Optimal distributed generation placement and size under uncertainties and contingencies in active distribution networks

In order to describe the uncertainties, such as distributed generation (DG) output, load fluctuation and contingencies, in active distribution networks comprehensively, this paper proposes an optimal DG planning model in the presence of active management schemes. A correlated sample matrix of wind speed, illumination intensity and load is generated using quasi Monte Carlo simulation and singular value decomposition. Fuzzy C-means clustering is utilized to classify scenarios of the sample matrix to improve the computation efficiency of optimal power flow. The optimal allocation model is mathematically formulated as a bi-level programming problem, which is solved by a hybrid algorithm combining dynamic niche differential evolution and primal-dual interior point algorithms. The solution provides the trade-off between system economy and security. Case study demonstrates the effectiveness of those techniques.