Multiple Regression Based Classifier for Static Security Assessment

Abstract—Power System Security is becoming an important part in the planning and operation studies. Security Assessment is the process of determining whether the current operational state is in a secure or emergency (insecure) state. Security evaluation performed by simulation program involves long computer time and generates voluminous results. This led to the development of ANN based approaches. This paper presents a practical and feasible Pattern Recognition (PR) approach for static security assessment in power systems. The feature selection stage uses a sequential method called single ranking method to select the best feature set from a large set of variables. The security function called classifier is designed by Multiple Regression technique and Perceptron Learning Rule. It is proved that Multiple Regression gives a better classification and minimizes the number of classification errors compared to the latter. The methodology of the PR approach is presented in detail. The performance of the algorithm is tested on IEEE 6 Bus, IEEE 14 bus and IEEE 30 Bus Systems. High recognition rates of security have been identified. Contingency ranking has also been addressed to meet the overall security assessment. I