SIMULATION OF TECHNOLOGICAL PROCESS BY USAGE NEURAL NETWORKS AND FACTORIAL DESIGN OF EXPERIMENTS

Abstract

The possibilities of simulation of technological process of aluminium anodic oxidation using the methodology of Design of Experiments (DOE) and theory of neural networks in order to monitor the anodizing process under various operating conditions are presented in this paper. The influence of chemical and physical input factors on the resulting AAO (anodic aluminium oxide) layer thickness at applied current density of 1 A•dm-2 and 6 A•dm-2 has been investigated. Based on the evaluation of experimentally obtained data, the computational predictive model describing the effect of individual input factors and their mutual interactions on the AAO layer thickness was developed in the form of cubic function. This model indicates which factors are important and how they combine to influence the response, it will enable us to optimize operating conditions. The most significant benefit of our research work in this field is the fact that all relevant factors were varied simultaneously.

Recommended articles

AUTOMATIZATION PROCESS OF DESIGNING THE TECHNOLOGICAL DOCUMENTATION BY TOOLS OF COMPREHENSIVE CAD-CAM-CAE SYSTEM

MARTIN POLLAK, PETR BARON, JOZEF ZAJAC
Keywords: generation | process documentation | NC program | automated processing

DESIGN OF ADJUSTABLE SMART VISION SYSTEM BASED ON ARTIFICIAL MUSCLE ACTUATORS

KAMIL ZIDEK, JAN PITEL, ALEXANDER HOSOVSKY
Keywords: automation | Segmentation | artificial muscles | image processing

MONITORING OF THE MACHINE MILKING PROCESS WITH APPLICATION OF INFRARED THERMOGRAPHY

PETR JUNGA, PETR TRAVNICEK, JURAJ RUZBARSKY
Keywords: surface temperature | machine milking | infrared thermography | monitoring