臺灣博碩士論文加值系統

摘 要
平行處理具有快速計算能力，因此被應用於負載潮流計算。
邊緣方塊對角形式的賈可比矩陣相當適合於平行負載潮流計算。
然而在求解邊緣方塊對角形式矩陣的切集分塊時，
仍須經過反覆傳值與疊代計算，因此求解是相當困難、費時。
為改善上述缺點，本文提出應用相量量測單元以平行求解連續
負載潮流的新演算法。本方法先應用量測單元量取切集分塊
的狀態變數，經靈敏度因子法判斷量測值後，將其視為切集分塊
的狀態變數解；再配合多微處理機以平行求解各子分塊的狀態變數。
最後模擬測試台電 及IEEE測試系統，得知本方法確實相當適用於
求解連續負載潮流。

ABSTRACT
An application of parallel processing to the load flow problem
is for
reducing the computation time. The bordered block diagonal form
(BBDF) matrix of the Jacobian matrix is quite suitable for
parallel processors
implemention. However, effectively solving the cutset block of a
BBDF matrix through iteration and transfer in parallel is a
relatively difficult and time-consuming task. To improve
the above method, a novel parallel algorithm based on
applying Phasor Measurement Units (PMUs) is proposed in this
thesis for solving the continuation load flow problem. For
identifying the measurement value of the PMUs, the
sensitivity factors method is used. The varying subset can
be solved by parallel processing. Simulation results of the
Taiwan Power Company and IEEE test systems are presented to
illustrate the performance of the proposed method and its
capability to implement parallel processing on the
continuation load flow problem.