Study On On-line Optimal Control Method Of Coke Oven Heating Combustion Process

Coke oven is an important thermal equipment for metallurgical use.Its structure is complex,and the process of heating and combustion is more complex physical and chemical reaction process.Coke oven flue temperature is one of the important parameters of coke oven production process control.The stability of flue temperature directly affects the quality of coke and production cost.However,the production process is restricted by various factors such as the quality of the production environment,the quality of the raw materials and the advanced testing equipment,etc.It makes improving the quality of coke,reducing energy consumption,saving production costs and protecting the environment the focus of the coking industry.For coke oven heating combustion process is difficult to establish mathematical model of the problem,put forward for coke oven fire way temperature control objectives.It based on the T-S fuzzy model and select fixed coke oven gas,blast furnace gas and blast furnace gas,adjust the coke oven gas,coke oven machine side of fire way temperature,two kinds of conditions to build a mathematical model of coke oven heating combustion process.The front part was identified by fuzzy C-means clustering algorithm,and the back part was identified by least square method.Due to the poor clustering effect of fuzzy C-means clustering algorithm,the sensitivity of initial clustering number setting and the tendency to fall into local extreme value during iteration,and combined with the advantages of good clustering effect and fast training speed of subtraction clustering algorithm,subtraction clustering algorithm and fuzzy C-means clustering algorithm were combined to identify the antecedent of T-S fuzzy model.Since the identification precision of the model is small,the least square method is adopted to identify the model.In this paper,based on T-S fuzzy model,through the fuzzy PID control algorithm to realize the focus fire way temperature control,due to the fuzzy PID control algorithm optimization of slow speed and low accuracy of control weaknesses,combined with differential evolution algorithm optimization capability is strong,the advantages of good stability,the introduction of differential evolution algorithm is combined with fuzzy PID control algorithm to optimize control.The simulation results show that the T-S fuzzy model has small root-mean-square error,high stability and high fitting degree.On this basis,the differential evolution improved fuzzy PID control algorithm has short response time and small overshoot.Based on T-S fuzzy model,the differential evolution improved fuzzy PID control algorithm can effectively control the coke oven fire path temperature,and well reduce the coking energy consumption,meeting the process requirements.This study not only provides a reliable mathematical model for the coke oven flue temperature control system,but also provides an optimized control algorithm for the coking industry.In order to achieve the goal of energy saving coke oven heating combustion process optimization control to provide an effective solution for the coking industry efficient and low-cost production provides a theoretical basis.