Dynamic balancing of the cleaning unit used in agricultural thresher using a non-dominated sorting Jaya algorithm

Abstract

Purpose This paper aims to propose a dynamically balanced mechanism for cleaning unit used in agricultural thresher machine using a dynamically equivalent system of point masses. Design/methodology/approach The cleaning unit works on crank-rocker Grashof mechanism. This mechanism can be balanced by optimizing the inertial properties of each link. These properties are defined by the dynamic equivalent system of point masses. Parameters of these point masses define the shaking forces and moments. Hence, the multi-objective optimization problem with minimization of shaking forces and shaking moments is formulated by considering the point mass parameters as the design variables. The formulated optimization problem is solved using a posteriori approach-based algorithm i.e. the non-dominated sorting Jaya algorithm (NSJAYA) and a priori approach-based algorithms i.e. Jaya algorithm and genetic algorithm (GA) under suitable design constraints. Findings The mass, center of mass and inertias of each link are calculated using optimum design variables. These optimum parameters improve the dynamic performance of the cleaning unit. The optimal Pareto set for the balancing problem is measured and outlined in this paper. The designer can choose any solution from the set and balance any real planar mechanism. Originality/value The efficiency of the proposed approach is tested through the existing cleaning mechanism of the thresher machine. It is found that the NSJAYA is computationally more efficient than the GA and Jaya algorithm. ADAMS software is used for the simulation of the mechanism.

Research on Enterprise Supply Chain Anti-Disturbance Management Based on Improved Particle Swarm Optimization Algorithm

“…Singh et al adopted non-dominated ranking index to verify the optimization algorithm of multi-objective problem in the dynamic balance mechanism of cleaning device of agricultural thresher. The results show that the proposed method is feasible, that is, non-dominated ranking index is universal for the detection of multi-objective optimization algorithm [21]. In the study of the basic method of non-orthogonal multiple access which has been envisaged as the fifth generation cellular network, Kumaresan's team used the fitness value of each iteration of the algorithm to evaluate the particles in the validation of the adaptive user clustering algorithm based on the particle swarm optimization algorithm.…”