Modern power system networks integrate communication systems to ensure efficient and reliable function of the power grids. The distance relay is an essential component of the power transmission system. With the rapid progress of communication technologies, distance relay systems are susceptible to cyber- attacks. This article presents a machine learning-based cyber attack detection model for distance relay. The model is capable of detecting FDI (False Data Injection) and FSI (False setting Injection) attacks. The proposed methodology begins by processing the current and voltage data. It then employs Principal Component Analysis (PCA) to calculate the eigenvalues and eigenvectors. This data is subsequently used to train and test the Random Forest ensemble algorithm for classifying faults, normal conditions, and attacks, and to generate labels. The proposed method’s efficacy is evaluated utilising the IEEE-9 bus system.