A Novel Stochastic Framework for the MHD Generator in Ocean

Abstract

This work aims to study the nonlinear ordinary differential equations (ODEs) system of magnetohydrodynamic (MHD) past over an inclined plate using Levenberg-Marquardt backpropagation neural networks (LMBNNs). The stochastic procedures LMBNNs are provided with three categories of sample statistics, testing, training, and verification. The nonlinear MHD system past over an inclined plate is divided into three profiles, dimensionless momentum, species (salinity), and energy (heat) conservations. The data is applied 15%, 10%, and 75% for validation, testing, and training to solve the nonlinear system of MHD past over an inclined plate. A reference data set is designed to compare the obtained and proposed solutions for the MHD system. The plots of the absolute error (AE) are provided to check the accuracy and precision of the considered nonlinear system of MHD. The obtained numerical solutions of the nonlinear magnetohydrodynamic system have been considered to reduce the mean square error (MSE). For the capability, dependability, and aptitude of the stochastic LMBNNs procedure, the numerical performances are provided to authenticate the relative arrangements of MSE, error histograms (EHs), state transitions (STs), correlation, and regression.