@Article{sdhm.2026.079074,
AUTHOR = {Guansi Liu, Hui Jin, Keqin Ding, Hao Wang, Violeta Mircevska, Maosen Cao},
TITLE = {Optimization Method for Sensor Placement in Fatigue Monitoring of Crane Welding Structures Based on Damage-Risk Fusion},
JOURNAL = {Structural Durability \& Health Monitoring},
VOLUME = {},
YEAR = {},
NUMBER = {},
PAGES = {{pages}},
URL = {http://www.techscience.com/sdhm/online/detail/26605},
ISSN = {1930-2991},
ABSTRACT = {In response to the dynamic changes in fatigue damage location of crane welding structures under lifting loads and the difficulty in accurately obtaining the stress concentration factor of welds, which results in limited effetiveness of traditional health monitoring sensor placement. This paper proposes aa sensor placement optimization method that integrates damage prediction and risk assessment. Firstly, the influence of weld geometry on fatigue performance is analyzed, and a rapid estimation model for the stress concentration factor is established using a radial basis function support vector machine. Furthermore, a fatigue damage prediction model for the welded structures is constructed. Secondly, the finite element simulation is employed to model typical crane operating conditions, and the structural stress distribution and cumulative fatigue damage are calculated. Considering the occurrence frequency of each operating condition in actual work, weighted cumulative damage is obtained. By identifying points with high weighted cumulative damage, quantifying their occurrence probability and consequences, and constructing a risk matrix, all potential fatigue points are ranked based on risk levels. Consequently, an optimal sensor placement method driven by both damage prediction and risk assessment is formed. Finally, the accuracy of the stress concentration factor estimation model is verified through a scaled model fatigue test of the main girder structure. The proposed sensor placement method for fatigue monitoring is further verified using a specific type of casting crane as a case study. The results indicate that the proposed method can effectively integrate damage information from multiple working conditions, characterize the structural damage status, and achieve optimized placement of sensors.},
DOI = {10.32604/sdhm.2026.079074}
}