The Innovation Graph Parameter Estimation In The Observability System Of PMU Measurement

With the development of the economic and the technology, it is the requirement of the future to develop smart grid based on the technology of moden information and communication. Smart grid put forward higher requirement of parameter estimation in power system. The appearance of PMU provides an important source of data for the safety, stability of the power system operation. So it is supposed to use the PMU data to improve the result of parameter estimation in power system. While as the result of the limit of economic and technology, the configuration of the PMU in out country still cannot reach the need of observability. Thus the researches concern with the using of limited quantity of PMU in the parameter estimation have important practical meanings.Baesd on the theory of innovation graph, this paper researches the observability of innovation graph technique in the AC load flow model firstly. To the unobservable network, we can use KCL to think of it as a general node, and establish the observability system of PMU measurement, which is observable for parameter estimation by use of the PMU measurement.The issue of whether the innovation graph technique is still applied is analyzed.The identifiability of the bad data and topology error is still applied in the observability system of PMU measurement.The identifiability of parameter error decrease the range of the branch which is identifiable because of the general node.The concept of non-boundary loop is given, the parameter error can be identified and estimatied in the non-boundary loop.The principle of innovation graph parameter estimation is depth studied. The situation of two parameter errors is discussed. Utilizing the innovation voltage sum which calculated by link deduction innovation is the main judgement for parameter errors. The correlation of parameter error and the classification mode of parameter errors is presented. This method can identify most situation of two parameter errors.The results of innovation graph parameter estimation by the observability system of PMU measurement and by SCADA system have been studied for comprehensive identification. According to the features and the distribution of PMU data and SCADA data, the principles to confirm each other of innovation graph technique to identify bad data, topology error, and parameter error is given to improve the accuracy of the result of parameter estimation.This research work is supported by the National Nature Science Foundation of China(Project Number : 50977017) and State Grid Corporation of China's technology project"The development and application of Innovation graph parameter identification and state estimation".