Comparison of Neural Capabilities of Hammerstein, Neural-Wiener, Hammerstein-Wiener Model In Modeling Limiting Current of The Lanthanum -Electro Deposition

Abstract

Lanthanum can be obtained by electrowinning process. Electrodeposition is a metal deposition process that uses DC current electrolysis. The presence of a magnetic field affects the growth of fractals or the morphology of the resulting deposit. The presence of a magnetic field leads to an increase in limiting current and uniform growth, as well as a more uniform metal deposition surface. Electrodeposition has the main problem, namely the roughness of the resulting layer (non-uniform crystal growth). The MED process will generate a limiting value (ib) to obtain the limiting value (ib) required of compounds and tools such as electrode area (A), electron-active concentration (C), kinematic electrolyte viscosity (V), diffusion coefficient (D), magnetic field strength (B), and the number of electrons involved in the MED process (n). Magneto-electrodeposition research tends to require expensive compounds. Therefore, there is a need for a solution to reduce the cost of expensive compounds from the magnetoelectrodepotion (MED) process in lanthanum. One solution offered through mathematical methods is to use the Neural Hammerstein Model modeling, Neural Wiener Model, and Hammerstein Wiener Model to guess the value of the best limiting current (iB) by comparing Mean Square Error (MSE), Mean Absolute Percentage Error (MAPE), Mean Absolute Error (MAE). The results and analysis of this study show that the Hammerstein Wiener Model has the smallest error value and can accurately estimate nonlinear processes and outperform other blocks considered.