A Novel Hybrid Approach for Modeling and Optimisation of Phosphoric Acid Production through the Integration of AspenTech, SciLab Unit Operation, Artificial Neural Networks and Genetic Algorithm

The purpose of the study was to identify and predict optimized parameters for phosphoric acid production. This involved modeling crystal reactor, UCEGO filter (as a detailed model is not available in Aspen Plus or other simulation software), separator using Sci-Lab develop Cape-Open models. conducted analyzing 10 different phosphates with varying qualities fractions P2O5 minerals. After successful simulation, sensitivity analysis by such as capacity, speed, vacuum, particle size, water temperature washing filtration cake, flow recycled strong from below filter, slurry reactor 1, reactors, H2SO4, resulting nearly one million combinations. To create an algorithm predicting process maximal extent recovering H3PO4 slurry, ANN models were developed determination coefficient 99%. Multi-objective optimization then performed genetic find most suitable that would lead higher reaction degree (96–97%) quantity separated lower losses gypsum. results indicated it possible influence on quality produced minimize during confirmed be viable when compared found literature.