Methodology for Evaluation and Intelligent Prediction of Fire and Safety Systems in an Electrical Power Plant Using Artificial Intelligence

Abstract

The risk assessment conducted for HMN3 highlighted several critical areas requiring continuous monitoring and preventive measures. The application of structured methodologies enabled the identification of potential hazards and the evaluation of their likelihood and severity, providing a solid basis for improving safety performance within the plant. In addition, this work explored the integration of Artificial Intelligence technologies as an innovative approach to industrial safety management. The proposed AI-based system, relying on real-time data acquisition, advanced preprocessing techniques, and a Random Forest predictive model, demonstrated significant potential for early fault detection, predictive maintenance, and intelligent decision-making. Such technologies can substantially reduce accident probability, improve response times, and enhance the overall reliability of power generation systems. Ultimately, the transition from traditional reactive safety practices to predictive and intelligent safety management represents a major opportunity for modern power plants. The implementation of these advanced solutions at HMN3 and similar facilities can contribute to safer operations, reduced economic losses, and greater sustainability of the energy sector in Algeria and worldwide.