| Because the performance of the flotation machine is greatly limited by the rotor speed,it is difficult to ensure that the flotation machine not only has sufficient inflation,but also produce the static flow separation environment required for mineral separation.In the early stage,the research group upgraded the stirring system of the traditional laboratory flotation machine,designed the double impeller stirring system flotation machine,and optimized some of its structural characteristics and working parameters.However,there are few studies concerning its performance,so it is difficult to scale up and apply it to industrial production.In order to solve the above problems,it is necessary to optimize and improve the performance of the mixing system of laboratory double impeller flotation machine.In this paper,five types of centrifugal impellers(a1-a5)and three types of mixing impellers(b1-b3)were taken as the research objects to compare the performance of the combined mixing system and determine the optimal mixing system;The performance prediction model of double impeller system was established by using BP neural network.Through the synchronous flotation comparison test device,the performance of the laboratory flotation machine was compared after expansion,so as to pave the way for further industrial application.The main research contents and conclusions of this paper were listed as follows:Through the experimental study on the structural optimization of centrifugal impeller,a5type was determined to be the optimal centrifugal impeller.Using this impeller,the inflation amount was 0.356m3/(m2·min),the concentration difference of methylene blue was 0.26mg/L,and the beneficiation efficiency was 19.21%.Compared with the traditional flotation machine,the performance indexes of the double impeller flotation machine were better than those of the traditional flotation machine,in which the inflation and the cavitation intensity was increased by0.073m3/(m2·min)and 2.3%,respectively.At the same rotation speed,the grade,recovery,and beneficiation efficiency of the concentrate were increased by 1.3%,4.4%,and 3.9%,respectively.Under the condition that the rotating speed of the double impeller flotation machine was lower than that of the traditional flotation machine,the concentrate grade obtained by the two types of flotation machines was nearly the same,while the recovery and beneficiation efficiency of phosphate using double impeller flotation machine was increased by about 6%and 3%.Through the experimental study on the structural optimization of the mixing impeller,it was concluded that the optimal mixing system was the matching of the optimal centrifugal impeller a5and the mixing impeller b3.With this mixing system,the inflation volume,suspension concentration,and beneficiation efficiency was0.327m3/(m2·min),4.6%,and 24.77%,respectively.Compared with the traditional flotation machine,at the same speed,the separation performance of the double impeller flotation machine was obviously better than that of the traditional flotation machine,in which the grade,recovery,and beneficiation efficiency were increased by about 2%,28%,and 12%,respectively.Further,the performance prediction model of double impeller system was established by using BP neural network.The obtained models were BP prediction model of inflation performance,suspension performance,and flotation efficiency.The training effect of these models was good,and the corresponding correlation coefficient R reached 0.99878,0.99973 and 0.9999,respectively.The research on the performance of the synchronous flotation comparative test device showed that the performance of the double impeller flotation machine was better at a lower rotation speed,in which the inflation volume was increased by0.0265m3/(m2·min)and the gas holdup was increased by 3.5%.In addition,the suspension dispersion state of minerals in double impeller flotation machine was better than that of traditional flotation machine.The static flow separation environment was achieved with double impeller flotation machine. |