Hybrid Model for Enhancing Fault Detection System in Nuclear Power Plant using Support Vector Machine and Multivariate State Estimation Techniques

Nuclear power plants (NPPs) are vital for sustainable energy production but require meticulous monitoring to ensure safety and operational stability. This study proposes a hybrid framework integrating Support Vector Machines (SVM) and Multivariate State Estimation (MSE) to enhance real-time fault detection and system reliability in NPPs. While SVMs effectively classify operational anomalies, challenges persist in optimizing kernel selection and managing high-dimensional sensor data. Our approach combines SVM’s predictive capabilities with MSE’s dynamic state estimation, enabling early detection of deviations and improved decision-making. By addressing limitations in real-time data processing and complex nuclear environments, this method supports predictive maintenance and reduces operational risks. The study reviews existing techniques and demonstrates how merging these tools advances monitoring accuracy, safety, and efficiency in NPPs. This paper explored critical gaps in the application of Support Vector Machines (SVM) for real-time monitoring in nuclear power plants (Nuclear Power Plants) and proposed a hybrid approach that combines SVM with the Multivariate State Estimation Technique (MSET).