Search

A power grid fault diagnosis method is proposed based on the graph signal modeling according to the fault recording start information and the network node eigenvector centrality algorithm. Firstly, the fault start information network diagram is constructed from the fault record start information of relay protection and fault information master station. Secondly, the graph smoothness analysis method is used to determine whether the grid fails and identify the fault type. Finally, the fault components are identified based on the feature vector centrality algorithm of network nodes and visualized. The simulation verifies the effectiveness of the proposed algorithm. As a part of the intelligent fault analysis of wave recording, this method helps to quickly grasp the fault situation of power grid, and focus on collecting wave recording data for fault analysis and verification, so as to improve the efficiency of wave recording analysis.

The removal of one of the transmission lines will cause power flow transfer. The transferred load of for the affected lines is different. According to the degree of the power flow transfer and using the flow transfer coefficient as the weight, this paper proposes an adaptive distance backup protection scheme. Based on the real-time flow distribution measured by the wide-area measurement system (WAMS), the flow transfer coefficient can be calculated. Then, using the coefficient as a weight, the degree of the setting impedance shrinkage can be adjusted. This scheme could not only prevent maloperation, but also identify the fault and it has the ability of enduring transition resistance. The simulation results on an IEEE 39-bus testing system illustrate the validity of the scheme.

In intelligent substation, the processes of data collection, logic processing, tripping process and GOOSE, SV and MMS messages among multiple devices are invisible. A kind of novel Multi-IED devices simulator based on IEC 61850 for intelligent substation is developed to simulate the behaviors devices. The system includes the visual interface program, main scheduling program of simulator, multiple IED modules, SV, GOOSE and MMS communication interfaces, and SV simulating program. The devices use configuration information in SCD to configure the operating parameters for each device. The parallel simulation method for multiple IED devices and transmission of SV, GOOSE and MMS among devices are given. The experimental results show that the simulator realizes the parallel simulation for Multi-IED as well as communication simulation of SV, GOOSE and MMS messages. The simulator system has applied to the practical training, has reduced the training costs and improved training efficiency.

Power systems get more unstable and insecure with more transmission congestion and smaller generation reserves. There is prospect of wide area control schemes meeting the challenge instead of traditional local ones. Communication network technologies, which have made a great stride during last decade, play an important role in the new schemes. Control systems based on wide area networks (WAN) are studied in this paper. Multi-agent concept is introduced to improve the control performance. A novel wide-area control system that utilizes phasor measurement unit (PMU), wide-area communication and multi-agent technology is presented. The distributed system architecture and multi-agent working mechanism are discussed in order to use phasor data and to realize optimal power system stability control.

In the traditional transformer fault monitoring system, the LPU (Local Process Unit) usually sends the data to the MMC (Main Monitoring Computer), which completes the complicated fault diagnosis and is the process center. In this paper, a novel transformer monitoring system is presented. The LPUs, the embedded systems, are designed to possess of the approximately functions to MMC such as monitoring, diagnosing and alarm. The TCP/IP and the Pub/Sub (Publish/Subscribe) technology are used to realize the design of LPU. The TCP/IP protocol and the network chip make the LPU with high transmission speed. The complicated diagnosis task is dealt with by a computer with strong data treatment capacity. The LPU and MMC can subscribe some fault diagnosis subjects.