A model-assisted adaptive controller fine-tuning methodology for efficient energy use in buildings
Authors:
- 1 June 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 54, 49-54
Abstract
Building Energy Management Systems are finding widespread use for the holistic control of all energy-influencing elements of buildings and are responsible for ensuring an effective and parsimonious energy use. In most cases, fixed logic controllers are deployed in the building to implement predetermined strategies. Good performance can not be guar anteed due to inherent uncertainties that can not be a priori ascertained, such as weather variations, occupant actions, and changes in the building state and characteristics. In this paper, a model-assisted tuning methodology is presented to adaptively and automatically fine-tune relevant controller parameters. In our approach, at the end of each day of the building operation, given "reasonable" predictions for the following day, and using an accurate thermal-simulation model to evaluate performance, a new set of controller parameters is generated to be used the following day. This way, good performance can be achieved using controllers with simple mathematical structure.Keywords
This publication has 23 references indexed in Scilit:
- Large Scale Nonlinear Control System Fine-Tuning Through LearningIEEE Transactions on Neural Networks, 2009
- An adaptive optimization scheme with satisfactory transient performanceAutomatica, 2009
- Adaptive Fine-Tuning of Nonlinear Control Systems With Application to the Urban Traffic Control Strategy TUCIEEE Transactions on Control Systems Technology, 2007
- Intelligent building energy management system using rule setsBuilding and Environment, 2007
- Reinforcement learning for energy conservation and comfort in buildingsBuilding and Environment, 2007
- Quasi-adaptive fuzzy heating control of solar buildingsBuilding and Environment, 2006
- Optimization of building thermal design and control by multi-criterion genetic algorithmEnergy and Buildings, 2002
- The Nature of Statistical Learning Theory2000
- Indoor air-quality control by a fuzzy-reasoning machine in naturally ventilated buildingsApplied Energy, 1996
- A training algorithm for optimal margin classifiers1992