Hybrid Neural Network and Genetic Algorithm method to Optimize Wastewater Treatment Process

摘要

Soft computing methodologies can handle the complexity of modeling and optimizing natural processes. Combination of these methods results in hybrid approaches that contain advantages of each one. In this paper, artificial neural network and genetic algorithm approaches are used to model and optimize the process of dye removal, which is an important part of waste water treatment. An experimental data set is used to approximate the relation between initial dye concentration, adsorbent, pH, and contact time as inputs and dye removal percentage as output through artificial neural network. Genetic algorithm approach is employed to suggest the best combination of input elements to maximizing dye removal for each initial dye concentration produced by factory. Results show that proposed input combination leads to 92% average dye removal with less economic cost