Research On Intelligent Integrated Optimal Control Technology Of Iron Ore Sintering Process And Application

As an important step in iron ore production, the sintering process has a direct influence on state of the blast furnace and output of steel. Since the sintering process is a dynamic system with long circuit, multivariable and complex mechanism, it is hard to perform optimal control of process by using traditional control theory and methods.In this paper, the intelligent integrated optimal controlling methods and strategies of sintering process are discussed, after an analysis of the characteristics of sintering process. The complicated optimal control of sintering process is divided into two parts:optimistic control of blending and heat status of sinter process. According to goals and characteristics of sintering process, controlling of blending, burning through point (BTP) and multi-objective optimal control technique are investigation respectively, and integrated optimal control system of sintering process is developed. The main study achievements include:(1) Intelligent integrated optimal control structureBased on the analysis of the complicated control characteristics of the iron ore sintering process, the basic frame of the intelligent integrated optimal control is proposed. The three parts of basic concept, which are the description of structure, the principles and steps of design, are put forward. This method provided a new idea for optimization control of the iron ore sintering process.(2) Optimal control of blendingBlending is the foundation of the sintering process. The predicted model of composition based on artificial neural networks are established, the production cost optimal model based on linear programming is proposed, and the expert rules model aimed to adjust the percentages is constructed, then structure of the blending in the sintering process is optimized.(3) Predictive model and multi-model transition controller based on fuzzy weight of BTPBTP is crucial factor during sintering, which affects heat status of sintering process. The strategies of controlling BTP are studied deeply in this paper. Due to the sintering process having the characteristics of long time-delay, the prediction models of BTP based on neural network are put forward. Based on the analysis of the characteristics of strong nonlinear and strong coupling in the process, the integrated control method of fuzzy control technology and predictive control technology is studied. The optimal control models based on the multi-model transition controller of BTP is established. The performances of the two models have been compared through simulation. And the multi-model transition control method has better stability and adaptability.(4) Multi-objective Optimization and Control based on Evaluation Function AlgorithmA multi-objective optimal control method based on evaluation function algorithm is proposed. The objective evaluation function of the system is designed, which transform the multi-objective optimization problems into the single-objective optimization problems and eventually results in coordinating control of the burning through point and the bunker-lever.(5) Intelligent integrated optimal control system of sintering processThe basic automation is realized through the computer control system of EIC. The intelligent integrated optimal control system is developed based on the proposed frame of the intelligent integrated optimal control system, which has been embedded in the EIC in the sintering process of some Iron & Steel Co. and eventually results in integrated optimal control of the iron ore sintering process.Through the application of the intelligent integrated optimal control technology in the iron-ore sintering process, the structure of blending is optimized, and the fluctuation of sinter compositions and BTP is restrained. The output and quality of sinter are improved effectively. The results of actual runs show that the system is effective.