A Soft-Sensing Approach to On-Line Predicting Ammonia-Nitrogen Based on RBF Neural Networks
Changhui Deng, Deyan Kong, Yanhong Song, Zhou Li, Jun Gu · 2009
Measuring ammonia-nitrogen in the aquaculture water is always a problem that how to carry out the on-line monitoring in the process of industrialized culture. There isnpsilat a more effective method to realize the real time on-line monitoring at present. Some even need expensive instruments and operators having high skills. The normal methods can only be performed in the laboratory, so it canpsilat be accomplished the requirement of the fast-field evaluation. Because of above factors, the development of industrialized culture in our country is not fast enough. In this paper it is built that the intelligent mathematic model which is used to predicting ammonia-nitrogen in the aquaculture water and which is based on RBF Neural Network (RBF NN). Through comparing the model values with the measured values, we can emend the predicting model the second time to realize the intelligent prediction of ammonia-nitrogen. The results show that the soft-sensing approach to on-line predicting ammonia-nitrogen based on RBF neural network is effective.