Study on prediction method for generation and consumption of coke oven gas

A direct prediction method based on empirical mode decomposition and echo state network is proposed to predict coke oven gas generation and consumption of the steel industry. First, the empirical mode decomposition is used to de-noise the practical data with high noise level. Then the direct relationship between the prediction origin and prediction horizon using echo state network is established without the need to close the network loop or iterate in the prediction process. Such a method has the advantage of avoiding the iteration error accumulation, and the corresponding forecasting precision is increased. The prediction results using practical production data show the validity of the proposed method and provide the scientific decision support for the gas resources scheduling.