An effective soft computing technology based on belief-rule-base and particle swarm optimization for tipping paper permeability measurement

This paper proposes a soft computing technology based on belief rule base (BRB) system for the tipping paper permeability measurement in tobacco factory. In current studies about BRB, both the referential values of the antecedent attributes and the utilities of the consequents are given in advance and are not trained by using the dedicated optimal algorithms. The limitations of expert knowledge may lead to error of BRB because both the referential values and the utilities make a real difference on the structure of BRB, and the appropriate structure is helpful for tuning parameters more accurately. Therefore, this paper focuses on the structure and parameters optimization of BRB (SPO-BRB) by taking the referential values of the antecedent attributes and the utilities of the consequents into account to improve the input–output modeling ability of BRB. However, SPO-BRB is a nonlinear nonconvex optimization problem (NNOP). To deal with the NNOP of SPO-BRB, a particle swarm optimization algorithm with improved velocity update way and repair methods (PSO_VR) is proposed. A case study based on the data collected from a tobacco factory of china is carried out. The test results demonstrate the functionality of SPO-BRB and the effectiveness of PSO_VR.