Multi Classification of Static Security Assessment using Teaching Learning based Optimization enhanced Support Vector Machine

the assessment of static security is very essential for reliable and safe functioning of the power system. The primary drawbacks of standard power flow techniques are lengthy calculation time and large memory need. The composite security index has been implemented here to avoid the problem of masking involved with the previously used performance index for assessment of static security. Multi-classifier for assessment of static security in power system has been implemented utilizing support vector machine. The parameters of Radial Basis Function are selected using Grid Search method and Teaching Learning based Optimization. Sequential forward selection method has been used to select features which are much lesser and best features. The results of the proposed Teaching learning based binary classifier are validated using two IEEE standard test systems.