Extension Neural Network in Water Quality Appraisal Application

The extension neural network is the extension theory and the artificial nerve neural network organic synthesis. The network takes the matter-element as the fundamental unit, and is connected by massive neurons. The network takes the matter-element extension and the correlation function as machine learning foundation, can better simulation human brain nervous system thoughts and intelligent behaviors. This article introduced the neural network element model and analyzed the neural network element extension, and the basic element transformation. By extension transformation, using a region in the outputting space to replace a point; the training speed is extraordinarily improved. The water quality appraisal is the water environment quality abbreviation is water adding quality fit and unfit, carries on qualitative or quota description, can reflect accurately water body quality and pollution condition. This paper weights the neural network by several major targets or quotas, which refers to neural network inputting samples. And we also establish extension neural network training water quality model, compare the extension neural network with the BP network, we can find the speed and effectiveness get better. The experiment proved that the appraisal effect with the extension neural network to the water quality to be more obvious.