Hybridization of Photovoltaic and Thermoelectric Energy Harvesting Systems at Seismic Nodes

Seismic data is one of the most important data for analysis and interpretation of subsurface, but remote seismic node fails frequently. This is one of the main constraints that limits seismic network to acquire continuous seismic data and near real-time prediction of seismicity of an area. There are several ambient energy sources at the remote seismic nodes. Optimization of their energy transducers and DC-DC converters is inevitable. In this study, solar and thermal sources are utilized with maximum power point tracking (MPPT) algorithm. The algorithm is based on the Neural Network model to supply the duty cycle across the converter optimally. Historical data were generated from Perturb and Observe algorithm in PSIM for the Neural Network Model to train the data and predict the duty cycles for both within and outside the data. The proposed system delivered an over 75% conversion rate of the Photovoltaic module's power. The system was modeled in Simulink under the ideal conditions of its components. It could face few constraints during prototype implementation due to the unusual characteristics of the thermoelectric elements. However, certain additional electrical energy was achieved for a low duty power load, such as a remote seismic node. The significant contributions are to identify operating constraints and design optimal hybrid energy harvesting systems at a remote seismic node.