New method based on neuro-fuzzy system and PSO algorithm for estimating phase equilibria properties

Abstract

The subject of this work is to propose a new method based on ANFIS system and\n PSO algorithm to conceive a model for estimating the solubility of solid\n drugs in sc-CO2. The high nonlinear process was modeled by neuro-fuzzy\n approach (NFS). The PSO algorithm was used for two purposes: replacing the\n standard back propagation in training the NFS and optimizing the process.\n The validation strategy have been carried out using a linear regression\n analysis of the predicted versus experimental outputs. The ANFIS approach is\n compared to the ANN in terms of accuracy. Statistical analysis of the\n predictability of the optimized model trained with PSO algorithm (ANFIS-PSO)\n shows very good agreement with reference data than ANN method. Furthermore,\n the comparison in terms of AARD deviation (%) between the predicted results,\n results predicted by density-based models and a set of equations of state\n demonstrates that the ANFIS-PSO model correlates far better the solubility\n of the solid drugs in scCO2.A control strategy was also developed for the\n first time in the field of phase equilibrium by using the neuro fuzzy\n inverse approach (ANFISi) to estimate pure component properties from the\n solubility data without passing through GCM methods.