直流集成光伏模块输出功率优化与并网

发布时间：2019-06-20 13:24

【摘要】：论文在了解国内外近年光伏发电系统研究现状的基础上,确定以直流模块式发电系统为研究对象。对光伏电池的基本理论和输出特性进行分析,介绍了适应不同特性曲线的传统单峰值MPPT方法和多峰值MPPT方法;对比研究常见的光伏系统结构和DC/DC拓扑,确定了本文最终采用直流模块集成式串并联结构,DC/DC变换器采用Buck-Boost拓扑。论文提出了一种基于抛物线预测的单峰值MPPT方法,该方法通过用光伏特性曲线上的三个点来拟合一个二次函数,用函数驻点的实际功率值来迭代地预测和修正最大功率点的位置。采用了合理的初始点选取策略,提升了MPPT的速度,该算法通过实验平台进行了验证;提出适用于多峰值MPPT的QPSO算法,采用基于电压闭环控制的探测搜索策略,使电压粒子在光伏特性曲线上进行具有量子行为的搜索寻优;通过设置聚散因子调整策略和迭代终止策略,有效改善了 QPSO算法在光伏模块最大功率跟踪系统中的寻优速度和稳态跟踪精度,消除了稳态功率振荡。与经典粒子群优化PSO算法及传统PO法进行仿真和实验对比,结果表明,QPSO算法具有调节参数少,全局寻优能力强,稳态性能好,对复杂曲线适应能力强的优良性能。研究了直流集成光伏模块串并联系统的功率-母线电压输出特性曲线,针对该曲线可能为多峰值曲线进行了母线电压的最大功率寻优控制,通过仿真验证了PSO算法和QPSO算法实现母线电压寻优的可行性,搭建了光伏模块并网系统的实验平台,并对直流集成光伏模块串并联系统进行了并网实验。
[Abstract]:On the basis of understanding the research status of photovoltaic power generation system at home and abroad in recent years, the DC modular power generation system is selected as the research object. The basic theory and output characteristics of photovoltaic cells are analyzed, and the traditional single peak MPPT method and multi-peak MPPT method suitable for different characteristic curves are introduced, and the common photovoltaic system structure and DC/DC topology are compared and studied, and the DC module integrated series-parallel structure is adopted in this paper, and the Buck-Boost topology is adopted in DC/DC converter. In this paper, a single peak MPPT method based on parabola prediction is proposed. In this method, a quadratic function is fitted by using three points on the photovoltaic characteristic curve, and the position of the maximum power point is predicted and modified iteratively by using the actual power value of the function standing point. The reasonable initial point selection strategy is adopted to improve the speed of MPPT, and the algorithm is verified by the experimental platform. The QPSO algorithm suitable for multi-peak MPPT is proposed, and the detection and search strategy based on voltage closed-loop control is adopted to make the voltage particles search and optimize with quantum behavior on the photovoltaic characteristic curve. By setting the dispersion factor adjustment strategy and iterative termination strategy, the optimization speed and steady-state tracking accuracy of QPSO algorithm in the maximum power tracking system of photovoltaic module are effectively improved, and the steady-state power oscillation is eliminated. Compared with the classical particle swarm optimization PSO algorithm and the traditional PO method, the simulation and experimental results show that the QPSO algorithm has the advantages of less adjustment parameters, strong global optimization ability, good steady-state performance and strong adaptability to complex curves. The power-bus voltage output characteristic curve of DC integrated photovoltaic module series-parallel system is studied. The maximum power optimization control of bus voltage is carried out because the curve may be multi-peak curve. The feasibility of realizing bus voltage optimization by PSO algorithm and QPSO algorithm is verified by simulation. The experimental platform of photovoltaic module grid-connected system is built, and the grid-connected experiment of DC integrated photovoltaic module series-parallel system is carried out.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM615

【相似文献】