Research On Controlling Method About The Burning Furnace-eyes Automatically Of The Operation Robot For Calcium Carbide Furnace

With the increasing production of calcium carbide in China,some problems in the process of discharging calcium carbide out of the furnace,such as labor intensity,severe working environment for workers(high temperature and dust),safety risks and so on,become more and more prominent.The whole calcium carbide production industry has made pressing needs about the automation,safe production,and high efficiency when producing the calcium carbide.Therefore,some operation robots for the process had been developed,that is,discharging CaC2 out of the furnace,which can accomplish most technologies of the process already.However,none of them is capable of achieving the technology automatically called heating the furnace-eyes,which hinders the full automation of the process.In response to this situation,this paper carries out a research on controlling method about the burning furnace-eyes automatically for calcium carbide furnace based on an operation robot developed by HIT Boshi Research Institute.Firstly,the physical characteristics of the AC arc are studied,and the arc of a burner is equivalent to a time-varying resistance.Based on the law of conservation of energy and the physical properties of the arc,a time-domain arc model described by differential equations is established which takes the arc length as input,the arc current as the output and the arc conductance as state variables.The Roche coil is chosen as the current detector for the burner based on the characteristics of the burner current and the current research of the current detection.The influence of each disturbance on the accuracy of the Roche coil is analyzed and the corresponding anti-jamming measures are put forward.Secondly,the model of the burner circuit system and the position servo tracking system model of the burner are established respectively through the analysis of the composition principle of the burner system.Based on the two model,the burner system model is established that takes the output of the controller of the control system as the input and the arc current as the output.In order to determine the unknown model parameters in the burner system model,a method of identifying the model parameters based on the measured data is proposed.Then,on the basis of the burner system,the control strategy of the automatic burner control method is designed from the requirements of the automatic burner control system,and the arc current under the maximum arc power of the burner is calculated by theoretical analysis.Considering the characteristics of non-linearity and time-varying nature of the burner system,a BP-PID control method is proposed combining the advantages of PID control and BP neural network control,which can realize the on-line tuning of PID parameters.The effect of BP-PID control is compared with PID control by the simulation.Finally,based on the experimental platform of the specific robot,the identification experiment of the parameters of the burner system,the verification experiment of the burner system model and the verification experiment of the automatic burner control method are carried out respectively.The results of the first two experimental show that the arc model and the burner system model are correct.The results of the last experiment show that the automatic burner control method based on PID control and BP-PID control can realize automatic burning and improve the efficiency of the thermal power of the burner arc.Compared with the thermal energy generated by the target power,the heat generated by the artificial burning arc is 70.6% of the maximum heat in the radial experiment,the heat generated by arc under PID control is 87.4% and the heat generated by arc under BP-PID control is 98.5%.In the axial experiment,the heat generated by arc under PID control is 85.5% and the heat generated by arc under BP-PID control is 94.8%.