| Summary: | 碩士 === 國立成功大學 === 電機工程學系碩博士班 === 91 === Because of the fast development of high-tech industry, the amount of data generated either from digital device or analog devices has been increased significantly. Particularly, the data to be analyzed are often affected by the characteristics of faulted equipment or elements. In other words, the unexpected disturbances may largely downgrade the accuracy of the data analysis. Hence, an effective approach that can help distinguish different disturbances would be useful to forewarn the operators, increasing the reliability of the system operation.
In this paper, a box counting-based fractal geometry approach is proposed for the disturbance detection, which was hardware-implemented by using the field programmable gate array. In the proposed approach, because the fractal geometry help quantize the behaviors of chaotics, a fractal dimension computation procedures has been embedded into the FPGA chips such that the disturbance-detection capability can be improved to a large scale. This detection approach has been tested on several power system disturbances. With the test results, they have validated the feasibility and practical values presented by the method for the application investigated.
|
|---|
