Investigation of power system transient stability using Clustering Based Support Vector Machines and preventive control by rescheduling generators

In the deregulated environment of power systems, the transmission networks are often operated with reduced security margins. This seeks the development of reliable and faster security monitoring methods. Support vector machines (SVM), a neural network technology has been presented as an important contributor for reaching the goals of online transient stability assessment (TSA). To ensure secure operation of power system, some preventive actions are required when potential instabilities are detected. This paper presents a new clustering based support vector machine (CB-SVM) to identify the stability status of power system and the identified unstable cases are brought back to a more stable economic operating point by rescheduling the generators optimally. To maximize the benefit of learning the SVM, it is combined with fuzzy-c-means data clustering technique. The proposed method is demonstrated using New England 39 bus test system and the results are shown to be promising.