2018 2 nd Int. conf. on Innovations in Science, Engineering and Technology (ICISET) 27-28 October 2018, Chittagong, Bangladesh Optimization of PV Energy Generation based on ANFIS Nesar Uddin Md. Saiful Islam Institute of Energy Technology Dept. of Electronics and Telecommunication Engineering Chittagong University of Engineering and Technology Chittagong University of Engineering and Technology Chittagong, Bangladesh Chittagong, Bangladesh nesar15cuet@gmail.com saiful05eee@gmail.com Abstract— The main point of this paper is to take bring out the maximum power for ever-increasing the efficiency of solar photovoltaic (PV) under unstable weather conditions. It is a method of hybrid small scale solar energy conversion system that use an adaptive neural fuzzy interface system (ANFIS). The conversion of solar energy, mostly depends on the irradiation and temperature. ANFIS is used to extract maximum feasible power under unstable solar irradiance and temperature by generating duty cycle through pulse width modulation (PWM). Duty cycle triggers the gate of DC-DC boost converter. This proposed system has been optimized by using MATLAB/Simulink software. The simulation results of the MPPT controller show that very effective and efficient than other ordinary conventional systems such as without MPPT system. Keywords— Adaptive neural fuzzy interface system (ANFIS); Maximum power point tracking (MPPT) system; DC to DC Boost converter; Solar energy. I. INTRODUCTION Energy plays a vital role for our social life and economy [1]. Because of inanition of fossil fuel reserves, greenhouse effect, environmental degradation and high cost, the consumption of renewable energy sources is increasing day by day for electric power generation [1]. In order to meet up the ever rising energy demand and overcome above problem, it is necessary on the way to the renewable energy sources such as wind, solar, tidal, sea wave, geothermal and hydro energy for maximum potential [2]. Al of them, solar energy is considered more reliable for Bangladesh mainly remote area [2]. This energy source is daily available, and environmentally friendly than other renewable source. Bangladesh lies in the sunny regions of the world. The majority of the parts of Bangladesh receive 4–7 kWh of solar radiation per square meter per day. About 250–300 sunny day occurs in a year, which can mitigate the total load demand of a family in that’s country [3]. On the other hand, solar energy systems usually suffer from their low efficiency. In order to rise above these drawbacks, MPPT techniques is a way to optimize greater efficiency for maximum power of PV panel. MPPT is a real-time control system that applied to the PV power converter in order to bring out the maximum possible power from the PV panel [4]. In this paper MPPT has been designed by artificial intelligence based on ANFIS controller. Artificial intelligence systems is a process which can take decision like a human brain by adjusting themselves. The situations and making correct decisions take automatically for future similar conditions [5]. An adaptive neural fuzzy inference system (ANFIS) is a type of artificial intelligence system that is based on Takagi–Sugeno fuzzy interface system. It is a combination both neural networks and fuzzy logic ideology [5]. Its inference system corresponds to a fuzzy set of IF-THEN rules based on training data that have learning ability to approximate nonlinear functions [6]. For using a hybrid learning procedure, ANFIS can build an input- output mapping based on both human knowledge and predetermined input-output data pairs which is more efficient and optimal way [6]. In this paper, designing and implementation of ANFIS based MPPT scheme which is interfaced with boost converter in Matlab/Simulink. II. METHODOLOGY A. PV model The system consists of PV module, a DC-DC boost converter, a control unit and load. Mainly, PV module depends on the solar irradiance and temperature as well as atmospheric condition. PV power is transferred to the load through boost converter. Duty cycle to trigger the gate of the MOSFET switch is continuously adjusted to track the maximum power point. ANFIS is used as the control scheme which takes irradiance and temperature as input from change of PV power and voltage ratio. ANFIS gives duty ratio as output for maintaining the on and off time of the MOSFET switch through PWM shown in Fig. 1. Fig.1 Block diagram of proposed system. 978-1-5386-8524-2/18/$31.00 ©2018 IEEE