Improved power system health state monitoring method and system

Abstract

The invention proposes an improved power system health state monitoring method and a system, belonging to the power system monitoring field. According to the invention, based on the installation nodes of a PMU device in a complicated power system, a power grid is divided into a linear area and a non-linear area. If there is some node provided with the PMU device, then, the node is regarded as the linear area that can be described by a linear model and can be solved directly. For the rest nonlinear area, the PMU and SCADA measurements are combined as a training sample; and an artificial neural network intelligent algorithm is utilized to estimate the running state of the power system. According to the invention, the algorithm applies the BP neural network algorithm with the homotopy algorithm to the state estimation of a power system, which avoids the establishment of a mathematic model and the various nonlinear iterative operations; and at the same time, the fault tolerance is not restricted by the ill-condition. The use of the neural network after sample training only requires seconds before the state amount result is achieved. In addition, the precision is high. With the method and the system, the shortcoming of the prior art regarding to the high cost and difficulty of technology can be overcome.

Description

technical field [0001] The invention relates to the field of power system monitoring, in particular to an improved method and system for monitoring the health state of the power system. Background technique [0002] The traditional power system operation state monitoring method is to use the real-time measurement and pseudo-measurement data obtained by the data acquisition and monitoring system (SCADA) to obtain the best estimated value of the system state variables through the iterative method of solving nonlinear equations. However, the problem with this method is that on the one hand, traditional devices are not synchronized in time and not wide-area in space. There is no unified time coordinate between different installation locations, and there is no accurate common time mark. The recorded data is only partially valid, and the analysis of the dynamic characteristics of the entire system is difficult to complete. On the other hand, with the rapid increase in the complex...

Images