Optimization of plasma deposition manufacturing parameters using a hybrid ANN-GAs method

Summary form only given, as follows. Plasma surfacing is an important enabling technology in high performance coating applications. Recently, it is being applied to rapid prototyping/tooling to reduce development time and manufacturing cost for the development of new product. In this technology, a plasma arc beam is used as thermal energy source, metal powders are preheated in plasma arc and deposited in melt pool on the base plate or deposited layer synchronously, with the movement of plasma gun and/or worktable controlled by CNC according to CAD slice model in computer, deposition layer grows gradually along z direction until the total part is fabricated. However, this technology is in its infancy, it is essential to understand clearly how process variables relate to deposit microstructure and properties for plasma deposition manufacturing (PDM) process control. In this article, the microstructure, mechanical properties and coating appearance for single surfacing and multiple surfacing under different parameters, such as plasma power, powder feedrate and scanning speed, etc., were studied in detail, and then a hybrid BP artificial neural network (ANN) and genetic algorithms (GAs) method which can overcome the shortcoming of BP neural network and obtain a global convergence result, was presented to optimize processing parameters. Using the optimized parameters, metal parts with complex shape and excellent mechanical properties and microstructure were fabricated successfully.