基于径向基神经网络的太阳能电池缺陷检测

沈凌云; 朱明; 陈小云

doi:10.3788/fgxb20153601.0099

您当前的位置：

首页 >

文章列表页 >

基于径向基神经网络的太阳能电池缺陷检测

发光学应用及交叉前沿 | 更新时间：2020-08-12

- 基于径向基神经网络的太阳能电池缺陷检测
- Solar Panels Defect Detection Based on Radial Basis Function Neural Network
- 发光学报 2015年36卷第1期页码：99-105
- 作者机构：
  
  1. 中国科学院长春光学精密机械与物理研究所,吉林长春,中国,130033
  2. 中国科学院大学北京,100049
  3. 长春理工大学电子信息工程学院,吉林长春,中国,130022
- 作者简介：
- 基金信息：
  
  国家自然科学基金(61203242)();中国科学院二期创新工程基金(C50T0P2)资助项目()
- DOI：10.3788/fgxb20153601.0099
  中图分类号： TP394.1;TH691.9
- 纸质出版日期：2015-1-3，
  
  网络出版日期：2014-11-18，
  
  收稿日期：2014-8-22，
  
  修回日期：2014-10-19，
扫描看全文
沈凌云, 朱明, 陈小云. 基于径向基神经网络的太阳能电池缺陷检测[J]. 发光学报, 2015,36(1): 99-105

SHEN Ling-yun, ZHU Ming, CHEN Xiao-yun. Solar Panels Defect Detection Based on Radial Basis Function Neural Network[J]. Chinese Journal of Luminescence, 2015,36(1): 99-105
沈凌云, 朱明, 陈小云. 基于径向基神经网络的太阳能电池缺陷检测[J]. 发光学报, 2015,36(1): 99-105 DOI： 10.3788/fgxb20153601.0099.

SHEN Ling-yun, ZHU Ming, CHEN Xiao-yun. Solar Panels Defect Detection Based on Radial Basis Function Neural Network[J]. Chinese Journal of Luminescence, 2015,36(1): 99-105 DOI： 10.3788/fgxb20153601.0099.

摘要

为了检测太阳能电池的缺陷

建立了太阳能电池板的电致发光(EL)图像与其缺陷类型间的神经网络预测模型

可以对太阳能电池板不同类型缺陷进行自适应检测。首先

采用主成分分量分析(PCA)算法对电致发光(EL)图像训练样本集降维;然后

将降维后得到的数据输入神经网络预测模型进行学习

对模型的参数进行优化选取;最后

将训练好的网络对测试样本集进行仿真。仿真结果表明:在采用相同的训练样本集和测试样本集条件下

与反向传播神经网络(BPNN)相比

径向基神经网络(RBFNN)具有全局最优特性

结构简单

最高识别率达96.25%

计算时间较短

能满足在线检测的要求。

Abstract

In order to detect the defect on solar panels and improve the conversion efficiency

two neural network models were established between solar panels electroluminescence (EL) images and defect types

which can detect different types of defects on solar panels adaptively. Firstly

the dimensions of EL images training samples set were reduced by using principal component analysis (PCA). Then

EL images training samples set after dimension reduction was put into the neural networks for training. Finally

the testing samples set was simulated by the trained network through choosing the best parameters. Compared with BPNN

RBFNN has the advantages of global optimization characteristics and simple structure

which leads to the highest accuracy rate of 96.25% and shorter computational time. The experiment results show that RBFNN can meet the requirements of online detection.

关键词

缺陷检测反向传播神经网络径向基神经网络主成分分析降维

Keywords

defect detectionback propagation neural network (BPNN)radial basis function neural network (RBFNN)principal component analysis (PCA)dimension reduction

references

Fuyuki T, Kitiyanan A. Photographic diagnosis of crystalline silicon solar cells utilizing electroluminescence [J]. Appl. Phys. A, 2009, 96:189-196.

Li Y H, Pan M, Pang A S. The application of electroluminescence imaging to detection the hidden defects in silicon solar cells [J]. Chin. J. Lumin.(发光学报), 2011, 32(4):378-382 (in Chinese).

Tsai D M, Wu S C, Chiu W Y. Automatic visual inspection of solar modules using independent component analysis[C]// Proceedings of The 14th Asia Pacific Industrial Engineering and Management Science Conference, Cebu: IEEE, 2013.

Yao M H, Li J, Wang X B. Solar cells surface defects detection using RPCA method [J]. Chin. J. Comput.(计算机学报), 2013, 36(9):1943-1952 (in Chinese).

Gong F, Zhang X W, Sun H. Detection system for solar module surface defects based on constrained ICA model and PSO method [J]. Acta Optica Sinica (光学学报), 2012, 32(4):1-9 (in Chinese).

Chiou Y C, Liu J Z, Liang Y T. Micro crack detection of multi-crystalline silicon solar wafer using machine vision techniques [J]. Sens. Rev., 2011, 31:154-165.

Jin J, Zhang Z G, Wang Z, et al. Temperature errors compensation for digital closed-loop fiber optic gyroscope using RBF neural networks [J]. Opt. Precision Eng.(光学精密工程), 2008, 16(2):235-240 (in Chinese).

Zeng L, He H J, Zhang Z B. Comparison inspection between ICT images & CAD model based on edge extracting by neural networks [J]. Opt. Precision Eng.(光学精密工程), 2011, 19(10):2533-2540 (in Chinese).

Deng X L, Cheng K, Wu W B. Detection of citrus huanglongbing based on principal component analysis and back propagation neural network [J]. Acta Photon. Sinica (光子学报), 2014, 43(4):1-7 (in Chinese).

Xu Z M, Zhou J Z, Huang S, et al. Quality prediction of laser milling based on optimized back propagation networks by genetic algorithms [J]. Chin. J. Lasers (中国激光), 2013, 40(6):1-8 (in Chinese).

Alexandridis A, Chondrodima E. A medical diagnostic tool based on radial basis function classifiers and evolutionary simulated annealing [J]. J. Biomed. Inform., 2014, 49(1):61-72

Lu C, Zhu Z, Gu X. An intelligent system for lung cancer diagnosis using a new genetic algorithm based feature selection method [J]. J. Med. Syst., 2014, 38(9):97-105

Tong D L, Schierz A C. Hybrid genetic algorithm-neural network: Feature extraction for unpreprocessed microarray data [J]. Artif. Intell. Med., 2011, 53(1):47-56.

Babu G S, Suresh S. Sequential projection-based metacognitive learning in a radial basis function network for classification problems [J]. IEEE Trans. Neur. Network Learning Syst., 2013, 24(2):194-206.

Ghosh-Dastidar S, Adeli H, Dadmehr N. Principal component analysis-enhanced cosine radial basis function neural network for robust epilepsy and seizure detection [J]. IEEE Trans. Biomed. Eng., 2008, 55(2):512-518.

Cybenko G. Approximation by superpositions of a sigmoidal function [J]. Mathematics of Control, Signals, and Systems, 1989, 2(4):303-314.

Xia K W, Li C B, Shen J Y. An optimizational algorithm on the number of hidden layer models in feed-forward neural network [J]. Comput. Sci., 2005, 32(10):143-145.

Li G N, Tan Q C, Zhang K, et al. Solar cells defect detection in electroluminescence images [J]. Chin. J. Lumin.(发光学报), 2013, 34(10):1400-1407 (in Chinese).

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据