Stochastic Optimization Model and Solution Algorithm for Blending Procedure of Process Industrial

Considering the uncertain quality parameters of raw materials in industrial blending processes, we propose a stochastic blending optimization model, which takes the cost of the raw materials as the optimization objective and incorporates uncertainty parameters into the quality constraints as random variables. Then, in order to overcome the shortcomings of the Monte Carlo sampling technique, we apply the more efficient Hammersley sequence sampling (HSS) technique to obtain an expectation optimization model that corresponds with the stochastic model. We use the HSS technique in the crossover and mutation steps of the genetic algorithm to maintain uniformity of the population and to effectively solve the stochastic blending optimization problem. The results of our industrial experiment show that the proposed method not only greatly reduces the consumption cost of raw materials, but also guarantees the quality of the blending product. This method has robustness and yields a good stochastic optimization mode for blending processes.