International Journal of Power Electronics and Drive System (IJPEDS) Vol. 10, No. 3, Sep 2019, pp. 1329~1338 ISSN: 2088-8694, DOI: 10.11591/ijpeds.v10.i3.pp1329-1338  1329 Journal homepage: http://iaescore.com/journals/index.php/IJPEDS High performance PV system based on artificial neural network MPPT with PI controller for direct current water pump applications Hussain A. Attia Department of Electrical, Electronics & Communications Engineering, American University of Ras Al Khaimah, United Arab Emirates Article Info ABSTRACT Article history: Received Dec 21, 2018 Revised Feb 22, 2019 Accepted Mar 23, 2019 This paper presents a new design of a standalone photovoltaic system which is supplying the required power to a direct current water pump that have difficulty to supply by the utility electricity. The system is controlled by an artificial neural networks (ANN) algorithm with function softening by PI controller that to guarantee the maximum power point tracking (MPPT) working conditions. A parallel connected PV array is designed to supply the required power to the water pump. The proposed design considers Permanent Magnet DC motor (PMDC) of 48 Volts, and 500 Watts as a water pump’s motor, the direct current (DC) pump is adopted to avoid the complexity of the alternating current AC pumping system which includes inverter, power filter, and insulated step up transformer, so the presented design avoids the mentioned AC system components. A feed forward ANN algorithm is adopted in this study to produce the reference voltage for the MPPT functioning of the PV system, Proportional Integral (PI) controller is inserted to soften the MPPT controller performance. System design, MATLAB simulation with results and the results’ analysis all are presented in this paper. The study conclusion confirms the effectiveness of the proposal as a successful system for practical applications. Keywords: ANN DC-DC converter MATLAB/Simulink MPPT PI controller Power requirement PV array Standalone PV system Copyright © 2019 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Hussain Attia, Electrical, Electronics & Communication Engineering Dept., American University of Ras Al Khaimah , Ras Al Khaimah, 10021, UAE. Email: lsntl@ccu.edu.tw 1. INTRODUCTION Due to the difficulty of reaching the grid electricity to the remote farms, the renewable energy sources are representing an ideal alternative to supply the required electrical power to the remote farms loads such as water pump. Solar energy is playing major role as an important renewable energy source due to its cleanness, zero cost and everywhere availability. Photovoltaic (PV) system function is producing the required electricity by converting the solar energy to direct current electrical power. The PV panel produced electricity is affected by the weather conditions, such as irradiation and temperature levels. These levels are fluctuate the instantaneous output power of the PV panel. The harvested power from the PV panel is affected positively by the irradiance while affected negatively by the ambient temperature [1-4]. The PV panel contents a matrix of basic solar unit which is a solar cell, these cells are connecting in series and/or in parallel based on the required specifications from the panel. In other words, to increase the level of panel’s output voltage, the number of serially connected cells will be increased whereas to increase the level of panel’s output current, the number of parallel connected cells will be increased. Due to the non-linearity of the solar power response during the fluctuation in the weather