Spray Operating Parameters Optimization Based on Artificial Intelligence During Process

Abstract

During plasma spray process, many operating intrinsic parameters allow tailoring in-flight particle characteristics (temperature and velocity), thus affecting the final coating quality: torch nozzle geometry, plasma flow mass enthalpy, flow thermal conductivity, momentum density, etc. themselves resulting from the selection of operating extrinsic parameters such as the plasma torch nozzle geometry, the composition and flow rate of plasma forming gases, the arc current intensity, etc.

Beside the coupled relationships between those operating parameters making difficult a full prediction of their effects, temporal fluctuations, among them the anode wear for example, require "real time" corrections to maintain particle characteristic to targeted values. Beside, substrate temperature has to be maintain to targeted values depending upon the feedstock to be sprayed, the geometry of the part to be coated, its thermal capacity, etc.

An expert system was built to optimize and control some of the main operating extrinsic parameters. This expert system includes two parts: 1) an artificial neural network (ANN) which was built to predict an extrinsic operating window and 2) a fuzzy logic controller (FLC) to control them.

The paper detailed the general architecture of the system, discuss about its limits and typical characteristic times and display an example.