System Response Processing and HHT Method on Dynamic Specification Determination using Cloud Computation

The vibration measurement tools are the primary sources to provide input data for the signal processing and system characterization subsystems for the mechanical structures' dynamic specification. The Hilbert-Huang transform (HHT), as the core of the identification systems, extracts the characteristic features to construct an accurate model of the system. In comparison with Fourier transform or wavelet transform, the HHT method can overcome their limitations while it can be implemented over cloud space. This paper uses empirical mode decomposition (EMD) and HHT by cloud computation to characterize a mechanical system. On the cloud space, the EMD method results in intrinsic mode functions (IMFs) of the structural acceleration responses, and the HHT calculations on the combination of these IMFs drives the value of natural frequency and system attenuation ratio. The acceleration responses of an eight-story structure are used to illustrate the performance of the proposed method. In the hardware implementation section, an ARM-based communication/hub board is used to transform and receive data from the cloud environment. The effect of the measurement noises, connection quality, and the different loading on the system identification process has been investigated. The implementation results show that the HHT calculates the system characteristics very accurately, and the effect of measurement noise, data communication link and loading type will have little effect on the system determination process.