Application of fuzzy control for iron ore sintering process.

Abstract

SynopsisIn the iron and steel making shops, some processes are so complex that it is difficult to formulate precise control models. In such cases, process control is made on the basis of the skilled operator's heuristic knowledge. However, utilization of skilled operator's heuristic knowledge is one of the major problems to control the process precisely. In the fuzzy set theory, skilled operator's knowledge is expressed as membership functions to control the process as well as skilled operators.The fuzzy set theory was applied to the on-line control of return fine hopper level at the sintering plant in Fukuyama Works, Nippon Kokan K.K. This was the first application of fuzzy control to the practical operation. The commercial operation has started in August 1986. As the result, the deviation of return fine hopper level decreased from 12 to 4 %/h and the amount of return fine decreased by about 2 kg/t-sinter.

Artificial intelligence and expert systems in the steel industry

“…The on-line system, put into operation in 1986, reduced the fluctuation of the return fine hopper level from 12% to 4% and also reduced the amount of return fine ore about 2 kg/t. 33 , 34 At Kawasaki Steel, the blast furnace control system uses three types of information to determine the appropriate action: the furnace heat level evaluation from the ES, the furnace heat predicted by a mathematical model, and the previously taken control action. 35 Another project at Kawasaki investigated the control of the hot-stove combustion flow rate at the Chiba Works' No.6 blast furnace.…”