Dynamic Modeling Of Iron Precipitation By Goethite Process Based On Multiple Characteristic Time Points

:The iron precipitation is a key process of hydrometallurgy, and a complex production process involving oxidation, hydrolysis and neutralization reaction. There are strong nonlinearity and coupling between state variables. At the same time, there is time-delay in the iron precipitation which is decided by goethite characteristics, and it is difficult to model for this process, resulting in the production process difficult to control, poor stability, high energy and material consumption, and low recovery of valuable metals. Therefore, researching the mechanism of iron precipitation by goethite process and establishing a dynamic mathematical model, which is better to reflect the reality of iron precipitation production, is great practical significance of optimal control for the process.The mechanism of iron precipitation by goethite process was analyzed and the kinetic mechanism of the process was studied, involving oxidation, hydrolysis and neutralization reaction, geting the related chemical reaction rate expression. The characteristics of three phase reaction and the influence of the pH values, amount of oxidation and zinc calcine, temperature, reaction time, seed crystal and stirring intensity on the seed to the precipitation of iron were analyzed, identified the main factors affecting the process efficency and quality; the different reactors in series of state time delay and the coupling relationship between the same reactor were also considered. Then, according to the law of conservation of materials and the characteristics of the CSTR model, the amount of oxygen intake and calcine and CSTR reactor inlet total flow were taken as input variables, the concentration of ferrous ion and ferric iron ion and pH value as state variables, the input and output nonlinear multiple time-delay dynamic mathematical model of iron precipitation process were established.Considering the presence of time-delay and kinetic parameters in the model were unknown, the time-delay system identification method based on multiple characteristic time points was used to solve dynamic mathematical model gradients on unknown parameters, and was combined with gradient-based optimization of sequential quadratic programming to identify the unknown time-delay and kinetic parameters. Field production data simulation results validated the validity and reliability of the dynamic model established in this paper. This paper contained23figures,5tables and79references.