Solution of structural mechanic’s problems by machine learning

This article proposes analysis procedure of structural mechanic’s problem as integral formulation. The analysis procedure was proposed as stressed-based analysis procedure as before plying the procedure, it is required to define stress distribution within the structural body by proper modelling and structural idealization assumptions. The methodology can suitable be applied for finding the solution of engineering applications with required accuracy. The methodology exploits the unfolded part of the structural analysis problems which were not so easy to solve such as geometric and material nonlinearity together with simple integration technique [11]. It has already unfolded the misery of physically exploiting plastic behaviour structures before the start of fracture of elastic materials [13]. The formulation is integral formulation rather than differential formulation in which whole stress –strain behaviour is utilised in the analysis procedure by using neural network as regression tool. In this article, one dimensional problem of uniaxial bar and plane strain axis symmetric problem of cylinder subjected to internal pressure is solved. The results are compared with the classical differential formulation or linear theory.