIVE CONTROL OF BUILDING TEMPERATURE
Abstract
A differential discrete particle swarm optimization method, aimed at designing a predictive air temperature controller for civil buildings, is presented. The predictive control allows the controller outputs to be computed in order to optimise the future behaviour for a heating/cooling plant in a building, regarding set-point tracking and control effort minimization to save energy. The controller is able to predict the building environmental conditions over a specified time interval, thus mathematical models describing the indoor building climate, as well as predicting the outside weather, are used. In this paper, the proposed design method is compared with a standard particle swarm algorithm, and preliminary experimental results are discussed.
Event details
- Event
- Instrumentation for the Information and Communication Technology Era
- Technical Committee
- TC4
- Place
- Kosice, SLOVAKIA
- Time
- 8 September 2010 - 10 September 2010
- Website
- http://www.imeko.tuke.sk