Design of a fuzzy logic approach based on genetic algorithms for robust plasma-sprayed zirconia depositions

Abstract This paper proposes a design for an adaptive system by modeling the relationship between coating surface roughnesses and the controling factors in plasma spray coating processes. A statistical design was used to obtain sufficient experimental information with the least number of experiments. Analysis of variance was then used to select significant control factors for reinforced coatings, and these identified factors used to construct an adaptive fuzzy logic control model. In order to model the process, a fuzzy logic controller (FLC) was utilized. A genetic algorithm (GA) was applied as a tool to optimize rule bases from traditional FLCs. Therefore, with the use of a GA-optimized FLC, robust reinforced deposition for coatings in the plasma spraying process can be obtained. The experimental results show that the obtained optimal rule base for FLC is capable of achieving the desired results. That is to say, the proposed design, which combines a statistical method and a GA-optimized FLC, is efficient and robust for the investigation of reinforced coatings in a plasma spraying process.