抄録
This paper presents fast load flow caluculation methods using a preconditioned technique. The load flow calculation may be expressed by solving a set of linear equations. A lot of studies on load flows have been made since Ward and Hale. The Newton-Raphson method is widely-spread due to the good convergence characteristics. The N-R method updates the Jacobian matrix and solves a set of linear equations at each iteration. On the other hand, the exact second-order load flow is attractive in terms of computational effort since it utilizes a constant Jacobian matrix. In this paper, the preconditioned conjugate residual and conjugate gradient squared methods are proposed to solve a set of linear equations in the load flow calculation. The precondition technique has an advantage that it can improve the convergence characteristics of the conjugate gradient type algolithms. Also, the algorithms are effective for solving a set of linear equations with large sparse coefficient matrices. The proposed methods are applied to the Newton-Raphson and the exact second-order load flow calculation. A comparison is made of the proposed and conventional methods in terms of computational effort. The effeciveness of the proposed methods are demonstrated in the IEEE 14, 30, 57 and 118 node systems.
