Real time spinning reserve allocation by a direct nonlinear interior point method

System spinning reserve constraint imposes an interdependency among real power generations of the units which are usually independent variables in optimal power flow (OPF). This paper presents a decoupling method for real time spinning reserve allocation problem (RTSRA) based on a direct nonlinear interior point method so that the constraints matrix of generation and that of network can be handled separately in OPF. A reduced Karush-Kuhn-Tucker (KKT) system is derived, whose scale is only determined by the network size itself and the influence of spinning reserve constraints can be taken into account by simply modifying corresponding entries of augmented Hessian matrix. The sparsity structure of the reduced KKT system is explored thoroughly to utilize the well developed super sparsity technique. Load curtailment, as another type of spinning reserve, is also considered in the proposed method.