Neural network method based on PMU data for voltage stability assessment and visualization

Voltage phase angle measurement between two points provides the indicator of the system state. This parameter as well as voltage magnitude have been used to identify how stress the system while load increased. In this paper we proposed a new technique for voltage stability assessment based on the Phasor Measurement Units (PMU) data. Firstly, a new method to visualized the difference of voltage phasor angle are presented. This method will be visualized the level of stress and proximity to voltage stability limit in the system. The indicator of stress in the system can be visualized by change of spectral frequency. It can be also be used to monitor such loss of generators or transmission line outages. Secondly, neural network data training which generated by load flow simulation are developed. The network actually represents the process of index stability calculation in the system from limited number of PMU in the system. The index stability are used for boundary stability limits while the voltage angle visualization in the same time provides continuous monitoring of entire system. This studied used 14 bus IEEE systems for testing the method. The result shown that the proposed technique is simple, fast and reliable, requiring minimum data measurement and be suitable for Wide Area Monitoring System.