A novel ensemble [formula omitted]-infinity proposal density based particle filtering to improve flux estimation in nonlinear point kinetic model

Abstract

This paper describes a novel Particle Filter algorithm that is both efficient and accurate in terms of computing and employs Ensemble H∞ Filter as the proposed density function to tackles the issue of degeneracy. The problem is considered for flux measurement in a Pressurized Water Reactor described by non-linear point kinetic model. A proposal density based on Ensemble H∞ is considered which uses empirical sample mean and variance to generate samples. The proposed algorithm is numerically less computationally intensive than the techniques currently available in the literature. The strength and suitability of proposed algorithm is compared with existing techniques in a variety of settings.