The framework of an automatic discovery system for critical flowgates and security operation rules in the power grid is designed. Based on the real-time operation data of the power grid

the system is able to automatically find the critical flowgates adapted to the online operation state of the grid. By using the automatic learning method

the system can find the fine rules of critical flowgates as the security operation knowledge of the grid. The functions of the core modules of the system are described

including online automatic discovery of critical flowgates

online calculation of limit transmission capacity of critical flowgates

online generation of mass simulation samples

feature selection

automatic discovery of security operation rules of the power grid. The online application of the system has been realized in Guangdong power grid. The application effect shows that the system is able to generate the security operation knowledge fast enough and refresh it every 15 minutes for online application as required. Meanwhile

the knowledge is capable of not only covering the existing security operation knowledge formulated by experts offline

but also finding new knowledge adapted to the online power grid operation mode

while improving the power grid security and economy.