The surface integrity plays a very important rule in this functional performance, being dependent of a large number of machining parameters. The major concern of the industry is to know which combination of machining parameters provides a better surface integrity of the machined components. AISI H13 tool steel has been applied widely to produce many different types of hot working dies due to its excellent mechanical properties, such as: good resistance to thermal softening, high hardenability, high strength and high toughness. Traditionally, the surface roughness is considered to be the principal parameter to assess the surface integrity of the machined component. However, residual stress becomes an important parameter because it may increase the mould/die lifetime and their ability to withstand more severe thermal and mechanical cyclic loadings (fatigue) during its service. Therefore, significant improvements in the quality of the mould/die can be achieved with the control of the residual stresses induced during its manufacturing. This paper examines the residual stresses induced by dry turning of AISI H13 tool steel. Residual stress was evaluated experimentally in function of the tool geometry, cutting speed, feed and depth of cut. The DOE method developed by G. Taguchi was used to reduce the number of experiments. An modelling and optimization procedure based in Artificial Neural Network (ANN) and a Genetic Algorithm (GA) was developed and applied to modelling the residual stresses and to identify the optimum combination of cutting parameters, which induces low tensile or compressive residual stresses, which contributes to a better surface integrity of machined components.
