基于模型优化与TR方法的水工弧形闸门损伤识别

Damage detection on hydraulic radial gates by model updating and TR method

针对水工弧形闸门的安全监测问题,基于结构动力模型优化技术提出水工闸门损伤识别与健康诊断一种解决方案。该方案将水工钢结构的损伤识别和健康诊断问题首先转化为一种非线性约束优化问题,通过建立带边界约束的非线性最小二乘目标函数,极小化结构实测模态与理论计算模态之间的误差,采用具有更强鲁棒性和可靠性的信赖域(TR)方法求解该优化问题,达到识别结构损伤位置及其程度的目的,从而为水工钢闸门的健康诊断提供技术措施和方案。最后通过一个水工弧形闸门模型的损伤识别数值仿真和试验研究,证明了该方案应用于水工钢闸门结构损伤识别与健康诊断的可行性和有效性。

This paper puts forth a new method for hydraulic gate structural damage identification based on model-updating techniques.This method transforms an identification problem into an optimization problem,defines anobjective function for a nonlinear least-squares problem with bound-constraints,and identifies the location and extent of structural damage by minimizing the sum of differences between the experimental and analytical modal data.Then a trust-region approach is used to solve the minimization problem to make such optimization process more robust and reliable.An application to damage detection of a model radial gate of steel structure that were numerically modeled and experimentally tested in laboratory,are presented.Results show advantages of the proposed method for damage detection and health diagnosis on hydraulic structures.