Optimal Reactive Power Planning Using Bender's Decomposition Technique And Interior Point Method

As one part of power system planning, the purpose of reactive power planning is to optimize sizes and locations of Var sources, so that the security and economic operation of power systems can be maintained, i.e. the investment of Var compensation equipments and the operation cost of power network can be minimized. From the view of mathematics, reactive power planning is a mixed-integer programming problem, and from the view of practical meaning, it includes two types constraint conditions and variables, which are investment and operation constraint conditions and variables, and they have coupled relation. So reactive power planning problem can be decomposed into the investment subproblem and the operation subproblem using Bender's decomposition technique, then it can be solved through two subproblems' iterative calculation.On the basis of formers' work, in this thesis reactive power planning problem is decomposed into investment and operation subproblems using Bender's decomposition technique, and the operation subproblem is solved using prime-dual affine scaling method. In this thesis, the selection of initial values and the dynamic modification of step size are discussed, and IEEE testing system is used to reactive optimal power flow calculation. The results show that prime-dual affine scaling method has stable iteration and convergence.In this thesis Bender's decomposition technique and prime-dual affine scaling method are used to solve reactive power planning of Anshan 220kV and 66kV system, and it is shown that this method has the ability to solve reactive power planning problems of large scale power systems. And the results are fairly satisfactory.