A Two-Stage Online Prediction Method for a Blast Furnace Gas System and Its Application

The byproduct gas in steel industry is one of the most significant energy resources of an enterprise. Due to the large quantity of yield, fluctuation, and various categories of users encountered in a blast furnace gas (BFG) system, it is very difficult to accurately predict the amount of gas to be generated and forecast the users' consumption demand. In this paper, a two-stage online prediction method based on an improved echo state network (ESN) is proposed to realize forecasting in the BFG system. In this method, one completes the prediction realized at the levels of BFG generation and consumption using a class of ESN with input compensation and parameter optimization. At the second stage, to predict gas holder level of the BFG system, the energy flows being predicted at the first stage are denoised, and their correlation with the holder level are determined by using a concept of grey correlation with time delay. Then the effect factors exhibiting high correlation levels are extracted to construct the model of the gas holder. The prediction system designed in this manner is applied in the Energy Center of Baosteel Co., Ltd, China. The results demonstrate that the prediction system exhibits high accuracy and can provide an effective guidance for balancing and scheduling of the byproduct energy.

[1]  Jun Zhao,et al.  Improved Echo State Network Based on Data-driven and Its Application to vskip0.3 baselineskip Prediction of Blast Furnace Gas Output: Improved Echo State Network Based on Data-driven and Its Application to vskip0.3 baselineskip Prediction of Blast Furnace Gas Output , 2009 .

[2]  Herbert Jaeger,et al.  A tutorial on training recurrent neural networks , covering BPPT , RTRL , EKF and the " echo state network " approach - Semantic Scholar , 2005 .

[3]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[4]  Yoshikazu Nishikawa,et al.  An optimal gas supply for a power plant using a mixed integer programming model , 1991, Autom..

[5]  J. Deng,et al.  Introduction to Grey system theory , 1989 .

[6]  Bernard Mulgrew,et al.  Nonlinear prediction of chaotic signals using a normalised radial basis function network , 2002, Signal Process..

[7]  Frank Pettersson,et al.  Nonlinear Prediction of the Hot Metal Silicon Content in the Blast Furnace , 2007 .

[8]  Chonghun Han,et al.  A Novel MILP Model for Plantwide Multiperiod Optimization of Byproduct Gas Supply System in the Iron- and Steel-Making Process , 2003 .

[9]  Ronald R. Mohler,et al.  Neural-network-based load modeling and its use in voltage stability analysis , 2003, IEEE Trans. Control. Syst. Technol..

[10]  B. Dong,et al.  Applying support vector machines to predict building energy consumption in tropical region , 2005 .

[11]  Frank Pettersson,et al.  Method for the selection of inputs and structure of feedforward neural networks , 2006, Comput. Chem. Eng..

[12]  Harald Haas,et al.  Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Communication , 2004, Science.

[13]  Chao-Hung Wang,et al.  Forecasting integrated circuit output using multivariate grey model and grey relational analysis , 2009, Expert Syst. Appl..

[14]  Chonghun Han,et al.  Short-Term Multiperiod Optimal Planning of Utility Systems Using Heuristics and Dynamic Programming , 2001 .

[15]  Gopal P. Sinha,et al.  Strategic and Operational Management with Optimization at Tata Steel , 1995 .

[16]  Peter Fogh Odgaard,et al.  Estimation of Uncertainty Bounds for the Future Performance of a Power Plant , 2009, IEEE Transactions on Control Systems Technology.