ANALYSIS OF CHEMICAL PROCESSES TOWARDS IMPROVEMENTS

Evolutionary Algorithms (EAs), which are a population based search algorithms that are inspired from nature, have proven their capability to solve several intricate problems in the field of engineering. EAs excel in solving complicated real world problems as they provide multiple of solutions compared to the traditional optimization algorithm. One important feature that differentiate between single and multi-objective optimization is the Pareto optimal front in the later, which gives the decision maker a multiple of equally good solutions. The objective of this project is to simulate and optimize complicated real world chemical engineering process problems. These problems are from various and different fields like biochemical engineering, chemical engineering and oil and gas processes using a selection of the mostly efficiently used EAs such as non-domination sorting genetic algorithm (NSGA-II) and multi-optimization differential evolution algorithm (MODE). The project also aims to develop a new version of MODE algorithm that will perform more efficiently in terms of finding better Pareto optimal front taking convergence and divergence into consideration. This developed algorithm will be validated by using it with the selected chemical processes and other well-known test problems. Other than multi-optimization, this approach will provide a deeper knowledge of the selected processes, which is difficult to gain using any other approach.