利用廣域同步相角量測與分散式計算技術進行智慧電網之即時電壓穩定度評估

Applications of Wide-Area Synchrophasor Measurements and Distributed Computation Techniques for Real-Time Voltage Stability Assessment in Smart Power Grids

劉建宏(Jian-Hong Liu)

指導教授 ： 朱家齊

國立清華大學/電機資訊學院/電機工程學系/博士(2014年)

摘要

隨著電網供電日益趨緊與大量的再生能源併入電網，電壓穩定度評估問題已成為監測與控 制目前電力系統安全性的關鍵議題。基於電網的安全考量，系統操作者需要仰賴可靠的電壓穩 定度評估工具，以即時分析電網的電壓安全度。傳統上，電壓穩定度評估法通常採用潮流方程式為基礎的解析法，雖然大多數的解析法已能精確地評估電網的電壓穩定度，但是其計算複雜 度卻成為應用於即時環境上的阻礙，近年來，隨著廣域同步相角量測技術與分散式計算技術的進步，這些新型技術開啟了發展即時電壓穩定度評估法的新方向。因此，本論文的目標將採用廣域同步相角量測技術與分散式計算技術，發展新型的即時電壓穩定度評估法。
首先，本論文將提出一修正型耦合單埠模型，用以改善現有耦合單埠模型在即時電壓穩定度評估上的低估情形。藉由負載的即時同步量測資訊，電網的擴增Ward-type等效電路中的虛功響應可定義出現有模型與實際電網的虛功不匹配程度，修正型耦合單埠模型再以此虛功響應定義一緩和因子，此緩和因子可做為修改現有耦合單埠模型電路參數之依據，以此修改完成修正模型之建構。此外，基於提出的修正模型，我們也發展具應用性的即時電壓穩定度指標。
之後，再提出一基於同步相角量測之短期電壓穩定度指標，用以即時偵測感應發電機之短期電壓不穩定。短期電壓不穩定現象很可能使得感應發電機過度加速，以致其轉速過高而無法回到原先的穩定平衡工作點。此短期電壓穩定度指標是藉由合併感應發電機之等效模型與修正型耦合單埠模型所建構而成。
此外本文將提出一可即時評估電網可傳輸容量的疊代型分散式演算法。三種特定的分散式演算法將應用於發展此疊代型分散式演算法，分別為(i)預測與修正型近似乘數法、(ii)輔助問題原理法與(iii)交替方向乘數法。疊代型分散式演算法可實現在一分散式系統中，電網可經由結構性的切割，分割為多個不重疊的子系統以及與其連結的邊界子系統之分散式系統，再以一平行運算架構計算電網的可傳輸容量。
最後，本論文將分析大型風能注入電網所造成的機率性負載裕度變化，因此擬提出一修正型直接法，同時結合C級級數機率分析法，用以分析負載裕度的機率分佈，此修正型直接法將分兩階段開發，首先研究一分散型雙向滑動法，採用分散式計算法計算連結多部風電系統的電網潮流解，第二階段再以一修正型直接法取代分散型雙向滑動法中的潮流計算步驟，以評估電網的負載裕度，因修正型直接法可以較少N + 1維度取代舊有2N + 1維度的負載裕度計算，所以具有應用於即時計算環境之優勢。本論文提出的即時電壓穩定度評估法將測試於多個典型的IEEE測試系統中，用以驗證提出之評估法的可行性與精確性。

關鍵字

修正型直接法 ； C級級數 ； 電壓穩定度 ； 耦合單埠模型 ； 虛功響應 ； 感應發電機 ； 短期電壓穩定度 ； 修正型耦合單埠模型 ； 可傳輸容量 ； 分散式演算法

並列摘要

As the power system becomes more stressed and the penetration of renewable energies increases, voltage stability assessment (VSA) becomes a key issue for monitoring and controlling the security of modern bulk power grids. For security considerations, system operators require powerful tools to analyze voltage security of the bulk power system in real-time environments. Traditionally, the VSA is accomplished based on model-based approaches. Although various analytical approaches have been proposed along this direction, their computational complexities may impede their real-time applications. In recent years, with advances of wide-area synchrophasor measurements and distributed computation techniques, these new technologies have opened new perspectives for developing real-time tools for VSA. To this end, this dissertation aims to develop new methodologies for enhancing the real-time voltage stability by using wide-area synchrophasor measurements and distributed computation techniques.
First, based on real-time PMU measurements of individual load bus, a modified coupled single-port model will be proposed for measurement-based VSA. This model will improve underestimations of existing coupled single-port models since the reactive power response extracted from the extended Ward-type equivalent is explored to compensate the reactive power mismatch in the existing coupled single-port model. Then, a mitigation factor based on this reactive power response will be defined to provide a direction for adjusting circuit parameters of the current model, and modified models can be constructed accordingly. In addition, based on this modified coupled single-port model, several voltage stability indicators are developed for real-time VSA.
Second, the phenomenon of the short-term voltage instability of induction generators will be investigated. This phenomenon will lead to over-accelerations of induction generators such that the induction generators with the high slip may not return to the pre-fault equilibrium point. In order to identify this short-term voltage instability of induction generators in real-time manners, a synchrophasor-based short-term voltage stability indicator will be developed by incorporating induction generator equivalent models into modified coupled single-port models.
Third, multi-area Available Transfer Capability (ATC) assessments will be investigated in the distributed computation environments. Three distributed schemes, including (i) Predictor-Corrector Proximal Multiplier Method, (ii) Auxiliary Problem Principle Method, and (iii) Alternative Direction Multiplier Method, will be applied to distributed ATC assessments. System partition with non-overlapping and boundary sub-systems will be employed to a distributed system for implementing proposed iterative distributed algorithms in a distributed manner.
Finally, probabilistic load margin predictions under large-scale penetration of wind generations will be studied. Under wind speed variations, a new computational framework will be proposed to conduct probabilistic load margin estimations. A distributed bi-directional sweep method will be employed in multiple wind generators connected to the main grid for power flow computations. A modified direct method will be presented such that conventional 2N + 1 non-linear equations can be replaced by N + 1 equations for load margin calculations. Accordingly, corresponding probabilistic estimations can be calculated by integrating both modified direct method and Gram-Charlier expansions. Simulations on several IEEE test systems have been used to validate the feasibility and the accuracy of our proposed techniques.

並列關鍵字

Modified Direct Methods ； Gram-Charlier Expansions ； Voltage Stability ； Coupled Single-Port Circuit ； Reactive Power Response ； Induction Generators ； Short-Term Voltage Stability ； Modified Coupled Single-Port Models ； Available Transfer Capability ； Distributed Algorithms

參考文獻

[1] IEEE/CIGRE Joint Task Force on Stability Terms and Definitions, “Definition and classification of power system stabilit,” IEEE Trans. on Power Systems, vol. 19, no. 2, pp. 1387–1401, May 2004.

[2] C. A. Canizares, F. L. Alvarado, C. L. DeMarco, I. Dobson, and W. F. Long, “Point of collapse methods applied to AC/DC power systems,” IEEE Trans. on Power Systems, vol.

[3] V. Ajjarapu and C. Christy, “The continuation power flow: a tool for steady state voltage stability analysis,” IEEE Trans. on Power Systems, vol. 7, no. 1, pp. 416–422, February 1992.

[4] I. Dobson, “Computing a closest bifurcation instability in multidimensional parameter space,” Journal of Nonlinear Science, vol. 3, no. 1, pp. 307–327, 1993.

[5] C. A. Canizares, “Calculating optimal system parameters to maximize the distance to saddle node bifurcations,” IEEE Trans. on Circuit and Systems-I Fundamental Theory and Applications, vol. 45, no. 3, pp. 225–237, March 1998.