Efficient operation of residential solar panels with determination of the optimal tilt angle and optimal intervals based on forecasting model

被引:18
作者
Akhlaghi, Shahrokh [1 ]
Sangrody, Hossein [1 ]
Sarailoo, Morteza [1 ]
Rezaeiahari, Mandana [2 ]
机构
[1] SUNY Binghamton, Elect & Comp Engn Dept, Binghamton, NY 13902 USA
[2] SUNY Binghamton, Syst Sci & Ind Engn Dept, Binghamton, NY 13902 USA
关键词
solar power stations; load forecasting; optimisation; sunlight; residential solar panels; optimal tilt angle determination; optimal intervals; forecasting model; solar panel tilt angle; solar tracking system; solar energy; optimisation problem; solar radiation; bee algorithm; stopping criterion; solar power generation; COLLECTORS; OPTIMIZATION; ORIENTATION; RADIATION;
D O I
10.1049/iet-rpg.2016.1033
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
A solar panel tilt angle plays a great role in the performance of the solar panel which is either fixed at an optimal tilt angle or continuously adjusted using a solar tracking system. Solar tracking systems are not cost efficient especially for residential usage. On the other hand, a fixed tilt angle results in a huge loss of solar energy. One resort to solve this problem is to adjust the tilt angle a limited number of times. In this paper, a novel procedure is proposed to select the number of intervals and their durations by solving an optimisation problem. The proposed algorithm is consisted of four major steps. First, the solar radiation of the next year is predicted using historical data. Second, using a bee algorithm the optimal tilt angle of each interval is computed. Third, an optimisation problem is solved to get new periods for each interval. Finally, a stopping criterion is checked to decide whether the previous step should be repeated or the algorithm has been converged. The effectiveness of the proposed approach is studied at nine different locations across the US. The results show improvement of the solar power generation by using the optimal intervals.
引用
收藏
页码:1261 / 1267
页数:7
相关论文
共 32 条