Blending Scheduling Under Uncertainty Based on Particle Swarm Optimization with Hypothesis Test
Blending is an important unit operation in process industry. As a nonlinear optimization problem with constraints, it is difficult to obtain optimal solution for blending scheduling, especially under uncertainty. As a novel evolutionary computing technique, particle swarm optimization (PSO) has powerful ability to solve nonlinear optimization problems with both continuous and discrete variables. In this paper, the performance of PSO under uncertainty for blending scheduling problem is investigated, and a new hybrid approach (namely PSOHT) that combines PSO and hypothesis test (HT) is proposed. The simulation results based on an example of gasoline blending problem show that the proposed PSOHT algorithm is valid and effective for solving problem under uncertainty.
KeywordsParticle Swarm Optimization Penalty Function Particle Swarm Optimization Algorithm Octane Number Nonlinear Optimization Problem
Unable to display preview. Download preview PDF.
- 3.Litvinenko, V.I., Burgher, J.A., Vyshemirskij, V.S., Sokolova, N.A.: Application of Genetic Algorithm for Optimization Gasoline Fractions Blending Compounding. In: Proceedings of The IEEE International Conference on Artificial Intelligences Systems (ICAIS 2002), Divnomorskoe, Russia (2002)Google Scholar
- 4.Zhao, X.Q., Rong, G.: Blending Scheduling under Uncertainty Based on Particle Swarm Optimization Algorithm. Chinese J. Chem. Eng. 13(4), 535–541 (2005)Google Scholar
- 5.Liu, B., Wang, L., Jin, Y.H., Huang, D.X.: Advances in Particle Swarm Optimization Algorithm. Control Instrum. Chem. Ind. 32(3), 1–6 (2005)Google Scholar