Open AccessWater quality assessment of the river network in Wenzhou city using PCA-BP neural network model
Author(s) -
Fang Zhou,
Chongguo Tian,
Jiliang Xu,
Jing Wei
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/612/1/012018
Subject(s) - principal component analysis , artificial neural network , dimensionality reduction , curse of dimensionality , artificial intelligence , computer science , backpropagation , data mining , disadvantage , quality (philosophy) , pattern recognition (psychology) , machine learning , philosophy , epistemology
The water quality assessment is often challenged by how to determine the index weight and simplify the evaluation model. This paper proposed an integrated model called PCA-BP neural network model to solve the problem, which combines the dimensionality reduction capability of the improved principal component analysis (PCA) method with the self-learning ability of back propagation (BP) artificial neural networks. And its application for the plain river network in the main districts of Wenzhou city indicated that the evaluation result of PCA-BP was consistent with the single-factor method and PCA method in overall trend. It was demonstrated that PCA-BP model could evaluate the water quality of the study area more reasonably and accurately, as it avoided the disadvantage of a certain factor completely covering the information of the other ones in the single-factor method and the risk of over-optimism in PCA method as well.