태양광 발전을 위한 MPPT 비교분석
Performance Analysis of MPPT Methods in a Photovoltaic System
- 주제(키워드) MPPT
- 발행기관 아주대학교
- 지도교수 이교범
- 발행년도 2013
- 학위수여년월 2013. 8
- 학위명 석사
- 학과 및 전공 산업대학원 정보전자공학과
- 실제URI http://www.dcollection.net/handler/ajou/000000015274
- 본문언어 한국어
- 저작권 아주대학교 논문은 저작권에 의해 보호받습니다.
초록/요약
We analyzed the pros and cons of the most frequently used MPPT algorithms in this paper. In a solar power module, energy output depends on external temperature and the amount of sunlight. Controlling this output is necessary to achieving minimal energy loss while at the same time maximizing electrical power. Maximum Power Point Tracking (MPPT) is a technique that attempts to maximize power output based on external conditions. MPPT may utilize one or more various methods, such as constant voltage control, Perturbation and Observation (P&O), Incremental Conduction (IncCond), “fuzzy” logic, neural networks, and the Newton method. Constant voltage control is an algorithm that varies the amount of current in the module to achieve constant voltage output. This is simple to achieve; however, if the amount of solar radiation changes rapidly, it may not find the maximum electric power, decreasing efficiency. P&O continually adjusts the voltage output of photovoltaic cells to maximize power output, and its method is fairly simple and accurate. However, depending on external conditions, it may cause oscillations around the maximum power points and fail to show the maximum electric power. Because of this, IncCond is often used in conjunction with P&O when power output is at or near the maximum power point, IncCond achieves its goal when the derivative of the power curve equals zero and it can track the maximum power point. The fuzzy logic method uses control rules based on expert knowledge, which makes it highly tolerant to noise. The control rules define relationships between input and output voltage to be used when mathematical calculations cannot produce accurate levels. However, this method can be very processor intensive if mathematical algorithms fail routinely. Neural network methods perform MPPT by “learning”: information is saved and used later. However, this method is also computationally expensive, so low-CPU systems cannot use this method. The Newton method is the most efficient way to approximate the actual power function. However, as the slope of the power curve approaches zero (the peak), the method becomes increasingly complex.
more목차
목 차 (List of Text)
Ⅰ. 서 론 1
Ⅱ. 태양광을 위한 MPPT 제어 3
2.1. 태양전지의 특성 3
2.1.1 태양전지의 동작특성 3
2.1.2 태양전지의 모델링 4
2.2. MPPT 제어 방법 7
2.2.1 일정전압 제어 방법 8
2.2.2 P&O 방법 9
2.2.3 IncCond 방법 10
2.2.4 최소자승법 13
2.2.5 퍼지제어 16
2.2.6 신경제어기 19
2.2.7 Newton Method 22
2.3. MPPT 제어 방법의 특성 비교 26
2.3.1 MPPT구현의 용이성 27
2.3.2 센서 27
2.3.3 MPPT 성능 27
2.3.4 CPU 연산량 28
Ⅲ. 결론 29
참고 문헌 30
