Research On Hybrid Modeling And Optimization Control For Hydrometallurgy Leaching Process

Mineral resources play an extremely important role in economic development and social life.China’s mineral resources are abundant,but the population base is large and the per capita resources are small.In order to respond to the national sustainable development strategy and ensure sustained and effective economic growth,it is necessary to increase the effective utilization of mineral resources.Hydrometallurgy is currently the main metal smelting extraction technology.It is a metallurgical process that uses a solvent to chemically react with mineral elements in minerals to extract and separate useful metals from minerals.The remarkable advantages of hydrometallurgy are high metal recovery rate,high process flexibility and simple equipment.Compared with the production process of pyrometallurgy,it is easier to achieve continuous and automated,and more adaptable to the requirements of sustainable resource development.A wide range of applications.The leaching process is one of the important production processes of hydrometallurgy.The leaching process directly affects the quality of the subsequent production process and determines the economic benefits of the enterprise.Although China’s hydrometallurgical leaching process has reached the world’s advanced level,the actual leaching production process control is still in the level of manual control,the experience adjustment,the degree of automation is not high,and resulting in low utilization of ore resources,low production efficiency,material consumption,and enterprise profits is low.Aiming at this problem,this paper systematically studied the mechanism modeling,hybrid modeling theory of leaching rate in the leaching process and application problems based on the hydrometallurgical production process of a gold smelter.Using the serial hybrid modeling method combining mechanism modeling and data modeling,comprehensively systematically carried out the research on the leaching rate prediction of hydrometallurgical leaching process.Based on the established hybrid model,the optimal control of the actual leaching process is studied.The main work of this paper includes:(1)Based on the reaction mechanism of the hydrometallurgical leaching process,a dynamic mechanism model of the leaching rate of the leaching process was established.The model consists of the solid-phase gold conservation equation,the liquid-phase gold conservation equation and the cyanide conservation equation in the liquid phase and the kinetics of gold and cyanide ions.Learn the composition of the reaction rate model.Through simulation,the influence of each input variable on the leaching rate in the model is analyzed,which lays a foundation for the subsequent optimization control research.(2)A hybrid modeling method consisting of a mechanism model and a dynamic reaction velocity estimation model is adopted.The mechanism model describes the known parameters of the process.The RBF neural network data model solves the dynamic reaction velocity that is difficult to measure in the actual process.Since the input and output of the data model are difficult to measure,the Tikhonov regularization method is used to estimate the dynamic reaction speed.This method can effectively avoid the influence of measurement noise on the calculation process.In order to make the model more accurate prediction and generalization ability,the model is corrected based on the deviation between the actual value and the predicted value.The generalization ability and accuracy of the hybrid model established by simulation analysis.(3)In depth analysis of the actual leaching process requirements,the establishment of a leaching process optimization model with economic benefits as the target,sodium cyanide addition as decision variables,respectively,using particle swarm optimization algorithm and sequential quadratic programming method to solve the model,to obtain steady state optimization aims.The simulation results show that the improved algorithm has better optimization performance.(4)Aiming at the real-time optimization problem of hydrometallurgical gold cyanide leaching process,a control method combining self-optimization and nonlinear model prediction is proposed.Based on the parameter uncertainty model of the system,the linear combination of output variables is selected and solved as the controlled variable.Under the action of the feedback controller,the constant setting value is tracked to realize the control of the leaching process under the uncertainty disturbance.Compared with the traditional PID control method,the results show that the control strategy combined with self-optimization and nonlinear predictive control has better control effect.