The three-dimensional energy resources demand-supply system is established for two regions of China, in which determination of unknown parameters is significant for the real energy resources demand-supply system. In this paper, based on Shanghai's energy statistic data (1999-2005), the parameters of the energy resources demand-supply system can be identified by using the neural network method, and an energy resources demand-supply system reflecting the real energy resources demand-supply relationship in Shanghai is obtained. The dynamical behavior of this system is also analyzed. Theoretical analysis and numerical simulation indicate that the obtained energy resources demand-supply system has the capacity of self-regulation and demonstrates a steady state. Finally, to ensure the steady development of the energy resources demand-supply system, corresponding measures are proposed. (C) 2011 Elsevier Ltd. All rights reserved.