Research On Modeling And Control Of Calcium Carbide Furnace

Calcium carbide is a commonly used chemical raw material and is widely used in the production of acetylene,organic synthesis and metal welding.In the current calcium carbide production industry,electric heating is mainly used to produce calcium carbide.Usually,calcium carbide furnace is used as the main production equipment.The temperature of the furnace is controlled by adjusting the position of the three-phase electrode.Calcium carbide furnace is a system with nonlinear,multivariable,severe random disturbance and strong coupling relationship.In actual production,calcium carbide furnaces are often controlled by manual operation.When the site personnel are improperly operated or judged to be wrong,it is easy to cause current fluctuations.This not only affects the quality of the calcium carbide,but also causes damage to the equipment.Therefore,in order to solve the problems in actual production,the control of the three-phase electrode regulation system of calcium carbide furnace has become the focus of this paper.Firstly,this paper takes the medium-sized closed calcium carbide furnace of a calcium carbide plant in Yangquan City,Shanxi Province as the research background,summarizes the development history and current situation of the calcium carbide industry,and introduces the research status of calcium carbide furnace modeling and calcium carbide furnace control.Based on the actual situation on the spot,the nature of calcium carbide,the production principle of calcium carbide,the structure of calcium carbide furnace equipment and its production process flow are introduced in detail.Secondly,because the designed electrode control strategy needs to be debugged and verified,the cost of debugging and verification in the field is very high.Therefore,it is necessary to establish a system model of the calcium carbide furnace.Firstly,relying on the input and output data of a large number of calcium carbide furnace systems collected on site,the basic framework of the calcium carbide furnace system model was established by using the neural network identification method.Then the system model of calcium carbide furnace was established by using extreme learning machine and support vector machine respectively.The same data samples were used to compare the effects of the two models.The results show that the extreme learning machine is more suitable for the establishment of the calcium carbide furnace system model.Finally,the electrode control strategy of the currently used calcium carbide furnace is analyzed and compared in detail,and the constant current method is taken as the basic control idea based on the actual situation on site.Then based on the constant current method and the actual control experience of the field,the fuzzy controller of the electrode regulation system is designed.Since fuzzy control is difficult to achieve precise control,iterative learning control is added on the basis of fuzzy control to improve the control precision.Finally,fuzzy iterative learning control is used to design a composite control strategy for the calcium carbide furnace electrode regulation system.And using the established model of calcium carbide furnace system,the effectiveness and feasibility of the control strategy are verified.