Release Time:2022-06-30  Hits:

Indexed by: Journal Article

Date of Publication: 2022-06-29

Journal: 测试技术学报

Institution: 后勤处（后勤党委）

Volume: 19

Issue: 2

Page Number: 146-151

ISSN: 1671-7449

Abstract: Fracture energy and crack length, as two imperative parameters of aggregates, can't be measured by usual methods. Based on neural network, a novel design of soft-sensor is presented, in which grey relation analysis as an assistant tool of secondary variables selection in the modeling course of soft - sensor is used to predict the fracture energy and crack length of concrete bending beams. Finally, the result shows that the design possesses high accuracy and can meet the practical demands.

Note: 新增回溯数据