Research And Realization Of The Optimization Of Thickener Operation Considering The Fluctuation Of Feed

The thickening-filter is one of the key processes in the hydrometallurgical process.The optimization of the thickening-filter process can reduce the operating energy consumption and improve production efficiency while ensuring safety.The current thickener optimization model designed by the research group is mainly based on the planned value of the thickener feed volume to plan the operation of the thickener.However,in actual production operations,the feed volume of the thickener is often affected by the upstream production process,causing fluctuations which may make the thickener’s feed volume to change drastically in a short period of time and deviate from the planned value.In order to adapt to the continuous fluctuation of the thickener feed volume,it is necessary to optimize and adjust the thickener operation according to the actual feed of the thickener.This thesis first introduces the research status of the optimization control of the thickeningfilter process,and then introduces the production process and primary production equipment of the grinding-floating process and the thickening-filter process.In order to realize the optimization of the operation under the fluctuation of the feed volume,this thesis constructs a two-stage thickener feed volume prediction model based on ARIMA-PLS,which is used to predict the average feed volume of the thickener in the next 4 hours.The optimization of the thickener operation is then conducted using the predicted value.First determine the delay time between the upstream process variables of the thickener and the feed volume of the thickener,and select the variables that have a greater correlation with the feed volume of the thickener as the input variables of the prediction model,and then use the rolling ARIMA prediction model to make predictions of the input variables through autoregression.The PLS regression model is then be used to predict the feed volume of the thickener by inputting the predicted value of the variable.The simulation results show that the prediction accuracy of the model meets actual production requirements.In order to avoid the influence of prediction errors on the operation safety of the thickener,this thesis takes the time series of the underflow pump discharge of the thickener as the operating variable,and the economic index of the thick filter press operation as the objective function,the safety of the thickener as the constraint,and takes the feed volume prediction error as an uncertain variable to bulid a two-level interval optimization model considering the uncertainty of the feed volume prediction error,and then a heuristic-based optimization method is applied to solve the optimization problem.The simulation results show that the proposed optimization model can make the the obtained operation meet the constraint conditions when the feed volume prediction error vary within a certain range,and the energy consumption economic index can be minimized.Aiming at the problem that the optimization results obtained may not meet the actual production requirements when the feed volume of the thickener changes drastically during the operation of the optimization model,a thickening-filter operation adjustment strategy is proposed.First,based on the operating energy consumption and operational safety of the thickener,a comprehensive evaluation index for the thickeningfilter operation is established to evaluate the optimization results,and then the comprehensive evaluation index is used every 4 hours to determine whether the interval optimization model is required to adjust the operation of the thickener.The simulation results show that the operation adjustment can effectively reduce the fault rate during the operation of the thickener and improve the running stability of the thickener.Finally,the optimization control and adjustment algorithm of the thickener was implemented in the actual beneficiation site.