Neural network model for predicting ferrite number in stainless steel welds

Abstract

Prcdicting the fcrrite content in stainlcss stccl welds is desirable in order to assess an alloy's susceptibility to liot cracking and to estimate the as-welded propertics. Scveriil methods liave been uscd over the years to estimate the ferrite content as a function of the alloy composition. A new technique is dcscribcd which uses a neural network analysis to deterinine the ferrite numbcr. The network was trained on tile same data set that was used to gcnente tlie WRC-1992 constitution diagram. Tlie accuracy of the ncural network predictions is coir pared to that for the WRC-1992 diagram as well as another recently proposed method. It was found that the neural network model was approximately 20% more accurate than either D f the other two methods. In addition, it is suggestcd that furthcr improvements to the ncural network model. including the consideration of process variablcs. can be made which can lcatl to even better accuracy.

Independence of EPR and PAP tests performed on resistance spot welding joints

“…The thermal energy associated with the RSW process gives rise to the formation of two new zones from the parent metal (Figure 2): (i) the weld nugget [37][38][39], formed from the solidification of the molten metal [40], with a cast dendritic microstructure of austenite with a high content of δ-ferrite in interdendritic regions as a consequence of the rapid cooling [14,[41][42][43] (Figure 3); and (ii) the heat-affected zone (HAZ), adjacent to the weld nugget, where the sensitisation phenomenon may occur.…”

Physicochemical analysis and the phenomenological model of secondary crystallization of a metal in the course of welding

“…As a consequence, the concentration of the i th carbide-forming element in the carbide phase can be written as (5) Here, w j is the fraction of carbon in the j th carbide phase with respect to the total amount of carbon involved in the formation of the carbide of the i th alloying element, is the carbon concentration in the products of austenite decomposition [given by expression (3)], and η i is the coefficient determined by the expression derived earlier in [38]; that is,…”

“…The kinetic analysis, as a rule, is carried out using the diagrams of both the isothermal and nonisothermal decompositions of austenite [1][2][3]. In recent years, the phase composition of a weld metal formed upon isothermal decomposition of austenite has been often predicted using numerical methods [4][5][6][7], which were originally developed to solve the equations describing the exponential and power laws for the nucleation and growth of new phases [8][9][10] and were then adapted to solid-state reactions. However, these methods cannot be applied to the conditions of nonequilibrium decomposition of alloyed austenite, even though it is these conditions that are characteristic of hardfacing processes and welding of special steels and alloys.…”