The CutMAG as a New Hybrid Method for Multi-edge Grinder Design Optimisation

This article is a part of the series dedicated to AI Methods Inspired by Nature and their implementation in the mechatronic systems. The CutMAG algorithm uses hybrid approach to optimisation, i.e. a combination of classic genetic algorithms (GA) with morphologic optimisation (M) thus creating innovative approach to optimisation of cutting disk design (Cut) for the multi-edge grinder. The input data include population of individuals. Each individual is represented by a set of cutting disks. Whereas the fitness function was assumed as a combination of several postulates of the mechanical design foundations. The method includes mechanical, design and energy aspects. Each individual constitutes a complete solution of the disk set whereas the population represents the entire class of solutions. The fitness function of an individual is calculated as the average fitness of each disk supplemented by information describing the relationship between both adjacent disks. The method for calculation of function values was selected so as to ensure its maximisation in the process of evolution. Although promising results of the genetic algorithms operation were achieved, one can consider further improvement of the method efficiency. The authors used morphological operations in order to better adopt the method to the task.