As cyber-physical systems (CPS) gain prevalence in sectors such as manufacturing, transportation, and critical infrastructure, ensuring their security and reliability is paramount. Traditional anomaly detection methods often fall short due to the dynamic and complex nature of CPS, leading to missed or false alarms. This study introduces an enhanced machine learning model that integrates statistical and deep learning techniques for real-time anomaly detection in CPS. By employing a hybrid approach of convolutional neural networks (CNNs) with statistical pattern recognition, the model demonstrates improved detection accuracy and responsiveness. Performance is evaluated using industry-standard CPS datasets, showing that the proposed model outperforms existing techniques in both accuracy and efficiency.
