Research on prediction of water resource based on LVQ network
Jian Wang, Yuanyuan Zhang · 2011
In the accelerated modernization, China confronts with the serious challenges including population explosion, water pollution, though the Chinese water resource is ample. The scientific and reasonable prediction of water resource requirement is essential for the environment protection and continually development. After analysis the use of a serial of artificial neural networks for the water resource requirement prediction, this paper presents a novel algorithm based on the LVQ network with fuzzy feedback function. This algorithm can not only unfold the future water resource requirement after the historic data analysis, but also generate new input data with some reasonable problem resolves, which makes the algorithms with feedback and evolvement scheme. The user can adjust the resolve accuracy to speed up the convergence, which demonstrates better time cost performance than the traditional BP network. In the preliminary experiment, the algorithm based on LVQ network with fuzzy feedback reveals better performance in the application background with inaccurate and incomplete water resource data.