Comparison between Artificial Neural Network and Rigorous Mathematical Model in Simulation of Industrial Heavy Naphtha Reforming Process

In this study, an artificial neural network (ANN) model was developed and compared with a rigorous mathematical (RMM) to estimate the performance of industrial heavy naphtha reforming process. The ANN model, represented by multilayer feed forward (MFFNN), had (36-10-10-10-34) topology, while RMM involved solving 34 ordinary differential equations (ODEs) (32 mass balance, 1 heat balance momentum balance) predict compositions, temperature, pressure distributions within All computations predictions were performed using MATLAB® software version 2015a. topology minimum MSE when number hidden layers, neurons in layer, training epochs 3, 10, 100,000, respectively. Extensive error analysis between experimental data predicted values conducted following functions: coefficient determination (R2), mean absolute (MAE), relative (MRE), square (MSE). results revealed that (R2 = 0.9403, MAE 0.0062) simulated process slightly better than 0.9318, 0.007). Moreover, computational time obviously reduced from 120 s for 18.3 ANN. However, one disadvantage is it cannot be used internal points reactors, temperatures, pressures weight fractions very well.