Optimal Sensor Placement Based On Improved Gravitational Search Algorithm

Structural health monitoring(SHM)is essential when detecting damage in large and complex structures in order to provide a comprehensive assessment of the structural health state.Optimal sensor placement is a part of structural health monitoring system,which aims to use a limited number of sensors to obtain high-quality structural health diagnosis data.Therefore,optimal sensor placement plays an important role in daily monitoring and guiding maintenance of large structures.There are still three shortcomings in the existing methods for optimal sensor placement.First,the problem of optimal sensor placement is essentially combinatorial optimization,which is better than the selected measuring points and has the feature of large amount of calculation.Intelligent optimization algorithm with better optimization performance is required to solve such problems;Secondly,most of the current research methods select the number of modes as the target mode of sensor optimization based on the experience of engineers.However,the contribution of different modes to the bridge structure is different,so it is necessary to select the mode which has a greater impact on the bridge structure as the initial data.Finally,different numbers of sensors have different effects on the results of optimal sensor placement,while some studies only consider a certain number of sensor optimal placement schemes.Because of the above shortcomings,this paper proposes an optimum sensor placement method based on the Adaptive Gravity Search algorithm(AGSA).The main research contents and conclusions are as follows:(1)To overcome the disadvantages of the low search accuracy of gravity search algorithm,the attenuation factor of the algorithm is adapted.The initial attenuation factor of the search is small,and the particles perform a global search with a large step size,which enhances the search efficiency of the algorithm.The search has a large attenuation factor in the late stage,and the particles perform a local search in a small step size,which improves the search ability of the algorithm.The result of the 6 test functions shows that the optimization accuracy and convergence speed of the AGSA algorithm has been significantly improved.(2)The effective modal quality participation ratio is introduced in the modal analysis of bridge structure,and the target modal quantity is selected according to the contribution rate of each mode to the bridge structure,to ensure the validity of the initial data of optimal sensor placement.(3)Three-dimension MAC(TMAC)is used as evaluation criterion.AGSA algorithm can be used to solve the problem of discrete sensor optimization by double coding.Through parameters sensitivity analysis,the optimal parameters of AGSA algorithm are obtained.(4)In the process of determining the optimal sensor placement scheme,the number of sensors is taken as variable and the maximum non-diagonal element of TMAC matrix is taken as objective function,and the change of objective function with the number of sensors is discussed.Compared with the GA algorithm and PSO algorithm,AGSA algorithm is found to have a smaller objective function and better optimization result,which further verifies that AGSA algorithm can be used to solve the optimal sensor placement problem.