Grain Temperature-Dependent Thermal Properties Estimation Using FI-QPSO Algorithm

Abstract

A hybrid fuzzy inference-quantum particle swarm optimization (FI-QPSO) algorithm is developed to estimate the temperature-dependent thermal properties of the grain. The fuzzy inference scheme is established to determine the contraction-expansion coefficient according to the aggregation degree of particles. The heat transfer process in the grain bulk is solved using the finite element method, and the estimation task is formulated as an inverse problem. Numerical experiments are performed to study the effects of the surface heat flux, measurement errors, and the individual space on the estimation results. Comparison with the quantum particle swarm optimization (QPSO) algorithm and conjugate gradient method (CGM) is also conducted, and it shows the validity of the estimation method established in this paper.