TOOLING SELECTION IN TECHNOLOGICAL PROCESSES USING NEURAL NETWORKS

Apply

TOOLING SELECTION IN TECHNOLOGICAL PROCESSES USING NEURAL NETWORKS

Journal Title: Archives of Mechanical Technology and Materials - Year 2015, Vol 35, Issue

Abstract

The idea of the author’s research is to develop a system aiding the design of a technological process (a CAPP system), namely a system for creation of a technological process plan, in which the sequence of technological operations is defined and for each operation in the technological process, the appropriate machine, tools, tooling and machining parameters are selected. The article discusses accessory selection in technological processes using neural networks. Tooling selection is a necessary stage in the design of technological processes if a tool that has been selected does not fit the machine. Tooling selection models were prepared using unidirectional multilayer neural networks with back propagation of error (MLP) and a self-organizing Kohonen network. Two completely different neural networks were selected for the selection of the tooling. MLP network represents a network with learning supervision, and network Kohonen network learning without supervision. The training data for the neural networks was prepared at a manufacturing company. The neural networks were made using the Statsoft STATISTICA Data Miner software.

Authors and Affiliations

Izabela Rojek

Keywords

Tooling (accessory) Technological process Neural networks

EP ID EP198799
DOI -
Views 48
Downloads 0