Naive Bayes based Intrusion Detection Scheme for False Data Injection Attack Detection in Smart Grid

Industrial control systems such as Smart Grids are becoming increasingly popular. However, there are some threats involved in the control systems. False data can be injected into the network during communication, leading to the belief that it is true data, causing some loss to the network that has been operating. While some attacks can be easily detected using traditional methods, False Data Injection Attack cannot be detected using traditional methods. These attacks not only cause the system to malfunction, but they can also result in the loss of human lives. In this work, machine learning models are used to predict the input data false data injection attacks. Various machine learning algorithms like logistic regression, Naive Bayes algorithm, Support vector machines, Decision tree, k-nearest neighborhood algorithm and Random Forest are implemented in python platform to find the best machine learning model for the data set considered. The data set is divided for training and testing. After the training and testing process, comparison between the models is done in order to find the best model. The best model so found is naïve bayes and is used to predict the new data whether the data is true data or false data.