The problem of the active and reactive optimum power dispatching solved by utilizing a primal-dual interior point method

This paper describes a procedure developed in ENEL S.p.A. (Italian Electric utility) to assess the coupled dispatching of active and reactive power for generation and transmission composite power systems. The problem is faced by utilizing a non-linens and sparse optimization model that explicitly takes into consideration the load-flow equations. For the solution of the optimization problem a non-linear extension of the primal-dual interior paint algorithm has been adopted. The linear system associated with each iteration of the algorithm is solved by exploiting its particular sparsity characteristics. The procedure, tested on several actual networks having various characteristics and sizes (up to 1500 nodes), can provide solutions of a higher degree of accuracy than the traditional decoupled methods of optimum power flow. In addition, considerations are made in view of a useful interpretation of the supplementary information provided by the Lagrange multipliers of the constraints supplied by the algorithm together with the optimum solution. (C) 1998 Elsevier Science Ltd. All rights reserved.