<p>This study presents a novel framework for forecasting volcanic activity by integrating an intelligent prediction system within a three-dimensional distributed mobile wireless sensor network (3D-MWSN). Recognizing the severe risks posed by volcanic eruptions to surrounding communities and infrastructure, the system is designed to enhance forecasting accuracy and support disaster preparedness. Leveraging advanced machine learning algorithms and real-time data acquisition, it analyzes diverse volcanic indicators to deliver highly accurate eruption predictions. The framework addresses two key challenges in volcanic monitoring: achieving robust prediction in dynamic, hazardous terrains and ensuring the limited energy autonomy of distributed sensors. To overcome these challenges, we propose an innovative energy harvesting mechanism based on power splitting technology, enabling long-term operation in remote areas. Unlike conventional monitoring systems, which rely on static sensors and fragmented data, the proposed 3D-MWSN provides dynamic, spatially rich and real-time observations. Experimental evaluations on real and simulated volcanic datasets highlight that the system improves forecasting accuracy and reduces false alarms. This work demonstrates that integrating mobile wireless sensors with intelligent prediction not only advances the state of the art in volcanic monitoring but also offers a sustainable and deployable solution for proactive disaster risk reduction in volcanic regions.</p>