Title :
Improved Differential Simulation Method for Oilfields Development Indices Forecast
Author :
Min, Chao ; Liu, Zhi-Bin
Author_Institution :
Coll. of Sci., Southwest Pet. Univ., Chengdu, China
Abstract :
A differential simulation method is proposed for the oilfields development indices forecast in this paper. Then it is improved through BP neural networks with varying learning rate, which is employed in the process of parameter identification and makes the accuracy of medium and long-term prediction better. This model based on time-varying systems is applied in a practical example in the end of this paper and the results demonstrate that the accuracy is indeed improved.
Keywords :
backpropagation; digital simulation; forecasting theory; neural nets; petroleum industry; production engineering computing; BP neural networks; differential simulation method; long-term prediction; medium prediction; oilfields development indices forecasting; parameter identification; time-varying systems; varying learning rate; Chaos; Computational modeling; Decision making; Educational institutions; Input variables; Neural networks; Parameter estimation; Petroleum; Predictive models; Time varying systems; differential simulation; varying learning rate;
Conference_Titel :
Natural Computation, 2009. ICNC '09. Fifth International Conference on
Conference_Location :
Tianjin
Print_ISBN :
978-0-7695-3736-8
DOI :
10.1109/ICNC.2009.277
