부산대학교 도서관

Iron precipitation is a key process in zinc hydrometallurgy. The process consists of a series of continuousreactors arranged in descending order, overflowing zinc leach solution from one reactor to the next. In this paper,according to the law of mass conservation and the reaction kinetics, a continuously stirred tank reactor model ofa single reactor is first established. Then, a distributed model of cascade reactors is built with coupled controlbased on the single reactor model, considering the unreacted oxygen in leaching solution. Secondly, four reactorsin the iron precipitation process are considered as four subsystems, the optimization control problem of the processis solved by a distributed model predictive control strategy. Moreover, the control information feedback betweensuccessive subsystems is used to solve the optimization problem of each subsystem, because of the existing controlcoupling in their optimization objective function of pre and post subsystems. Next, considering the intractability ofthe optimization problem for subsystems with various constraints, a distributed dual iterative algorithm is proposedto simplify the calculation. With the consideration of its cascade structure and control couplings, the proposedalgorithm iteratively solves the primal problem and the dual problem of each subsystem. The application case showsthat distributed model predictive control based on dual iteration algorithm can handle coupled control effectivelyand reduce the oxygen consumption.