Stochastic optimization for real-time operation of alumina blending process

In this paper, the stochastic optimization blending operation is applied to the alumina production in this paper. A new binomial distribution based stochastic scenario optimization used together with the sample selection approach is utilized to design the optimal set point for control, under which the probability of quality indices of the raw slurry being within the tolerance region is high enough in the presence of uncertainties caused by fluctuation of the raw material and disturbances. Through practical industrial experiments, it is observed that the proposed stochastic optimization method is effective and the computational cost is low.