 |
    |
Run-Time Parameter Selection and Tuning for Energy Optimization Algorithms
Ingo Mauser1, Marita Dorscheid2, and Hartmut Schmeck2
1FZI Research Center for Information Technology 76131, Karlsruhe, Germany
mauser@fzi.de
2Karlsruhe Institute of Technology – Institute AIFB 76128, Karlsruhe, Germany
marita.dorscheid@partner.kit.edu
schmeck@kit.edu
Abstract. Energy Management Systems (EMS) promise a great potential to enable the sustainable and efficient integration of distributed energy generation from renewable sources by optimization of energy flows. In this paper, we present a run-time selection and meta-evolutionary parameter tuning component for optimization algorithms in EMS and an approach for the distributed application of this component. These have been applied to an existing EMS, which uses an Evolutionary Algorithm. Evaluations of the component in realistic scenarios show reduced run-times with similar or even improved solution quality, while the distributed application reduces the risk of over-confidence and over-tuning.
LNCS 8672, p. 80 ff.
lncs@springer.com
© Springer International Publishing Switzerland 2014