Stochastic Operational Optimization for Metallurgical Blending Process Under Uncertainty

This paper presents a stochastic operational optimization method for metallurgical blending processes by focusing on the uncertainty of composition of raw materials, where the stochastic optimization model is firstly formulated by explicitly incorporating uncertain parameters into the constraints as random variables. Then an efficient sampling technique is utilized to construct the expectation counterpart of the stochastic model, in which Monte Carlo sampling is commonly employed. Finally, an HSS stochastic genetic algorithm is proposed to solving the stochastic optimization problem, where HSS technique is used the initial population generation and population updating so as to improve the population diversity and uniformity of random operations. The results show that the proposed method can guarantee the quality of blending products as well as greatly reduce the consuming cost of raw materials under uncertainty.