Optimization Control Of Goethite Iron Precipitation Process Based On Virtual Motion Camouflage

The goethite iron precipitation process is a key process in zinc hydrometallurgy,and its purpose is to remove the impurity iron ions in the leached liquid.In this process,the ferrous ion concentration at the outlet of the reactor is an important index to measure the effect of iron removal.Therefore,it is necessary to make the ferrous ion concentration at the outlet of reactors meet the process requirements.At the same time,the oxygen addition amount can be dynamically optimized to achieve resource saving on the basis of ensuring the iron removal effect.This paper has carried out the research on the modeling and the optimization control of the iron precipitation process.The main work and innovative achievements are reflected in the following aspects:(1)Considering the complex mechanism of the iron precipitation process,which leads to the low accuracy of the mechanism modeling method and the difficulty in data modeling due to the lack of real-time process data,a hybrid modeling method for the goethite iron precipitation process that integrates mechanism and data is proposed.Firstly,the mechanism model of iron precipitation process is established according to the reaction kinetics.Then,since the mass transfer coefficient in the mechanism model is difficult to obtain,it is established based on the process data.On this basis,combined with the mechanism model,a hybrid model of the iron precipitation process is established,and the parameters in the model can be optimized by particle swarm optimization(PSO).Finally,the proposed method is verified by simulation.The results show that the proposed method can accurately reflect the dynamic changes of the iron precipitation process.(2)In order to overcome the challenge that the optimization control problem is difficult to solve and the high computational cost due to multiple reaction couplings and many nonlinear process constraints in the iron precipitation process,a decentralized control method based on Virtual Motion Camouflage(VMC)is studied.Firstly,the optimization control problem of the iron ion concentrations at the outlet of reactor is constructed on the basis of the hybrid model of the iron precipitation process.The outlet iron ion concentrations are represented by the path control parameter through VMC,and the original optimization control problem is transformed into a nonlinear trajectory planning problem in this way.Then,the optimal solution satisfying the constraints on each collocation point is solved by Legendre pseudo-spectral,and the optimal trajectories of the ion concentrations are obtained.Through the above method,the optimal solution considering multiple constraints can be obtained on the basis of reducing the dimensions of the optimization control problem.The simulation results show that the proposed control method can track the fluctuation of outlet iron concentrations with a nice real-time performance.(3)Considering a series of disturbance factors which are unavoidable in the actual process,a distributed optimization control method based on the double-layer structure is proposed to make the system run smoothly.In this method,the upper layer builds and solves the optimization problem of the iron concentrations at the outlet of reactors by aiming at the least amount of oxygen added and the quality of the goethite slag best meeting the process requirements together.For the process oxygen utilization in this problem is difficult to measure online,a PSO-RBF neural network is used for estimation.Since the dissolved oxygen that is not exhausted in the previous stage reactor will enter the next stage reactor to participate the reaction,the oxygen addition amount of the current reactor can be reduced by designing a distribution optimization controller.Therefore,the lower layer proposes a distributed optimization control method based on VMC.Construct the distribution optimization control problem of the iron precipitation process,and use VMC to solve the oxygen addition rate.Finally,the effectiveness of the proposed method can be verified by the simulation results.Compared with the manual control and decentralized control,the oxygen addition is significantly reduced.