نهمین کنفرانس بین المللی فناوری و مدیریت انرژی

رفتن به صفحه اصلی

ارائه مجازی مقاله

کد مقاله icemt9-1188

عنوان Employing Machine Learning Algorithms to Identify False Data Injection in Smart Grid

نویسندگان Mohsen Tajdinian - Mostafa Mohammadpourfard - Ali Goodarzi

چکیده Enhancement in demand of electrical energy from industrial or residential consumers has resulted in encouraging policy makers to utilize of new generation of power network named as smart grid. It is obvious that cost of overall monitoring will be decreased as a result of making smart grid, however it may result in enhancement of risk of cyber-attacks. Stealth attack or false data injection which has been recently introduced has become one of the important challenges of smart grids. This is because conventional bad data detection methods are not developed with regard to stealthy attacks. This paper presents a new algorithm which demonstrates that under false data injection condition in power systems, state of system is clearly individualized from the normal state. Stealthy attack detection has been carried out utilizing supervised machine learning based techniques. In all employed machine learning techniques, this paper proposes a new preprocessing algorithm to decrease the number of the features that need to be processed. Therefore, computation complexities are reduced substantially. The proposed method has been evaluated on the IEEE 14-bus standard test system. The results show the effectiveness of proposed method in detecting stealthy attacks.

محتوای ارائه مجازی مقاله

نظرات