Power System State Estimation

The state estimation of the power system is an important component of the energy management system (EMS), especially playing an even more important role in the electric market environment. Though the theory of state estimation in power system had right away established in the early of the 1970s, improving constantly in the nearing more than ten years. But power system state estimate have a lot of fields that worth study further still. For example, to specific system, it is still worth to find a method of computing accurately, steady and fast. This paper researches the subject of the observability analysis algorithm and state estimation algorithm of power system. At first, this paper reviews three kinds existing methods of states estimation observability analysis --topological, number and mix method, emphasing the method of mix model based on the solvable condition of power flow. This paper present the concept of state variable of flow island based on the fact that there is not more than an unknowned complex voltage in a flow island. It can combinate successly flow islands according to several rules based on the solvable condition of power flow using reduced Jacobian matrix. Several kinds of classical state estimation algorithms are proposed in the paper, their use in different systems analysized. The standard which estimates the stability of algorithms is introduced. This paper presents a weighted least squares method for state estimation based on block Jacobian matrix. In the proposed method, whole injection measurements, highly weighted virtual measurements and a few essential flow measurements make up measures which can be used for resolving all state variables of a power system, while other flow measurements are considered redundant to resolve state variables. In the following, the Jacobian matrix is divided into two blocks. The first block matrix is diagonally dominant positive definite matrix, the other block matrix does not include injection measurements that induce numerical morbidity.Thus the proposed arithmetic can reduce the loss of information brought from multiplying Jacobian matrix and calculation of multiplying matrix and enhances the numerical stabilization. It is remarkable that this method can retrain the degree of numerical morbidity caused by highly weighted virtual injections.