The Gas Predicting And Scheduling Model For Steel Enterprises

Steel enterprises are high-energy consuming. It is critical to use energy more effectively, improve energy efficiency and reduce the waste of energy. In the process of steel making, it gives off a large amount of by-product gases (blast furnace gas, coke gas and converter gas). These gases are not only an important secondary energy, but also a main atmospheric pollution sources. If they are not effectively collected and used, it will lead to the waste of energy, the pollution of atmosphere, and the increase the cost of the products.In this thesis, we compare and discuss the utilization of the by-product gases in both domestic and foreign steel enterprises. Based on the comparison results, we point out the gap between the domestic enterprises and those advanced enterprises. We then introduce the model of time series analysis and prediction briefly and propose a minimum variance adaptive prediction algorithm. Furthermore, we improve the algorithm and put forward the multi-step adaptive prediction algorithm. We use time series prediction theories to make optimization process of the by-product gases of steel enterprises and compare the advantages and disadvantages among several algorithms. Finally we establish the gas flow prediction model. Comparing the difference between the predicted results and the actual gas measuring data, we using least squares regression and partial least squares regression theory to establish the gas optimization scheduling model. By establishing the gas flow prediction and scheduling model, we can rationally use the by-product gases. In this way, steel enterprise will not only guarantee the gas consumption of its users but also reduce the cost and increases the profits.