With the development of smart grid technology and information technology in recent years

the possibility of cyber attacks in power system increases. A bi-level nonlinear optimization model of false data injection(FDI)attack is proposed based on the state estimation. In the upper level

the cyber attackers hack the measurement data of the power system with the purpose of finding the optimal attack strategy to maximize the economic loss of the power system while subjected to a list of constraints including the measuring attack range and state estimation residual error. The security-constrained economic dispatch model is used in the lower level and the power system operators perform the optimization dispatch of power system based on the load data disposed by the state estimation. In order to handle the complexity of the bi-level optimization model

the proposed model is transformed to single-level nonlinear planning model with KKT(Karush-Kuhn-Tucker)conditions. The programming simulation of nonlinear FDI attack model is implemented in GAMS and solved by using the nonlinear planning solver of BARON. The numerical results indicate that the bi-level FDI optimization attack could seriously deteriorate the secure and economic operation of power systems

and the effectiveness of the proposed approach is consequently validated.