Research Of Different Measurement Weights In Power System State Estimation

State estimation is an important part of the energy management system (EMS). With grid scale and the level of the power system automation increased year after year, which requires higher computing speed and more accurate calculation of power system state estimation.This paper makes a deep study of state estimation theory and algorithms, which has a comprehensive analysis of the factors on the impact of state estimation accuracy, especially focusing on the weight setting. The reasonable set of measurement weights has a great influence on state estimation calculation accuracy, and the traditional measurement weight setting method is generally selecting the reciprocal of the measurement error variance as weights, without considering the different types of measurements. This thesis has a deep study and discussion of the different types of measurements on the impact of state estimation calculation accuracy.Based on the perturbation analysis on the states affecting the sensitivity of different types of measurements, it can find out that the states have a lower impact on the sensitivity of the branch power measurements; the states have a higher impact on the sensitivity of the voltage measurements and bus input power measurements. When making a qualitative analysis of network equations, it can find out that the state variables can be equivalent to the bus in the network, thus analyzing different types of measurements impact on the states. This paper proposes a new measurement weights setting method:partially use the voltage measurements in the state estimation; fully use the branch power measurements in the state estimation. And weigh different types of measurements coordination, making low weights of voltage measurements and bus input power measurements, making high weights of branch power measurements.By simulating IEEE4system, the correctness of the theoretical analysis can be verified when calculating the deviation of the states by setting the error of the different measurements. By simulating IEEE30system, an improvement of the weights setting method for different types of measurements is made out, which can certify the weight setting is reasonable.