Online-Optimization Control Method Based On Performance Assessment For Combustion Process Of Coke Oven

Coke, as the main product of the coking production process, is an important raw material of national economy. With the dramatic increasing demand of the coke, coking enterprises make higher production indexes which include coke quality, coke consumption and production efficiency. The combustion process of coke oven is susceptible to the external factors. The flue temperature is always fluctuant, which may cause performance reduction of the control system and make a serious impact on the improving of the quality of the coke. In response to these problems, this paper studies online-optimization control method for the combustion process of coke oven.Firstly, on the basis of in-depth analysis of the combustion process of coke oven, this paper sums up control problems of coke production process. In response to these problems, this paper presents optimal control system for combustion process of coke oven which is composed of the two-level performance evaluation system and the online-optimization module.Secondly, to meet the demands of real-time performance evaluation and process-oriented performance evaluation of the control system, this paper designs two-level performance evaluation system for both of them. The two performance evaluation subsystems run independently, and the former one’s period is less than the latter one. When the performance of control system does not meet the production requirements, the system will use online-optimization algorithm to update the controller parameters. This paper further analyzes the production process’s demands for the stability and other performance demands of flue temperature, then establishes different multi-objective optimization model for different subsystems, and use multi-objective evolutionary algorithm to solve them. Simultaneously, in order to avoid the disadvantage of the intelligent optimization algorithm during its converging process, this paper presents a knowledge-guide based multi-objective differential evolution algorithm. The algorithm obtains evolution knowledge from an evolution population and feedbacks the knowledge to guide the search process. The algorithm improves the efficiency of the algorithm.Finally, according to the heating balance principle of the combustion process of coke oven, this paper derives of the simulation model and establishes a simulation system. In this paper, we assume that the system kindles the blast furnace gas. For the coking time and the heating gas calorific value have greate impact on the flue temperature, two simulation experiments are conducted. Simulations verify the effectiveness of the method.