Applied Soft Computing 65 (2018) 333–348 Contents lists available at ScienceDirect Applied Soft Computing journal homepage: www.elsevier.com/locate/asoc A maximum power point tracking method for PV system with improved gravitational search algorithm Ling-Ling Li a , Guo-Qian Lin a , Ming-Lang Tseng b,c,∗ , Kimhua Tan d , Ming K. Lim e a School of Electrical Engineering, Hebei University of Technology, Tianjin, 300130, China b Institute of Innovation and Circular Economy, Asia University, Taiwan c School of Business, Dalian University of Technology, Panjin, China d School of Business, University of Nottingham, UK e Chongqing University, China a r t i c l e i n f o Article history: Received 18 July 2017 Received in revised form 18 December 2017 Accepted 22 January 2018 Keywords: PV system MPPT Partial shading Improved gravitational search algorithm Particle swarm algorithm a b s t r a c t Photovoltaic (PV) system has gradually become research focus in the field of renewable energy power generation, and the output efficiency of PV system is the major concern of researchers. There are obvious non-linear characteristics in the output of PV system, and it will be greatly affected by external envi- ronment. For achieving the maximum output power, PV system must operate under the guidance of maximum power point tracking (MPPT) methods The tracking time and accuracy of these methods need to be improved. Therefore, this study contributes to increase output efficiency of PV system by improving the tracking time and accuracy of existing MPPT methods Specifically, a MPPT method with improved gravitational search algorithm (IGSA-MPPT) was proposed. The dynamic weight was added in the change factor of the gravity constant and the related factors of memory and population information exchange were added into the updating formula of particle velocity. IGSA-MPPT not only reduced the tracking time, but also improved the tracking accuracy and mitigated the fluctuations of the reference voltage. Finally, simulation results are compared with the of MPPT methods with particle swarm Optimization (PSO-MPPT) and gravitational search algorithm (GSA-MPPT). The average tracking time of IGSA-MPPT was reduced by 0.023 s and 0.0116s, and the average increase rates of maximum power were increased by 1.7071% and 0.7001% compared with PSO-MPPT and GSA-MPPT. In the simulations of PV system under the varying irradiance and temperature, the tracking speed and tracking accuracy of IGSA-MPPT were higher than those of PSO-MPPT, GSA-MPPT, GWO-MPPT, ICO-MPPT, and FCGSA-MPPT. In summary, IGSA- MPPT has better performance in tracking time and accuracy than other comparison algorithms. It can improve output efficiency of PV system in practical application. © 2018 Elsevier B.V. All rights reserved. 1. Introduction With the continuous improvement of the world economy and the human living standards, the consumption of energy in the world is increasing. Among these consumed energy sources, pri- mary energy represented by traditional fossil fuels (such as coal, oil, natural gas, etc.) still occupies a high proportion. These energy sources will inevitably cause environmental pollution and other issues. And as non-renewable energy sources, the reserves of them are becoming increasingly exhausted. Therefore, these energies are ∗ Corresponding author at: Institute of Innovation and Circular Economy, Asia University, Taiwan. E-mail addresses: lilinglinglaoshi@126.com (L.-L. Li), 505270173@qq.com (G.-Q. Lin), tsengminglang@asia.edu.tw (M.-L. Tseng). being replaced by clean energies such as solar energy gradually [1,2]. Compared to fossil fuels, solar energy is one of the most ideal energies since it is widely distributed, and will not produce any pollution in the operation [3]. For these advantages of solar energy, the proportion of photovoltaic (PV) system in all power generations is also increasing. PV system is of great significance in improving power transmission structure and reducing environmental pollu- tion. Nonetheless, there are still some shortcomings in PV system. The most important one is the low output efficiency. According to the existing statistics, efficiency of converting from solar energy to electric power is only 15%–20% [4]. While, for the fluctuations in irradiance and temperature of external environment, the con- verted power cannot be completely injected to the grid or stored in the energy storage system. This would cause the energy waste in electric power [5,6]. https://doi.org/10.1016/j.asoc.2018.01.030 1568-4946/© 2018 Elsevier B.V. All rights reserved.