Papers by Author: Xin Liu

Abstract: For the bridge health monitoring it is very important to understand the information what factors lead to the bridge damage. Clearly know these factors and the correlation between them can improve the efficiency of the bridge operation. But, the informed researches of damage factors are based on the expert's opinion and not on a systematic method of as the basis. First around 30 candidate factors are Scientific selected in reinforced concrete bridge related database in this paper, then these factors are divided into four common type, and in turn using rough clustering in the light of the four types, the rank of each type and comprehensive class are partition. Finally, the most significant influence factors to the structural damage are searched by the rough set theory (RST), and the association rules of each factor are analyzed.

Abstract: In order to solve optimal placement of bridge sensors based on the modal information, a multi-objective integer programming expected value model is established and the number of modal is considered as a stochastic variable in this paper, and here Fisher and MAC matrix are combined as the objective functions . Since DNA Genetic algorithm has the merits of plentiful coding, and decoding, conveying complex knowledge flexibly, these merits and the technique of stochastic simulation are also combined, which for estimating stochastic integer programming expected value models problem. And finally the feasibility of the algorithm is showed by XuGe Bridge as an example.