Fuzzy neural network compound control and its applications for dust trap of electric arc furnace

The dust produced by steel-making in an electric arc furnace is one of the most serious pollution sources. A kind of intelligent control strategy is presented and some presently-existing problems are discussed based on the complexity of the dust trapping process of an arc furnace. The tendency of the quantity of smoke, produced in different smelting steps, to change can be predicted with a neural network. A basic output value can be obtained from the basic controller, and then the operating points of the system can be changed easily and the revolution rate of the dust-trap draught fan can be adjusted in time. A good effect has been obtained and the energy cost has been reduced in this dust trap process by means of fuzzy inference to compensate for the effect of the bag-house on the system