Research On Damage Self-diagnosing And Sensor Optimal Placement Of Smart Structures

The self-diagnosing damage function is one all-important applied aspect of smart materials and structures. It is one of the main research contents of smart materials and structures. There are two important problems that are related to the self-diagnosing damage function. One is the method of damage diagnoses, the other is the problem of optimal sensor placement. It has been of an important theoretical meaning and a great practical value for applications of smart materials and structures to research on these two problems. The dissertation is supported by the grant from National Natural Science Foundation of China (90205031). In the dissertation, the self-diagnosing damage method and optimal sensor placement of smart materials and structures are mainly studied in order to offer theory guide for the design and realization of the fiber smart layer with self-diagnosing function. The main research contents in this dissertation are as follows:At first, On the base of the traditional frequency method for damage detection, an improved frequency damage indexes are given. The improved indexes include a damage orientation index and a damage degree identification index. Finally, the results of numerical value simulation experiment show that the improved indexes are more effective than the original indexes.Secondly, In order to eliminate the effect of damage degree, the strain modal difference vector is standardized, and to give the theory proving. In addition, in the dissertation a combination damage index based on the improved damage orientation index and the standardization strain modal difference vector is advanced for the neural network input character parameters.Thirdly, In order to avoid compound factors identification and to predigest the complexity of damage diagnoses, a multilevel damage diagnoses strategy of smart materials and structures based on modal analysis and neural network is advanced. The main research contents include the damage alarm based on RBFNN, the damage orientation based on auto-adapting PNN and the damage degree identification based on BPNN, And to study their basic principles and technology realization methods. Finally, the results of numerical value simulation experiment of composite material cantilever girder and the steel rule girder experiment show that the strategy is feasible for damage diagnoses.In the end, Aiming at the problem of optimal placement of strain sensor, the optimal placement criterion based on strain modal assurance criterion is advanced, and to study on the method that choosing some sensor placements by using strain modal information matrix. In addition, to apply the improved genetic algorithm to optimize residual sensor placement is studied.