ニューラルネットワークに基づくクレーン走行ばりの損傷評価

概要

論文の詳細を見る
The welded girders of cranes with high loading cycles develop cracks in the welds on upper flange and/or lower flange. It needs the maintenance and periodical inspection to keep the safety of crane service. This paper showed the method of damage level evaluation of welded crane runway girders, especially in iron works using artificial neural network, and showed the important factors which contribute to the damages on crane girders by sensitivity analysis. The damage evaluation for 20 examples using this system showed the satisfactory results on the decision concerning fatigue crack occurred or not. And the results of the sensitivity analysis showed that the damages in the vicinity of the upper flange were influenced by the kind of cranes, maximum wheel load, and the product of used years and frequency of use, and damages in the lower flange were influenced by runway girder span, and the product of used years and frequency of use.
社団法人日本建築学会の論文
1997-10-30