This study aims to develop an Intrusion Detection System (IDS) using deep learning and machine learning algorithms to detect cyber attacks in the network traffic of critical infrastructures using an artificial intelligence-based approach. The research investigates various machine learning algorithms, datasets, and performance evaluations to detect the security vulnerabilities commonly found in industrial networks. Implemented in Python, the system has been tested on hybrid dataset, demonstrating the performance of different algorithms in terms of accuracy, precision, and other metrics. From artificial intelligence perspective, this study contributes machine learning and deep learning in cybersecurity, showing how normal and ensemble models can effectively detect complex threats, with fewer features but more relevant. The research employs supervised learning techniques, leveraging labeled datasets to train models that can accurately classify network traffic as either normal or attack, ensuring high detection accuracy. From an engineering standpoint, the system's Python implementation addresses the practical challenges of real-world deployment in industrial control systems (ICS) and facilitates integration with existing infrastructures. Additionally, the custom dataset and post-dissector code contribute to the field of industrial cybersecurity, providing engineers with tools for testing, validating, and optimizing IDS solutions. As cyber-physical systems are increasingly integrated into ICS, the proposed IDS provides a crucial layer of defense against cyber threats, safeguarding both the digital and physical components of critical infrastructure. The findings reveal that the proposed system exhibits high performance in terms of detection accuracy. The results show that the system provides an effective and reliable detection mechanism using artificial intelligence techniques.