Photovoltaic Power Generation Data Preprocessing Using Wavelet Thresholding

Abstract:

The accuracy of power prediction of photovoltaic power generation will be reduced if there are noises in the historical data of photovoltaic power generation. It is necessary to eliminate noises in the historical data. First, the power generation can be arranged in a two-dimensional data set by date. Secondly, the grayscale image matrix will be obtained after data normalized. Thirdly, two-dimensional wavelet based image de-noising method will be used in the matrix de-noising. Finally the real historical data will be obtained through anti-normalization the matrix which is already de-noised. We will analyze its results feasible and effectiveness.

• Periodical: Advanced Materials Research (Volume 936)
• Pages: 2179-2183
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.936.2179
• Online since: June 2014
• Authors: Kai Bai, Hong Da Qu, Jin Zong, Zhi Li*, Hua Xin Zhai, Zhou Zhao
• Keywords: De-Noising, Historical Data, Photovoltaic Power Generation, Power Predication

References

[1] Zhanglan, ZhangYanxia, GuoChangmin, ZhaoJie, PV systems based on neural network predictive power generation [J]. China Power，2010, 43(9)：75-78.

[2] XieLimei, Grid-connected PV system modeling and on the distribution grid voltage and the impact loss of grid[D], Peking：Thesis of North China Electric Power University，(2009).

[3] WuYaohua, Based on wavelet soft thresholding Term Load Forecasting [J]. Guangdong Electric Power，2007, 12(20)：5-8.

[4] XiongChuanping, CaoJunjie, ChenQian, JuPing, Power load based on wavelet analysis denoising model identification [J]. Journal of Hohai University，2011, 39(4)：470-474.

[5] DONOHO D L，JOHNSOTNE I M. Idea spatial adaptation by wavelet shrinkage[J]. Biometrika, 1994, 81(3); 425-455.

DOI: 10.1093/biomet/81.3.425

[6] DONOHO D L，De-noising by soft thresholding[J]. IEEE Trans on Information Theory, 1995, 38(2): 613-627.

DOI: 10.1109/18.382009

[7] XieJiecheng, ZhangDali, XuWenli, Image denoising Review [J]. Chinese Journal of Image and Graphics，2002，7（3）：209-217.

[8] ShiHongbo, Based on wavelet transform image denoising [D]，JiLin：Jilin University，(2007).

[9] ShiJiachen, Image Denoising Based on Wavelet Transform Algorithm and Its FPGA Implementation [D]. NanJing：Nanjing University，(2010).

[10] ZhangMing, LiKaicheng, HuYisheng, Neighborhood threshold based on wavelet classification power quality signal denoising algorithm [J]. Power Systems，2010, 34(10): 84-89.

[11] YangMing, HanXueshan, FanShu, Component based Sparse Bayesian learning probabilistic wind power output prediction method [J]. Power Systems，2012, 36(14)：125-130.

[12] DuanQing, Learning method based on Sparse Bayesian regression and classification in the prediction of the power system [D], Shandong：Dissertation of Shandong University，(2010).