Research On Optimal Sensor Placement Of Health Monitoring System On Spatial Grid Structure

In the large-span spatial grid structure health monitoring system,the optimal placement of sensors is a key issue in the process of structural damage identification,health assessment,and post-maintenance decision-making and implementation.How to use a limited number of sensors in engineering structural health monitoring Realizing the most effective configuration to obtain as much structural modal information as possible has become a growing concern.Therefore,this paper has conducted a comprehensive discussion and research on a series of problems such as the reduction of degrees of freedom,the selection of modal order,the determination of the number of sensors,and the optimization of sensors in the optimal arrangement of sensors in the health monitoring of space grid structures.First,considering the large number of degrees of freedom of the nodes of the large-scale grid structure,the model polycondensation technology is introduced to remove the secondary degrees of freedom with small vibration displacement on the structure and little influence on the measurement results,so as to simplify the model.Using the 2-norm method based on Fisher information matrix,the most suitable modal order is selected.For the determination of the number of sensors,first use the QR decomposition method to select the initial measuring point of the structure,and then use the MAC criterion to determine the optimal number of sensors.The corresponding modal calculation parameters were extracted,and the optimal sensor placement program was compiled using MATLAB software in combination with the above algorithms.Secondly,through a 14×1×0.7m space grid structure,using the three traditional optimization algorithms of EI,MKE and MSE to select the 10th and 20th order modal information to optimize the layout calculation of the grid structure,the results show that The sensor layout results obtained by different methods can reflect the modal characteristics of the grid structure itself,which is consistent with the results of ANSYS modal analysis and model polycondensation.At the same time,the optimization effect is evaluated through three different evaluation criteria.The results show that the more modal orders involved in the calculation,the more comprehensive the information obtained,but as the modal order increases,the amount of calculation may increase,resulting in The result is poor stability.Observing the evaluation indexes of various algorithms,the effect of MSE method and El method is relatively better.In addition,through the intelligent optimization algorithm-PSO algorithm,a fitness function based on the MAC method is constructed,and the influence of selecting different modal orders on the optimal arrangement of measuring points is compared and analyzed.By observing the distribution of the degrees of freedom of the measuring points on the grid and the calculated value of the MAC non-diagonal elements,it shows that the improved PSO method has the characteristics of simple and efficient algorithm,rapid iteration convergence,and not easy to fall into the local optimal solution.Finally,through the space grid structure model corresponding to the finite element model,the three traditional optimization algorithms introduced in this paper are experimentally verified.The modal parameters of the grid are extracted,and the dynamic characteristics of the modal data obtained by the experimental grid under the excitation of the hammer and the theoretical results of the finite element analysis are compared.The experimental results show that the first-order mode shapes of the measured structures of each algorithm are very close to those of ANSYS analysis,and the frequency difference is not large,which verifies the feasibility of different optimization algorithms and the reliability of the experimental model.It provides important reference value for the sensor arrangement in the health monitoring of the spatial grid structure.