Distribution network adaptive protection system based on artificial intelligence

Abstract

With the continuous development of society, the requirements of enterprises and the public for the reliability of the power system are also improving, and the fault identification and protection action of the distribution network is conducive to the comprehensive analysis and overhaul of the line, which is of great significance. In this paper, a short-circuit fault identification method for distribution networks based on artificial intelligence technology is proposed to categorize and predict power system faults using an association rule algorithm. Then, a distribution network adaptive protection strategy based on the equivalent impedance of the distributed energy system is proposed, which real-time adjusts the setting value and criterion of the protection through the distribution network containing distributed energy devices connected to the grid and adjusts the working status in normal operation, realizing that the current protection of the distribution network adapts itself to the change of the working conditions of the distribution network system of IIDG. Simulation results show that when two-phase grounding and three-phase short-circuit faults occur in the distribution network under the protection of the adaptive system. The protection circuits start instantaneously in 0.005s, which realizes the functions of the fast startup, self-sustained operation, and controllable shutdown when faults occur in the distribution network. Meanwhile, the feasibility of the scheme is verified by utilizing the system to output short-circuit faults in the case of IIDG grid-connected or off-grid when the respective integrating values are greater than the positive sequence short-circuit current values.

Keywords

• distribution network
• artificial intelligence
• association rule algorithm
• classification prediction
• distributed energy
• equivalent impedance
• adaptive protection
• 68m12