2-D nonlinear corretion in an improved BP neural network
-
摘要: 为了减少位置敏感传感器(PSD)的非线性的影响,分析了PSD的工作原理及其非线性成因,提出一种基于Levenberg-Morquardt算法改进的反向传播(BP)神经网络方法进行非线性修正,并进行了理论分析和MAT-LAB仿真比较.结果表明,改进的BP神经网络方法能有效地减少非线性影响,且相对传统的BP神经网络而言,收敛速度更快,使修正后的PSD器件在非线性区里获得与线性区近似的线性度.这一结果对PSD更好的应用是有帮助的.
-
关键词:
- 光学器件 /
- 位置敏感传感器 /
- 非线性修正 /
- 改进型反向传播算法 /
- Levenberg-Morquardt算法
Abstract: In order to reduce the effect of nonlinearity of a position sensitive detector(PSD),after analyzing its working principle and the reasons of nonlinearity formation,nonlinearity correction was carried out in an improved back propagation(BP)neural network based on Levenberg-Morquardt algorithm.MATLAB simulation results show that the improved BP neural network can reduce nonlinearity more effectively,and converge faster than a traditional BP neural network.After revision the PSD obtains approximate linearity in non-linear area within the linear area.This result is helpful for PSD application. -
-
[1] ZENG Ch.PSD research and application based on two-dimensional position sensitive detector[D].Hangzhou:Zhejiang University,2009:11-12(in Chinese).
[2] LI Zh K,QIN Y Y,GUO Q.The artificial neural networks in the twodimensional PSD component misalignment revises the application[J].Measurement and Control of Computer,2006,12 (5):463-466 (in Chinese).
[3] ZHOU Zh H,JIANG W J,ZHU W L.Investigations of PSD and its integration equip mark settle experimental[J].The Electron Measures with Instrument College Journal,2008,14 (4):6-9 (in Chinese).
[4] LU Zh Y,ZHOU Y J,LI W J.The BP neural network of the feeling machine nonlinear error correction method study[J].Feeling Technique,2005,24(2):11-12(in Chinese).
[5] ZHANG J T,DONG X L.The method of feeling machine of nonlinear correction[J].Technique of Calculate,2007,18 (6):27-29 (in Chinese).
[6] ZHOU B Y,CHEN P F,ZENG W,et al.Nonlinear error corrects experimental investigations of position sensitive detector[J].Optical Instrument,2008,27(5):11-15(in Chinese).
[7] XU L N.Control of neural networks[M].Beijing:Electronic Industries Press,2005:18-20(in Chinese).
[8] YAO W J.The improvment of BP algorithm studies in the MATLAB realize[J].Modern Electronics,2006,21 (6):95-98(in Chinese).
[9] NI F Zh,LI G Q.Spread the feeling machine nonlinear error correction method according to the measurement system of LM algorithm[J].Measure to Control a Technique,2008,13 (6):70-71 (in Chinese).
[10] LI G Y.Intelligence control and its MATLAB carry out[M].Beijing:Electronic Industries Press,2005:39-40(in Chinese).
[11] QI Zh G,WANG X G.A studies algorithm of advanced BP network[J].Automate Technique and Application,2005,17 (2):13-14 (in Chinese).
[12] ZHANG Sh N,GUO W Y,XIAO L Y.The design and training of MATLAB BP neural networks[J].Science and Economies of Neimenggu,2005,14(6):95-97(in Chinese).
[13] MIN X L,LIU G H.Appliment of tapping BP network use the MATLAB nerve network tool box[J].Appliment of Computer,2007,23(8):27-29(in Chinese).
计量
- 文章访问数: 2
- HTML全文浏览量: 0
- PDF下载量: 6
下载: