Decompositions for optimal power flows

Abstract

Two major difficulties are experienced in most of the optimization algorithms: computational inefficiencies associated with large power systems, and problems regarding handling functional inequality constraints. The Han-Powell method has been proved to be extremely fast, robust and efficient for the small optimal power flow problems but balks at full size problems. The method can be extended to large size problems provided an efficient decomposition method is developed to overcome the problem associated with large power systems. In view of this an efficient reliable, simple and more realistic decomposition method developed by Ganga Agnihati et al. (1988) is used which overcomes the problem of large size power networks and solves the large power network using Han-Powell method. This decomposition method decomposes the large size power network into a number of blocks from the weak coupling points, and the boundary buses are modeled as swing buses. >