Title :
A fuzzy clustering method for coherent generator groups identification based on A-K
Author :
Liu Tianqi ; Wen Jun ; Liu Xuan ; Xingyuan, Li
Author_Institution :
Sch. of Electr. Eng. & Inf., Sichuan Univ., Chengdu, China
Abstract :
A coherent groups recognition method using fuzzy clustering method based on A-K networks is proposed. Firstly, a fuzzy similarity matrix is formed by applying maximum-minimum algorithm. Then the A-K networks are trained with each row of the fuzzy similarity matrix as inputs. The nerves of output layer which win ultimately represent different dynamic styles. Finally, it is tested on the EPRI-36 bus model of PSASP. The results based on A-K fuzzy method are more similar to the results based on time simulation compared to A-K method. Moreover, A-K fuzzy method can identify coherent generator groups in greater time range.
Keywords :
ART neural nets; electric generators; electric machine analysis computing; fuzzy set theory; learning (artificial intelligence); minimax techniques; pattern clustering; self-organising feature maps; A-K network training; ART network; EPRI-36 bus model; Kohonen neural network; Kohonen self organizing neural network; PSASP; coherent generator group identification; fuzzy clustering method; fuzzy similarity matrix; maximum-minimum algorithm; Artificial neural networks; Clustering methods; Fuzzy neural networks; Fuzzy systems; Neural networks; Pattern recognition; Power system dynamics; Power system modeling; Power systems; Subspace constraints; Coherency identification; Coherent generator groups; Dynamic equivalent; Fuzzy clustering; Fuzzy similar matrix; Self-organization;
Conference_Titel :
Sustainable Power Generation and Supply, 2009. SUPERGEN '09. International Conference on
Conference_Location :
Nanjing
Print_ISBN :
978-1-4244-4934-7
DOI :
10.1109/SUPERGEN.2009.5348289