VOCAL 2024
Abstract Submission

46. Particle Filter Optimisation algorithms for robust optimisation

Invited abstract in session FB-4: Methods of optimization, stream contributed papers.

Friday, 10:15 - 11:45
Room: C105

Authors (first author is the speaker)

1. Éva Kenyeres
Department of Process Engineering, University of Pannonia
2. Alex Kummer
Department of Process Engineering, University of Pannonia
3. János Abonyi
Department of Process Engineering, University of Pannonia

Abstract

Population-based optimization algorithms gained high interest in the last few decades as they reach outstanding performance by propagating not only one but many candidate solutions in the search space. The Particle Filter Optimisation (PFO) concept is one novel representative of these techniques, which is based on the probabilistic approach of the Particle Filter state estimation algorithm.
The PFO algorithm moves weighted sample elements in the search space thus a probability distribution of the elements is evolved. This carries information about the shape of the objective function. The related literature contains the description of several variants of the PFO aiming to find the global optimum. However, the opportunity to use it for robust optimization has not been investigated yet. As in practical problems besides uncertain factors often not the global but a more stable local optimum with high performance gives the most desirable solution, our research deals with the possible robust optimization applications of the algorithm.
Our presentation will address the following points: 1) Introduction of the PFO algorithm and its crucial tunable elements. 2) Some ideas on how to use PFO for robust optimization, e.g., combined with clustering. 3) Experiment results on a benchmark function and on a practical problem from the chemical engineering field.
Results verify, that the PFO algorithm holds a promising potential for robust optimization aims due to its probabilistic nature.

Keywords

Status: accepted

Back to the list of papers