臺灣博碩士論文加值系統

近年來，由於過度使用石化能源，其排放的含碳廢氣，造成全球氣候的變遷與暖化。為了減少含碳化合物的排放，世界各國致力於發展綠色能源(如太陽能發電、風力發電、潮汐發電……等)與提高能源使用的效率。而太陽能發電主要是藉著將太陽照度轉化成電能，其照度會隨著季節、時間、雲的遮蔽、氣候等因素所影響，把這些不穩定因素加入太陽能發電系統後，也可能造成系統在可靠度上的疑慮，進而衍生出太陽照度預測的重要性。本研究希望透過影像處理的技術，對全天空影像進行分析，預測雲朵在後續的時間內的移動是否會遮蔽住太陽？藉以降低“雲的遮蔽”在太陽照度的預測上的不穩定性。
本研究首先會讀取連續的全天空影像，偵測影像中雲的區域當作遮罩，利用連續影像相減法，取得影像中有移動的部分，再透過遮罩可擷取出影像中雲所移動的區域。利用雲所移動的區域，來擷取局部特徵點，並根據雲的區域遮罩來做分群，分完群的各群會被視為是獨立的雲。接著對連續影像中的各群做追蹤，保留追蹤成功的群組。透過對保留群組中的資訊做計算可取得特徵向量，使用機器學習的方法對特徵向量進行訓練，最後利用訓練得到的模組對其他的連續影像進行預測，並將預測的結果和實際影像情況作比對及分析。

In recent years, due to the excessive use of the fossil energy, carbon emissions have caused the global climate warming. In order to reduce the carbon emissions, countries around the world committed to the development of green energy which includes solar power, wind power and hydropower. In Taiwan, research of the solar power gets more attention gradually. But the solar irradiance would change dramatically due to season, time, weather and occlusion of clouds. These factors may cause the worries on the reliability of the solar power system. And forecasting short-term irradiance is important for the operators to manage and allocate resources. In our research, we use the image processing technology to analyze the all-sky images, and use the analysis results to predict the occlusive situation between sun and clouds. The prediction would help increase the reliability of the short-term solar irradiance forecasting.
In our research, we read the all-sky images and detect the area of clouds in the images as a mask first. Then, we use the image difference to get the motion region. Applying the cloud mask to motion region, we can get the cloud motion region in the images. Afterwards, we use the cloud motion region to detect the feature points. Then the feature points will be clustered by a clustering algorithm. After obtaining the clustering results, we perform tracking of feature clusters in continuous images. After tracking, we use the tracking information to calculate the feature vector. Then, we use this vector to train the predictive model. Finally, we do the prediction and validate the results with the ground truth. And we get a good performance that the prediction accuracy is higher than 85%.

第一章 緒論 P1
第二章 特徵偵測和分群 P6
第三章 特徵向量及太陽遮蔽預測 P29
第四章 實驗結果 P40
第五章 結論與未來工作 P49
參考文獻 50

[1] Bryan Urquhart,“Sky Imager Solar Forecasting for Microgrid
Optimization”,Nov. 2001
[2] 工業技術研究院,綠能與環境研究所,工程及自動化研究室,”太陽能出力預
測技術”,Jan. 2012
[3] M.S.Ghonima, B.Urquhart, C.W.Chow, J.E.Shields, A.Cazorla,
J.Kleissl,“A method for cloud detection and opacity classification
based on ground based sky imagery”Atmos.Meas.Tech,5,2881-2892, July. 2012
[4] J.Zhang,B.M.Hodge,A.Florita,S.Lu,H.F.,”Metrics for Evaluating the
Accuracy of Solar Power Forecasting”, NREL, Oct. 2013
[5] T.Stoffel, D.Renne,, D.Myers, S.Wilcox, M.Sengupta, R.George,
C.Turchi,”Best Practices Handbook for the Collection and Use of Solar Resource Data”,NREL, Sep.2010
[6] W.Fenga, B.Zhang ,Z.Caob, X.Zongb, J.Röningc,”Calibration and
Rectification Research for Fish-eye lens Application”,2011
[7] D.W.Slater, C.N.Long,” Total Sky Imager/Whole Sky Imager Cloud
Fraction Comparison”,Mar. 2001
[8] Santa Barbara Instrument Group,“SG-4 Autonomous Guider and
AllSky-340/340C All-Sky Camera Serial Interface Specification”,
July.2009
[9] David Campbell,“Widefield Imaging at Bayfordbury Observatory”,
Apr.2010
[10] C.N.Long, D.W.Slater, T.Tooman,” Total Sky Imager Model 880 Status and Testing Results”, Nov. 2001
[11] C.N.Long, J.M.Sabburg, J.Calbo, D.Pages,” Retrieving Cloud
Characteristics from Ground-Based Daytime Color All-Sky
Images”,Oct.2005
[12] Pierre Ineichen, Richard Perez,” A New Airmass Independent Formulation For The Linke Turbidity Coefficient”,Nov.2001
[13] Debasish Basak, Srimanta Pal, Dipak Chandra Patranabis,” Support Vector Regression”,July.2007
[14] C.Cortes, V.Vapnik,” Support Vector Machine”,Sep.1995
[15] Ibrahim Reda, Afshin Andreas,” Solar Position Algorithm for Solar Radiation Applications”,Jan.2008
[16] 周瑞雄,陳梧桐,陳春明,孫惠民,”卡爾曼預估器與多目標追蹤法則(Kalman Predictor and Multitarget Tracking Algorithm)”
[17] Lindsay Kleeman, Monash University, Clayton,” Understanding and Applying Kalman Filtering”
[18] A. Heinle, A. Macke, A. Srivastav,” Automatic cloud classification of whole sky images”,Apr.2010
[19] M. Martínez-Chico, F.J. Batlles, J.L. Bosch,” Cloud classification in a mediterranean location using radiation data and sky images”,Apr.2011
[20] Hao Huang, Shinjae Yoo, Dantong Yu, Dong Huang, Hong Qin,” Correlation and Local Feature Based Cloud Motion Estimation”,2012
[21] Janet E.Shields, Monette E.Karr, Art R.Burden, Richard
W.Johnson,Vincent W.Mikuls, Jacob R.Streeter,” Research toward Multi-Site Characterization Of Sky Obscuration by Clouds”,June.2007
[22]Ricardo Marquez, Carlos F.M.Coimbra,” Intra-hour DNI forecasting based on cloud tracking image analysis”,Oct.2012
[23]Bryan Urquhart, Chi Wai Chow, Andu Nguyen, Jan Kleissl,” Towards Intra-Hour Solar Forecasting Using Two Sky Imagers At A Large Solar Power Plant”,2010
[24]Hsu-Yung Cheng, Chih-Chang Yu, Sian-Jing Lin,” Bi-model short-term solar irradiance prediction using support vector regressors”,
Jan.2014
[25]Fei Wang, Zengqiang Mi, Shi Su, Hongshan Zhao,” Short-Term Solar
Irradiance Forecasting Model Based on Artificial Neural Network Using
Statistical Feature Parameters”,Feb.2012
[26]Ricardo Marquez, Carlos F.M.Coimbra,” Forecasting of global and
direct solar irradiance using stochastic learning methods, ground experiments and the NWS database”,Feb.2011
[27]Chia-Lin Fu, Hsu-Yung Cheng,” Predicting solar irradiance with
all-sky image features via regression”,Oct.2013
[28]Joshua S.Stein, Matthew J.Reno, govClifford W.Hansen,” The
Variability Index: A New And Novel Metric For Quantifying Irradiance And PV Output Variability”,2012
[29]Bryan Urquhart, Chi Wai Chow, Matt Lave, Jan Kleissl,” Intra-hour
forecasting with a total sky imager at the UC San Diego solar energy
testbed”
[30]Hsin-Chien Huang, Yung-Yu Chuang, Chu-Song Chen,” Affinity
Aggregation for Spectral Clustering”,IEEE.2012
[31]Dipti Prasad Mukherjee, Scott T.Acton,” Cloud Tracking by Scale
Space Classification”,Feb.2012
[32]Thomas M.Hamill, Thomas Nehrkorn,” A Short-Term Cloud Forecast
Scheme Using Cross Correlations”,June.1993
[33] Marko Heikkil, Matti Pietik¨ainen, Cordelia Schmid,” Description of Interest Regions with Local Binary Patterns”,June.2008
[34] Chris Harris, Mike Stephens,” A Combined Corner And Edge
Detector”,1988
[35] David G. Lowe,” Object Recognition from Local Scale-Invariant Features”
[36] Herbert Bay, Andreas Ess, Tinne Tuytelaars, Luc Van Gool,”
Speeded-Up Robust Features”,Dec.2007
[37] John Canny,” A Computational Approach to Edge Detection”,Nov.1986
[38] Richard O. Duda, Peter E. Hart,” Use Of The Hough Transformation To
Detect Lines And Curves In Picture”,Apr.1971
[39]Chris Stauffer, W.Eric L.Grimson,”Learning Patterns of Activity
Using Real-Time Tracking”,Aug.2000
[40]Tianzhu Zhang, Hanqing Lu, Stan Z.Li,”Learning Semantic Scene Models
by Object Classification and Trajectory Clustering”,June.2009 [41]Stefan Emilov Atev,”Using Asymmetry in the Spectral Clustering of
Trajectoies”,2011