A Feed Forward Network Mode with Prior Knowledge and Its Application to Forecasting Ocean-water Corrosion

YANG Xiaoming; XIE Tiejun; XIONG Liqing; DAI Mingan; ZHU Xiangrong

doi:10.13374/j.issn1001-053x.2000.03.014

Volume 22 Issue 3

Aug. 2021

Turn off MathJax

Article Contents

Article Navigation > Chinese Journal of Engineering > 2000 > 22(3): 242-244

YANG Xiaoming, XIE Tiejun, XIONG Liqing, DAI Mingan, ZHU Xiangrong. A Feed Forward Network Mode with Prior Knowledge and Its Application to Forecasting Ocean-water Corrosion[J]. Chinese Journal of Engineering, 2000, 22(3): 242-244. doi: 10.13374/j.issn1001-053x.2000.03.014

Citation:

YANG Xiaoming, XIE Tiejun, XIONG Liqing, DAI Mingan, ZHU Xiangrong. A Feed Forward Network Mode with Prior Knowledge and Its Application to Forecasting Ocean-water Corrosion[J]. Chinese Journal of Engineering, 2000, 22(3): 242-244. doi: 10.13374/j.issn1001-053x.2000.03.014

Citation:

YANG Xiaoming, XIE Tiejun, XIONG Liqing, DAI Mingan, ZHU Xiangrong. A Feed Forward Network Mode with Prior Knowledge and Its Application to Forecasting Ocean-water Corrosion[J]. Chinese Journal of Engineering, 2000, 22(3): 242-244. doi: 10.13374/j.issn1001-053x.2000.03.014

PDF( 352 KB)

A Feed Forward Network Mode with Prior Knowledge and Its Application to Forecasting Ocean-water Corrosion

doi: 10.13374/j.issn1001-053x.2000.03.014

1. Inst of Higher Edu, UST Bejing Beijing 100083, China;
2. Applied Science School, UST Beijing, Beijing 100083, China;
3. Qingdao Inst of Ocean Corrosion, Qingdao 266071, China

Received Date: 1999-04-27
Available Online: 2021-08-27

Abstract

Abstract

The relatonship between material corrosion and ocean envirnmental factors is studied.A feed forward network incorporated with prior knowledge is used to model the mapping between the corrosion rate and the environmental factors.The calculation results show that the model can give a better prediction of the corrosion rate than those given by BP model.
- forecast of ocean-water corrosion,
- prior knowledge,
- feed forward network model