Technical note Adaptive fuzzy gain scheduling PID controller for maximum power point tracking of photovoltaic system Anastasios I. Dounis * , Panagiotis Kofinas, Constantine Alafodimos, Dimitrios Tseles Technological Education Institute of Piraeus, Department of Automation, 250 P. Ralli & Thivon Str., Egaleo 122 44, Greece article info Article history: Received 24 November 2012 Accepted 26 April 2013 Available online 5 June 2013 Keywords: Maximum power point tracking Adaptive fuzzy PID controller Fuzzy gain scheduling Adaptive scaling factors Two-level control system architecture Fuzzy logic system abstract This paper proposes a methodology of designing a Maximum Power Point Tracking (MPPT) controller for photovoltaic systems (PV) using a Fuzzy Gain Scheduling of Proportional-Integral-Derivative (PID) type controller (FGS-PID) with adaptation of scaling factors (SF) for the input signals of FGS. The proposed adaptive FGS-PID method is based on a two-level control system architecture, which combines the advantages of fuzzy logic and conventional PID control. The initial values of the PID’s gains are deter- mined by the ZieglereNichols tuning method. During transient and steady states, the PID’s gains are adapted by the FGS-PID to damp out the transient oscillations, to reduce settling time and to guarantee system stability and accuracy. Also, the conditioned input signals of the FGS-PID are tuned dynamically by gain factors which are based on fuzzy logic system (FLS). The FLS is characterized by a set of fuzzy rules which are fuzzy conditional statements expressing the relationship between inputs (error and change of error) and outputs. This approach creates an adaptive MPPT controller and achieves better overall system performance. The simulation results demonstrate the effectiveness of the proposed adaptive FGS-PID and show that this approach can achieve a good maximum power operation under any conditions such as different levels of solar radiation and PV cell temperature for varying PV sources. Compared to conventional methods (PID, perturb and observe method P&O), this method shows a considerable high tracking performance. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction A photovoltaic system directly converts solar radiation into electricity. The PV electric power generation has two limitations: a) very low conversion efficiency, especially under low irradiation and b) the power generation depends on atmospheric conditions (solar radiation and temperature), ageing and load conditions [1,2]. In order to reduce the cost of energy, it is ideal to maintain the PV operation at its maximum achievable efficiency at any time. However, this goal is complicated by uncertain nonlinear currentevoltage (IeV) and po- werevoltage (PeV) characteristics due to changes in intrinsic and extrinsic factors. The maximum power point is a unique point on the PeV curves and at this point the PV system produces its maximum power. Although there are many factors influencing the energy conversion efficiency, the maximum power point tracking is the most vital aspect of control design for PV generation. The MPPT is substantially a nonlinear control problem. This is due to the nonlinear nature of PV and the continuous changes of its parameters with the unpredictable variations of the environmental conditions. A significant number of MPPT algorithms have been presented in the literature such as the perturb and observe method and the incremental conductance method [3,4]. The P&O method is widely used because of its low implementa- tion complexity [3]. The shortcoming of this method is that the operating point of the PV fluctuates around the MPP. Therefore, the available energy is decreased. Furthermore, if the solar irradiance changes rapidly, the P&O technique fails to track the real point of maximum power. This creates a slower tracking time response. The convergence speed is varied due to the slow trial and error process. The incremental conductance MPPT method is based on comparing the instantaneous conductance to the incremental conductance. At the maximum power point, the values of conduc- tance and incremental conductance are equal but with opposite signs. This method has medium implementation complexity compared to P&O [3]. The drawback of this technique is the oscil- lations around the MPP. MPPT fuzzy logic controllers (FLC) have the advantage of being fast robust and of having quiet good performance (time response, stability, tracking speed, small oscillations) under varying atmo- spheric conditions. MPPT FLCs are more effective under sudden changes of atmospheric conditions compared to the traditional * Corresponding author. E-mail addresses: aidounis@teipir.gr, aidounis@otenet.gr (A.I. Dounis). Contents lists available at SciVerse ScienceDirect Renewable Energy journal homepage: www.elsevier.com/locate/renene 0960-1481/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.renene.2013.04.014 Renewable Energy 60 (2013) 202e214