Hybrid energy harvesting system design and optimization for Seismic network

Remote seismic nodes frequently experience failure because of the short lifecycle of lead–acid batteries. Even the use of solar cells is insufficient because these cells have low energy conversion and requires the deployment of relatively expensive solar energy harvesting systems at the node. The conventional solar energy harvesting system requires hardware upgrades and charging to sustain the power at the remote seismic node. Lead–acid batteries and Direct Current-Direct Current (DC–DC) converters lose electrical energy due to low energy conversion, energy leakage, and shorter lifecycle. This study aims to develop an energy harvesting system with hybrid energy storage characterized by optimized constraints to enable the creation of a continuous and long-term remote seismic node database. This is anticipated to aid in the near real-time monitoring of the seismicity of an area and prediction of earthquakes.

A single-diode model of the solar cell and photovoltaic module was adopted and developed in MATLAB, Simulink, and PSIM programs based on fundamental equations. The output observations of voltage, current, and power were measured, compared, and presented in graphs and tabulated forms. A Maximum Power Point Tracking (MPPT) algorithm is required to interface with the DC–DC converter for optimal solar energy conversion; rapid changes in irradiance and temperature influence the effectiveness of the algorithm. However, conventional algorithms inaccurately respond under these environmental conditions and change the duty cycle levels across the converter. They exhibit steady-state oscillations that cause continuous power losses while tracking possible Maximum Power Points (MPPs). Alternatively, an algorithm capable of managing the irregularities of inputs and nonlinearities of the photovoltaic module using curve fitting equations derived from the parameters of the selected module is adopted. The three-dimensional lookup table function yields the optimal photovoltaic voltage, current, and power outputs. The table requires a low computational process, and low oscillations

at maximum voltage, current and power points are observed. The table consists of precomputed and possible combinations of optimal targets and input variables transformed by the artificial neural network model to supply maximum power levels. The proposed approach is simulated and tabulated by employing a single 120-W photovoltaic module. The algorithm, encoded using C programming, is embedded in an 8-bit microcontroller to feed optimal duty cycle levels across the pulse width modulation of the DC–DC converter. To improve the efficiency beyond 5 %, a synchronous buck DC–DC converter is adopted in this research. In the DC–DC converter, the investigation targets achieving a 0.1-ms processing time with an input voltage between 12.5 and 13.5 V. The fixed output voltage is 12 V through all responses while it is interfaced with the photovoltaic module. The converter simulations were performed in PSIM and Proteus

environments with a remote seismic node mimicked with a drawn load current level of 0.8 A. The synchronous DC–DC buck converter is connected to the hybrid energy storage of the lead–acid battery and supercapacitor via a shunt connection of the converter. The novelty of this research is its delivery of two power solutions at the remote seismic node: (1) low computational effort and a low-cost MPP algorithm by adopting a lookup table and the curve fitting equations of a selected photovoltaic module; (2) a hybrid of lead–acid battery and supercapacitor with a synchronous buck converter to extend the lifecycle and ensure continuous current flow through the remote seismic node. The proposed approach was evaluated and verified along with the computer-controlled photovoltaic system. Based on tabulated results and graphs, the deviation VII was between 2.5 % and 5 %. The results show that the proposed MPPT algorithm has less complexities, and the supercapacitor serves as buffer for the lead–acid battery to achieve an efficiency level of approximately 75 %.