| Summary: | 碩士 === 國立雲林科技大學 === 電機工程系 === 106 === This thesis proposes a hybrid maximum power point tracking (MPPT) method with zero oscillations in steady-state by combining genetic algorithm (GA) and perturb and observe (P&O) method. The proposed MPPT can track the global maximum power point (GMPP) quickly for a photovoltaic (PV) system even under partial shaded conditions (PSC). The oscillations around the GMPP are eliminated and the power loss can be reduced significantly. In addition, the proposed MPPT can make the PV system operate at the highest efficiencies under different atmospheric conditions. The P-V characteristic curves appear multiple peak points when PSC occurs, which will lead the conventional MPPT methods (P&O, INCs, HC…etc.) to misjudge, thus make the PV system work at local maximum power points (LMPP) instead of the GMPP. During the MPP tracking, the system will oscillate around the MPP’s, resulting in a large and unnecessary power loss. To solve the problem, the artificial intelligence (AI) algorithms, such as PSO, Bee Colony optimization, GA…etc., were developed to deal with this issue. However, the problem with the AI algorithm is that the time for convergence may be too long if the range of the MPP search space is large. In addition, if the atmospheric conditions change fast, the PV system may operate at wrong working point for a long time. In this thesis, a method combining the P&O’s fast tracking and GA’s GMPP tracking ability is proposed. The proposed system can stop the oscillation as soon as the GMPP is found, thus minimizing the power loss. The proposed hybrid MPPT can achieve superior overall performance while maintaining the simplicity of implementation. At the last, the simulation and experimental results are presented to demonstrate the feasibility of the proposed system.
|
|---|
