Utilizing an average method to extract the dynamic properties of structures with image processing

Dynamic properties of structures, such as natural frequency, shape modes, and damping ratios, play a decisive role in structure behavior against dynamic loads like earthquakes. Determining the earthquake forces imposed on structures based on the design spectra is one of the utilization of these properties. Other applications of these features could be utilized to update the finite element model, detect potential damage in structures, long-term health monitoring of structures, and evaluating the safety of structures after heavy loading. Therefore, accuracy and reducing the cost of extracting these properties would have a significant role in improving the efficiency and consequently of the useful life of structures. In other word, the more accurate of structural dynamic properties means, the more accurate determination of the seismic response of the structures. These properties depend on a great deal of detail, such as material behavior and the geometry of the structure, which could not be easily simulated in analytical models. So, performing seismic tests on structures is the most reliable method for obtaining these properties. Determination of these properties have been done in a variety of ways. However, various methods have been developed in many studies to extract these quantities by image processing. The aim of this article is presenting a novel approach to increase the accuracy of remote sensing by image processing. Therefore, the current paper is an attempt to apply the average method to improve efficiency as a cheaper method for obtaining the dynamic properties of structures. For this purpose, a cantilever aluminum beam and a Three-story frame with additional mass have been considered, and the commercial camera captures the vibration of the structure. The extracted displacements of each four points on the edge of the specimen are recorded as the input signals of the system. With two numerical derivatives of these displacements, the acceleration of the structure is obtained. Peak survey method utilized to extract natural frequencies, damping ratios, and mode shapes of the each selected point. The averaging method applied to calculate the final properties of the structure.

At last, the results are compared with the values ​​obtained from the acceleration sensors embedded on the structure and the finite element results. Then the accuracy and error of the algorithm are evaluated. However, these results could be utilized as the input information in the health-monitoring of the structures. The results show that the novel method did not improve the accuracy of the first three natural frequencies modes of vibration in comparison with the standard method for the cantilever beam. It is also observed that the new method wouldn’t make a significant difference in the calculation of damping ratios of the system. On the other hand, although the existence of cables and sensors would reduce the accuracy of image processing and recorded displacements, the new algorithm improves the estimation of the first three shapes modes. In the same way, the same function was performed for the Three-story frame structure. Although its natural frequencies did not change for the first three modes of vibration, the mode shapes are closer to the values ​​obtained from the accelerometer sensors.