The electricity industry must be restructured due to the rising trend of electric power demand, resource restrictions, and the degradation of existing grid infrastructure. Besides, in addition to the numerous advantages, the introduction of Internet of Things (IoT) technology and the conversion of the electrical grid to a Smart Grid (SG) creates security concerns. This study aims to incorporate Machine Learning algorithm-based Intrusion Detection System (IDS) to battle cyber-attacks, as this is one of the ways ahead in detecting and mitigating security attacks. A Random Forest classifier is employed in this method to detect intrusion, and the proposed method's performance is compared to the XGBoost Algorithm. The suggested method outperforms other methods for Intrusion Detection in IoT-based SG in terms of Accuracy and Precision, according to experimental data.TheXGBoost classifier has an 80 percent accuracy rate, while the Random Forest classifier has an 83 percent accuracy rate. Similarly, the calculated precision rate for XGBoost and Random Forest models is 81 percent and 84 percent, respectively.