摘要
日光跟踪技术能明显提高光伏系统的效率,传统的光电检测日光跟踪技术的跟踪精度低,基于图像处理的日光检测跟踪技术精度虽高但算法复杂,系统成本高。通过图像处理进行太阳光斑检测跟踪的一般方法,包括太阳光斑图像的二值化、形心提取、差量计算等算法的分析,提出了一种简化的形心提取方法,通过与传统形心提取算法比较,误差很小并不足以对跟踪精度造成实质性的影响,最后用ARM-CortexM3内核的芯片作为载体实现了对太阳光斑图像的检测跟踪算法。
The efficiency of photovoltaic system can be improved by solar trace technique. The precision of conventional solar detection technique is low, and the trace technique based on solar image processing is of high precision, but also of high complexity algorithm and high cost. In this paper, several image processing algorithms, including image graying processing,image filtering, centroid processing, were analyzed. Aiming at reducing the cost of hardware and promoting the processing rate,a new method with low complexity for computing the offset of solar image position was proposed. By comparing to conventional centroid extraction algorithm, the error is tiny which can not bring a serious effect on the track precision. The implementation using ARM-CortexM3 MCU was presented to test the simplified solar trace algorithm.
出处
《电源技术》
CAS
CSCD
北大核心
2017年第3期429-431,共3页
Chinese Journal of Power Sources
基金
山东省自然科学基金项目(ZR2014JL044)
关键词
日光检测
图像处理算法
形心
solar detection
image processing algorithm
centroid
