| As the key element in power management system, power system state estimation is indispensible. As a result of the ever-growing demand of safe transmission through a more complicate network, it is much more important for the dispatchers to know exactly what the state of the system is. So some improvements have to be made in order to enhance the performance of power system state estimation.Consisted of GPS and phasor measurement unit, wide area measurement system provides a brand new method for state estimation. Since there can be data with high accuracy, once introduced into the procedure of state estimation, it can greatly improve the speed and convergence rate. This paper introduced not only the PMU measurements but also the calculated values of the nodes which are connected to the one that has PMU. According to the error theory, these calculated values are still relatively highly believable.While enlarging the redundancy of measurements, the PMU measurements make the dimensions of the matrix bigger as well. This is extremely observable when there are a large number of PMUs. One most negative effect is that the iteration time has been prolonged. Therefore the placement of PMUs with a certain number should be studied.This paper uses the minimum spanning tree algorithm in the graph theory to configure the PMUs.'Finding the bus that cover the network most' is taken as the goal of the MST. Since there can be several zero-injection nodes in the system, a recursive MST algorithm is proposed in order to pinpoint the nodes and delete them in the optimal configuration sets found by MST.The algorithms mentioned above are programmed in the MATLAB. From the result of the simulation of IEEE-14 and IEEE-30, the RMST algorithm can effectively carry out PMU placement plans with more sets. And meanwhile, the zero-injection nodes are correctly deleted in advance. As for the state estimation algorithm, which makes the most use of PMU measurements, its convergent speed and accuracy are improved.
|
PDF Full Text Request