A sustainable and efficient alternative for water pumping in electrically isolated rural areas of Ecuador

The main economic activities in rural areas of Ecuador are agriculture and livestock rearing, with water being the primary resource for these activities. Approximately 3% of the rural population lacks access to the conventional electrical grid, so photovoltaic pumping systems can fulfil the need to capture and distribute water without emitting greenhouse gases. This research aims to study the appropriate technologies to develop a photovoltaic pumping system for future implementation in rural areas of Ecuador, based on a centrifugal pump, induction motor, variable speed drive, and powered by a photovoltaic array. Simulink/MATLAB R2019B is used for system modelling and simulation. Additionally, two maximum power point tracking methods are studied, the perturb and observe algorithm and the artificial neural network algorithm. The simulation results show that when using the artificial neural network maximum power point tracking technique, the system performs better, being faster in transitions and presenting fewer oscillations in the steady state. The EN50530 standard is also employed to test the performance of the maximum power point tracker, and as a result, the artificial neural network-based algorithm achieves the highest efficiency of 99.5%. Finally, it is shown how the automatic regulation of speed allows adjustment of the pumping capacity in response to variations in irradiation and temperature. Even in unfavourable levels of irradiation and temperature, the water supply is not interrupted, and the pump capacity is adjusted to the maximum power supplied by the photovoltaic array.