Prediction of Iron Ore Demand Based on Coupled Phase-Space Reconstruction and Neural Network
- Xiaojun Yan, Zhiya Chen
- Corresponding Author
- Xiaojun Yan
Available Online November 2015.
- https://doi.org/10.2991/itms-15.2015.350How to use a DOI?
- chaos; coupled phase-space reconstruction; neural network; iron ore demand prediction
- The over capacity of steel and structure adjustment of iron production cause the producing elasticity. As the major raw material, the demand of iron ore fluctuates significantly and it brings great trouble for steel enterprise in business decision-making. In order to improve the management decisions, the steel enterprises must carry out the effective predictions of the iron ore demands. Based on the coupled phase space reconstruction and neural network, we proposed a prediction model of the iron ore demand, which first used the raw demand data for the coupled phase-space reconstruction, then trained these reconstructed data with the neural network, finally predicted the iron ore demand according to the predicted time. Besides, the iron ore quarter demand data at 2001-2011 from a typical steel enterprise was used for verifying this prediction model. Results show that this prediction model of the iron ore demand is easy to operate and its predicted data is reliable, which can provide theoretical guidance to the scientific and reasonable management decisions.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Xiaojun Yan AU - Zhiya Chen PY - 2015/11 DA - 2015/11 TI - Prediction of Iron Ore Demand Based on Coupled Phase-Space Reconstruction and Neural Network BT - 2015 International Conference on Industrial Technology and Management Science PB - Atlantis Press SN - 2352-538X UR - https://doi.org/10.2991/itms-15.2015.350 DO - https://doi.org/10.2991/itms-15.2015.350 ID - Yan2015/11 ER -