Research And Application Of Density Wide Domain Intelligent Control Method In Heavy Media Separation Process

In recent years,with the continuous promotion of intelligent coal washing plants,the heavy medium separation is an indispensable link in coal washing.However,due to the change of coal quality and the development of ash recovery in actual production,higher requirements are put forward for the control range of density and the adjustment speed of density.Through the investigation and analysis on the production site,the main reasons are that the buffer capacity of the qualified medium bucket is limited,which cannot meet the change of suspension volume caused by the large density adjustment,and the control flow of the water supply valve and the diverter valve is limited,which cannot adjust the density quickly and in a large range.Therefore,this paper proposes intelligent density wide domain control of heavy medium separation process,analyzes and studies the process of heavy medium separation,and divides the separation process into steady-state control of density,rise control of density step and fall control of density step from the perspective of control,and studies three control modes and switching strategies.In view of the problem of insufficient buffer capacity of qualified medium barrel in the existing separation process of three products' heavy medium cyclone,the reverse diversion process is put forward.The suspension in qualified medium barrel is transported to the concentrated medium barrel through the reverse diversion pump to increase the buffer capacity of qualified medium barrel.Rapid adjustment of density adds the dosing valve to the control of density adjustment to increase the speed of density increase.The wide area density control scheme adopts system switching and different control modes for different working conditions.In the steady-state mode,the diverter valve adopts fuzzy control,the feed valve adopts PID control,and the feed valve adopts constant value control.The reverse shunt and water refill valve of density step down mode are controlled by PID.The density step-up valve is controlled by sampling.Mode switching is completed by using BP neural network and SVM classification.BP neural network is used to predict the liquid level of qualified media barrel after adjustment.SVM judges the working conditions according to the predicted liquid level and density deviation value,and realizes the switching between modes.After making the control scheme,this paper analyzes the change of the flow rate in the qualified medium bucket in the process of heavy media separation in detail,and finally establishes the transfer function matrix of heavy media separation based on the equilibrium relationship of the volume in the qualified medium bucket and the equilibrium relationship of the medium.The parameters of the transfer function are obtained through industrial field experiments and data collection analysis.Finally,Simulink is used for simulation verification.The simulation results show that the switching between modes can be realized smoothly and the adjustment effect is better.In this paper,AB production of 1769 series of PLC as the underlying core hardware,the upper computer software uses kingview to complete the connection between the underlying hardware and the upper software platform.In the software platform software platform is realized by using OPC communications between the communication,including Python,database and communication between the OPC server.The system has been put into trial operation in xinliu coal preparation plant.In the trial operation stage,the system operates reliably and can well adapt to the problems of repeated adjustment of density set value and large adjustment range caused by coal seam gangue on site.The stability of density has been greatly improved.