An error reducing algorithm of Innovation Graph Technique

• Published in: 2007 International Power Engineering Conference (IPEC 2007) pp. 177 - 182
• Main Authors: Suquan Zhou, Yanjun Zhang, Tianyu An, Yongjie Leng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2007
• Subjects: Artificial neural networks; Electric variables measurement; Error analysis; Generalize bus; Large-scale electric network; Large-scale systems; Network partitioning; Network topology; Noise measurement; State estimation; Switches; Technological innovation; Velocity measurement
• ISBN: 9810594232; 9789810594237
• ISSN: 1947-1262; 1947-1270

This paper presents a method which reduces error of the innovation graph technique that researches power system state estimation. This method makes sure that Innovation graph technique can identify multiple bad data, topology changes and topology errors more accurately in large-scale electric network. First, the innovation graph technique is introduced and the error source of the innovation graph technique is analyzed. Second a method which can reduce the error of the innovation graph technique by network partitioning is proposed. At last, the merits and demerits of the method which can reduce error are analyzed. Based on innovation graph technique, a simplified network which equals to the original network is constructed and then the simplified network is used to calculate reckoned innovation of boundary branches in each sub-network. The reckoned innovation of sub-network boundary branches are used to correct the reckoned innovation of sub-network inner branches. The method reduces error superposition in the innovation graph technique, improves the accuracy of branch reckoned innovation, and blocks the error diffusion among subnetworks. The method improves the ability of the innovation graph technique for identifying bad data, topology errors and topology changes. The method can be applied on real-time large-scale electric network.