On the Implementation of Neural Network Concept to Optimize Thermal Spray Deposition Process
- PDF / 47,868 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 38 Downloads / 158 Views
On the Implementation of Neural Network Concept to Optimize Thermal Spray Deposition Process
Sofiane Guessasma, Ghislain Montavon, Christian Coddet LERMPS, Université de Technologie de Belfort-Montbéliard 90 010 Belfort Cedex, France ABSTRACT Numerous processing parameters, up to fifty, characterize the plasma spray deposition process. A better quality control of the resulting deposits induces a better understanding of their effects on coating formation mechanisms. Numerical models can help to provide such an understanding. From a mathematical point of view, d.c. plasma spray deposition process is assimilated to a nonlinear problem in regards to its variables (operating parameters, environment, etc.). This paper develops a global approach based on an implicit describing of the mechanisms implementing Artificial Neural Networks (ANNs). The global concept and the protocols to implement are presented and developed for an example related to d.c. plasma spray process. INTRODUCTION Thermal spraying is a generic term describing several covering "dry" surface treatment processes. They are widely used for industrial applications in order to confer to components of a mechanical system specific in-service properties permitting to specifically adapt their resistance against various solicitation modes (i.e., wear, thermal insulation, corrosion, etc.) [1]. The basic principle of this technique can be easily described [2]. A feedstock material, usually powder particles of a given particle size distribution are injected into a thermal jet exiting a high velocity from a nozzle. Hence, the jet simultaneously heats until the melting temperature, at least, and accelerates, the individual particles, which form a particle stream. The component to be covered is placed in front of this stream and the individual particles impact, flatten and solidify to form individual thin lamellae. The deposit of matter results from the stacking of those lamellae obtained by the successive relative passes of the stream in front of the component. The nature of the jet permits to distinguish among the several techniques. The jet can result: • either from an enthalpy of reaction provided by a combustion reaction, the processes consisting in this case in the flame spraying (case of a diffuse flame) or the High Velocity Oxy-Fuel spraying (case of a pre-mixing flame) or the D-Gun spraying (case of a detonation flame), • or from an energy provided by an electric arc, the processes consisting in this case in plasma spraying (a plasma is generated by molecular dissociation and atomic ionization of a gas mixture and ejected through a nozzle), or in electric arc spraying (the arc melts the tips of two electrodes that are atomized by a gas flow to form the particle stream). The important level of "global" energy in the system induces extremely short and transient interactions between the feedstock material and the jet (characteristic duration varying from 0.3 ms for the average particle / jet interaction duration to 50 µs for the average lamella solidification durat
Data Loading...
