一种远程水污染神经网络监测系统

介绍了一种采用紫外分光光度计及人工神经网络分析的计算机远程水污染监测系统； 介绍了系统构成原理和用于废水紫外光谱图识别的多层前向神经网络的设计， 分析了含有不同隐节点数的前馈网络的精度； 该系统可对远程水质进行监测；不需要测量废水的各成份含量而通过对废水光谱图的神经网络识别直接给出水质合格与否的结论； 该系统还可推广应用于红外及可见光范围的光谱图分析中， 从而大大拓宽该系统在水质监测乃至成份分析方面的应用；

A LONG-DISTANCE WATER POLLUTION MONITORING SYSTEM BASED ON NEURAL NETWORK

A long distance water pollution monitoring system is introduced, which mainly consists of an ultraviolet spectrophotometer and a computer, and is based on artificial neural network. The structure of the system and the design of the multiple feedforward neural network used for identification of spectrogram of wastewater are discussed. Precision of the networks with different numbers of hidden layer nodes is analyzed. Long distance water can be monitored by the system, and the judge of the water quality can be made through the identification for the spectrogram of wastewater by neural network, without the content measure of pollutants in wastewater. The system can be also used in the analysis of infrared or visible light spectrograms.