Optimization of Betulinic Acid Ester Enzymatic Synthesis by Artificial Intelligence

The reaction conditions of the enzymatic synthesis of betulinic acid ester are a practical and vital reaction. Its results have been previously modeled by artificial intelligence. In this studyr, mentioned reaction has been not only simulated but also optimized by multi-objective meta-heuristic algorithms. The Multi-Layer Perceptron (MLP) is a type of feed forward artificial neural network (ANN) which its performance improves by reduction of train and test data errors. It depends on the trained method and the number of neurons in the hidden layer. In an appropriate ANN, errors for train data and test data have to be closed. Radial Basis Function (RBF) hasn't been utilized as ANN already. The RBF consists of an extraordinary advantage: it determines the number of neurons due to desired error and its parameters have been set by advanced particle swarm optimization (PSO) algorithm. Further, PSO's parameters including c۱, c۲, and ω are determined as fuzzy. Finally, the results of the proposed method will be compared with those of previous methods.

کلیدواژه ها:

train data ، test data ، error minimization ، goal ، spread