This paper presents a comprehensive analysis and fault diagnosis approach for wave energy conversion (WEC) systems, specifically focusing on point absorber technology, using Bayesian Networks (BNs). The main objective of this work is to develop a probabilistic framework that enhances fault detection and diagnosis by modeling the interdependencies between key subsystems, including the power take-off (PTO) mechanism, mooring lines, and electrical components. Wave energy conversion systems offer a promising solution for sustainable energy generation, but fault detection remains a critical challenge in ensuring continuous and efficient operation. The proposed approach enables a probabilistic evaluation of failure modes and their impact on overall system performance by modeling the complex interdependencies between system components. By integrating environmental factors, historical failure logs, and operational data, the Bayesian network allows real-time dynamic updates of fault probabilities, facilitating predictive maintenance techniques. The proposed approach aims to improve system reliability, reduce downtime, and optimize maintenance strategies. Case studies are provided to validate the approach, demonstrating significant improvements in early fault detection. The results underscore the potential of Bayesian networks as a powerful tool for enhancing the operational resilience and sustainability of wave energy conversion systems.