On-line Yield Optimization Utilizing Oscillatory Autocatalytic Reaction in CSTR by Using A Fast Searching Method

摘要

In this paper, we solve an optimal control problem to get a high yield of a product by using the characteristics of oscillatory autocatalytic reaction in continuous stirred-tank reactor (CSTR). First, we formulate a problem of maximizing the yield on a finite horizon at each present time. Next, we propose a new method to get a set of optimal input moves in real-time at each sampling step by a searching method which utilizes random inputs. The effectiveness of the proposed method is shown through some simulations.