RESUMO
OBJECTIVE: To optimize the preparation process of Erigeron breviscapus sustained-release pellets. METHODS: A mathematical model of relationship between the independent variables and dependent variable of the preparation process of Erigeron breviscapus sustained-release pellets was established by using back-propagation (BP) artificial neural networks (ANN), and the preparation process parameters were optimized with particle swarm optimization (PSO) algorithm. RESULTS: The pellets prepared according to the optimized preparation process parameters had significant effect of sustained-releasing. Drug release from the pellets was controlled by both diffusion and matrix corrosion. CONCLUSION: Combining BP ANN modeling with PSO algorithm provides an effective way to solve the multi-dimensional optimization problem of complicated nonlinear systems in pharmaceutical technology.