The Research Of The Reheating Furnace Temperature Control Method Based On Fuzzy Neural Network

Crude oil is an important energy source in industrial life and the long-distance pipeline transportation of crude oil is an important way of oil storage and transportation. Because of the characteristics of high wax and easy solidification,we must keep heating oil in the course of transmission in order to ensure the smooth transportation of crude oil in the pipeline. Oil furnace is widely used in the crude oil gathering system of oil industry which mainly used to heat crude oiK natural gas and well products in order to achieve its transportation, settlement, separation, dehydration and initial processing. It is one of the largest consumption equipments in the oil and gas field. How to improve the automatic control levels and efficiency of the furnace has a high research value and practical value.The control of furnace combustion system mainly includes three aspects:First, control the furnace temperature within reasonable limits to ensure the smooth transportation of crude oil industry needs.Second, maintain optimum air-fuel ratio. Because only when the air-fuel ratio is the best,the fuel can be fully burning. Otherwise, not only waste fuel but also pollute the environment. Third, adjust the wind withdrawals and the air supply to maintain negative pressure or positive pressure of the heating furnace and to guarantee the security needs of industrial operation. This article mainly researchs on how to keep the oil temperature within the requirements of industry assumed that the air-fuel ratio is under the premise of effective control. Because the furnace is a nonlinear time system with the characteristics of large inertia, delay and strong coupling,it is difficult to establish the mathematical model precisely.Using traditional control methods based on mathematical model is difficult to achieve effective control of the temperature. As an intelligent control, fuzzy neural network integrated both the advantages of fuzzy reasoning and neural network. This article uses the method of fuzzy neural network to control the temperature of the furnace.In optimizing the parameters, in the early the global search ability of genetic algorithm is powerful,but in the late its local search ability is bad; in the early BP algorithm is badly influenced by the initial weights and the global search ability is bad,but in the late its local search ability is powerful. This article integrated the advantages of genetic algorithm and BP algorithm to use GA-BP algorithm to optimize the parameters of fuzzy neural network off-line.Then use BP algorithm to optimize the parameters of fuzzy neural network. Finally, to simulate in the MATLAB environment and achieved a good result. First,the comparison of control effect of the fuzzy neural network and the traditional method of PID shows that the method proposed in this paper is better than PID method at response time, overshoot, and interference. Then the comparison of the optimization results of GA-BP algorithm and BP algorithm shows that the optimization method of this paper is superior to BP algorithm.