Dynamic Parameters Research For Reinforced Concrete Silos Based On Improved Data-driven Stochastic Subspace Method

Reinforced concrete silos are widely used in various fields such as food, coal, structure and metallurgy engineerings, et. al, whereas, it's well known that their damage problem under earthquake disaster is still very serious. Therefore, it is necessary that one has to determine the dynamic response of reinforced concrete silos by further theoretical and experimental research for the requirement of structure design, application and disaster prevention.Currently, dynamic analysis of reinforced concrete silos is just involved in few literatures and no unified theories are available. For example, a simplified calculating method is given in the national standard"Seismic Assessment of Industrial Structures"for the frequency determination of single silo only, rather than the grouped silos, for the anti-seismic design. Among various dynamic analysis methods, the approach of ambient excitation has been successfully applied in engineering structures including huge buildings, bridges and transmission towers, and so on, while it hasn't been extended to dynamic analysis of reinforced concrete silos. Hence, it's interest to evaluate the dynamic parameters of silos using the ambient excitation method.In the present thesis, on the basis of available structural experiments based on ambient excitation and modal parameters identification methods, the coupled scheme with theoretical derivation, numerical simulation and experiments are developed to investigate the dynamic response of single and grouped reinforced concrete silos for the cases of various supports and loads. The corresponding research works are summarized as follows:1. Numerical simulation for reinforced concrete silos. In the present analysis, the finite element procedures of four silo models and two practical silos in service were separately established under different load cases. By comparison of numerical results of single silo and grouped silos, the dynamic behaviors of them are analyzed and the influence of granular materials on silos mode is investigated simultaneously.2. Dynamic parameters identification method for reinforced concrete silos. The thesis respectively establishes dynamic parameter identification techniques in frequency domain, time domain and combined time-frequency domain with the ambient excitation method, and then an improved data-driven stochastic subspace method is proposed and programmed for the purpose of parameter identification. Subsequently, to validate the computational efficiency, stability and accuracy of the proposed approach, the dynamic vibration of classical cantilever beam is analyzed using the present procedure.3. Experimental research for reinforced concrete silos based on ambient excitation. For the purpose of mode analysis, the single and grouped silos with specified column or wall supports are firstly tested with the specified ambient excitation and the related collocation scheme of sample points are investigated in detail. Next the practical bunkers in Chao Hua, Xin Mi City, and grouped silos in the eastern suburb, Zheng Zhou City, are tested using these procedures, respectively, for the acceleration data.4. Modal identification and analysis for reinforced concrete silos. Once the frequencies and modes are obtained by finite element analysis, the ambient excitation test can be conducted to determine the acceleration of the silo structures, and then, the peak picking method and the developed improved data-driven stochastic subspace method are respectively employed to obtain associated dynamic parameters. The comparison of numerical results of these two methods shows the numerical stability, efficiency and accuracy of the proposed method. Simultaneously, the dynamic parameters obtained by modal identification technique are useful to further improve the established finite element model.