基于改进遗传算法的反向传播神经网络拟合LED光谱模型

发布时间：2018-01-13 09:18

本文关键词：基于改进遗传算法的反向传播神经网络拟合LED光谱模型　出处：《激光与光电子学进展》2017年07期 　论文类型：期刊论文

【摘要】：发光二极管(LED)太阳光模拟器的设计需要对LED光谱建立精度高且稳定性好的数学模型。针对LED光谱数学模型非线性的特点,提出利用一种经改进遗传算法(GA)优化的反向传播(BP)神经网络对LED光谱模型进行辨识。通过改进GA的算子,提高算法收敛效果和辨识精度,利用改进GA对BP神经网络初始和权值阈值进行优化,用于建立可靠的LED光谱模型。选取不同驱动电流条件下的白色、红色LED光谱进行实验验证,实验结果表明该算法拟合的LED光谱模型与实际测量光谱分布非常接近,相比其他模型精度更高,普适性更好。
[Abstract]:The design of light-emitting diode (LED) solar simulator requires the establishment of a precise and stable mathematical model for the LED spectrum, aiming at the nonlinear characteristics of the LED spectral mathematical model. An improved genetic algorithm (GA) -optimized backpropagation BP neural network is proposed to identify the LED spectral model. By improving the GA operator, the convergence effect and identification accuracy of the algorithm are improved. The improved GA is used to optimize the initial and weight threshold of BP neural network to establish a reliable LED spectral model. The white and red LED spectra under different driving current conditions are selected for experimental verification. The experimental results show that the LED spectral model fitted by the proposed algorithm is very close to the measured spectral distribution and has higher accuracy and better universality than other models.
【作者单位】：杭州电子科技大学生命信息与仪器工程学院;
【基金】：国家重大科研仪器研制项目(61427808)
【分类号】：TN312.8;TP18
【正文快照】： 1引言随着发光二极管(LED)技术的日益发展,凭借其响应时间短、可控性能好、产品类型丰富等优点,LED已逐步成为太阳光模拟器的主流光源[1]。在LED太阳光模拟器的设计中,需要对LED光谱分布建立等效的数学模型,在此基础上才能完成多种LED合成标准太阳光谱的研究,得到满足性能要

【相似文献】