Systolic Array-Based Wavelet and Hilbert Transform Computations with Application to Electric Power Quality Signal Monitoring

• Main Authors: Feng-Hao Hsu, 許峯豪
• Other Authors: Shyh-Jier Huang
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/75091068636960463234

碩士 === 國立成功大學 === 電機工程學系碩博士班 === 96 === Electric power quality monitoring has emerged to be a critical issue from perspectives of power producers and high-tech industries. By grasping the polluted source as well as adopting efficient strategies, electricity cost can be significantly reduced, while the life of equipment will become longer. In view of this importance, a novel method with high-efficiency time-frequency computation is proposed in this thesis, in which the wavelet transform is integrated with Hilbert transform such that the monitoring performance is significantly improved. Meanwhile, a systolic array-based design of field programmable gate array (FPGA) is also employed to reach the hardware realization, where the analog-to-digital circuit is combined such that a platform of electric power quality monitoring is overall realized. In order to assess the performance of this prototype, signals including voltage interruption, voltage sag, voltage swell, voltage harmonics, and capacitor-switching transients are all tested. Software simulations and experimental results help support the proposed method for the application considered.