| 摘要: | 關於智慧電網的發展，有效的電力品質事件檢測，對於由不同分散式的能源系統組成的微電網和電力系統的監測辨識是非常重要的課題。電力品質事件的發生，嚴重會造成商業和工業用戶大量損失。加上近年電力電子技術廣泛的應用，以及電力系統中的非線性負載使用量大增，使得諧波問題越來越重要。電壓驟升驟降、中斷、閃爍以及頻率偏移等也是常見的電力品質問題。
傳統上電力事件的分析方法有很多種，例如檢測基頻的方法有零交越演算法(Zero crossing)、檢測諧波可使用快速傅立葉演算法(Fast Fourier Transform,FFT)等等。以快速傅立葉轉換為例，雖然簡單且計算速度快，若是訊號特性不符合FFT的使用限制，分析結果會受到洩漏效應與欄柵效應影響而產生誤差，雖然可以透過增加分析訊號取樣點數增加頻率解析度來改善，但會導致計算速度降低。
如何達到高解析度及提昇效能，又能囊括多種電力事件的分析，是本文的重點。基於上述考量，採用普羅尼法做為判斷電力事件的主軸，開發一電力品質事件監測辨識系統，既可囊括多種電力事件判斷，使用限制及運算速度都在標準之上。透過判斷電力品質事件的發生，系統可觸發相對應的自動保護機制以防止電力系統損壞造成大量損失。
;With the development of smart grid, the effective detection and identification of voltage variation events become important tasks for the protection of microgrid and power system monitoring integrated different distributed energy systems. When power quality events happened, these events usually caused large losses in commercial and industrial consumers are the main events in power systems due to the sensitivity of equipment to these voltage variations. In addition, in recent years, the widely applications of power electronics technique, and the amount of non-linear loads increased on power system, harmonic problem become more important.
There are a lot of analysis methods in tradition. For example, Zero-crossing is used to analyze the fundamental frequency, Fast Fourier Transform is used to analyze the harmonics, etc. Take FFT for example, although the simple calculation procedure and the high computational efficiency are the advantages of FFT, there exist some limitations needed to meet. If it didn′t, the aliasing, the leakage effect, and the picket-fence effect may occur. Though It could increase the sampling points to improve the frequency resolution, it will deteriorate the computational efficiency.
The keypoint of this thesis is high resolution, great efficacy and including analysis of multiple power events. According to aforementioned considerations, Prony method is used and plays an important role in analysis of power events. A SCADA of power quality events is developed. With advantages of multiple power events analysis, less limitation, better computational efficiency, power quality events can be detected correctly and activate the protection or compensation devices avoiding damage of power system. |
