跳到主要內容

臺灣博碩士論文加值系統

(44.211.31.134) 您好!臺灣時間:2024/07/13 02:35
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:邱文澔
研究生(外文):CHIU,WEN-HAO
論文名稱:無人機結合AI影像辨識 應用於太陽能板巡檢
論文名稱(外文):Drones combined with AI image recognition applied to solar panel inspection
指導教授:王紹宇
指導教授(外文):WANG,SHAO-YU
口試委員:侯帝光毛世威王紹宇
口試委員(外文):HOU,DI-GUANGMAO,SHI-WEIWANG,SHAO-YU
口試日期:2024-03-14
學位類別:碩士
校院名稱:國立聯合大學
系所名稱:機械工程學系碩士班
學門:工程學門
學類:機械工程學類
論文種類:學術論文
論文出版年:2024
畢業學年度:112
語文別:中文
論文頁數:88
中文關鍵詞:無人飛行載具VGG-16YOLOv8影像辨識太陽能板巡檢
外文關鍵詞:Unmanned Aerial VehicleVGG-16YOLOv8Image RecognitionSolar panel inspection
相關次數:
  • 被引用被引用:0
  • 點閱點閱:148
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
太陽能已經成為再生能源不可或缺的重要一環,為了使太陽能板發電功率維持良好效果,即時監測的重要性是不可缺少的手段,但以往人工巡檢的程序相當複雜,且過程中耗時耗力,同時政府於通過立法,於新的建案與相關建物等都需加裝太陽能裝置,以致快速的監測及巡檢顯得格外重要。本研究主要使用無人飛行載具獲得太陽能板狀況相關影像實施巡檢,並透過AI的深度學習(Deep learning, DL)之影像辨識,來識別太陽能板的硬體設備有無故障異常等現象,分類並檢測出有狀況的太陽能板位置,以 (Visual geometry group,VGG-16) 檢測靜態、(You only look once,YOLOv8) 檢測動態,結果顯示靜態準確率達到了95%、動態檢測準確率達到99.3%。因此本研究主要針對人力巡檢程序進行優化,利用無人飛行載具進一步解決了人工檢查時的耗時耗力,同時減少檢修人員暴露於高處的危險性,將提前訓練好得不同網路模型區分動態、靜態實施檢測,增加故障識別的準確性,可更加快速的識別有問題的太陽能板,進而提高了能源產生率,能有效的將效果最大化。
Solar energy has become an integral component of renewable energy, with most solar panels installed in large power stations or on rooftops to efficiently harness sunlight. To ensure optimal power generation, real-time monitoring is imperative. Traditional manual inspection procedures were intricate and time-consuming. The 2023 implementation of the Renewable Energy Development Regulations by the government mandates the installation of solar energy devices in new construction projects, emphasizing the need for swift monitoring and inspection. This research employs unmanned aerial vehicles (UAVs) to capture images for inspecting the condition of solar panels. Utilizing AI deep learning (DL) image recognition, a rapid and straightforward identification method is employed to assess the integrity of solar panels, identifying faults and abnormalities. The study distinguishes between static and dynamic detection using the Visual Geometry Group (VGG-16) for static accuracy and You Only Look Once (YOLOv8) for dynamic accuracy. The results demonstrate a static detection accuracy of 95% and a dynamic detection accuracy of 99.3%. The primary focus of this study is to optimize the manual inspection process, leveraging UAVs to address the time-consuming and labor-intensive nature of manual inspections. This approach not only reduces the risk associated with maintenance personnel working at elevated locations but also enhances fault identification accuracy. Pre-training various networks enables the model to differentiate between dynamic and static implementations, ultimately leading to quicker identification of problematic solar panels. This approach aims to boost energy generation rates, effectively maximizing the overall impact.
摘要 I
Abstract II
致謝 IV
目錄 V
圖目錄 VII
表目錄 X
第1章 前言 1
1.1 研究背景 1
1.2 研究動機 3
1.3 研究目的 5
第2章 文獻回顧 7
2.1 無人機的應用 7
2.2 太陽能板巡檢 9
2.3 人工智慧深度學習(DL) 11
2.4 影像辨識 13
第3章 研究原理與方法 17
3.1 無人機 18
3.1.1 系統及結構 18
3.1.2 平衡及控制原理 21
3.2 AI暨影像辨識 26
3.2.1 VGG-16之太陽能板靜態辨識 27
3.2.2 YOLOv8之太陽能板動態辨識 31
3.2.3 混淆矩陣 41
3.2.4 平均精確均值mAP (Mean average precision,mAP) 43
第4章 結果與討論 46
4.1 靜態影像辨識之VGG-16分析結果 46
4.2 動態影像辨識之YOLOv8分析結果 49
4.2.1 訓練、驗證損失 52
4.2.2 DFL損失函數 (Distribution Focal Loss,DFL) 57
4.2.3 混淆矩陣結果 62
4.2.4 平均精確均值mAP 65
第5章 結論 69
未來展望 71
參考文獻 72


[1] "Renewables 2023 Global Status Report," REN 21, 2023.
[2] 全國法規資料庫, "再生能源發展條例," 21 6 112. [Online]. Available: https://law.moj.gov.tw/LawClass/LawAll.aspx?pcode=J0130032.
[3] M. N. Kumar , S. Kumarasamy, . M. Samykano and J. Veerasundram, "On the technologies empowering drones for intelligent monitoring of solar photovoltaic power plants," Procedia Computer Science, vol. 133, pp. 585-593, 2018.
[4] E. S. Berman, M. Fladeland, J. Liem , R. Kolyer and M. Gupta, "Greenhouse gas analyzer for measurements of carbon dioxide, methane, and water vapor aboard an unmanned aerial vehicle," Sensors and Actuators B: Chemical, vol. 169, pp. 128-135, 2012.
[5] G. Rohi, O. Ejofodomi and G. Ofualagba, "Autonomous monitoring, analysis, and countering of air pollution using environmental drones," Heliyon, vol. 6, no. 1, 2020.
[6] J. Burgués, S. Doñate, M. D. Esclapez, L. Saúco and S. Marco, "Characterization of odour emissions in a wastewater treatment plant using a drone-based chemical sensor system," Science of The Total Environment, vol. 846, no. 157290, 2022.
[7] R. R. Hemamalini, R. Vinodhini, B. Shanthini, P. Partheeban, M. Charumathy and K. Cornelius, "Air quality monitoring and forecasting using smart drones and recurrent neural network for sustainable development in Chennai city," Sustainable Cities and Society, vol. 85, no. 104077, 2022.
[8] S. M. Jena, S. M. Sant, R. Mishra, S. K. Nayak and S. Nayak, "Design and analysis of flying u a V drone for sanitization purpose," Materialstoday:PROCEEDINGS, vol. 62, no. 12, pp. 6583-6589, 2022.
[9] P. F. Peña, A. R. Ragab, M. A. Luna, M. S. A. Isaac and P. Campoy, "WILD HOPPER: A heavy-duty UAV for day and night firefighting operations," Heliyon, vol. 8, no. 6, 2022.
[10] S. Prathibha, K. Saradha, P. Sharmila and S. Kaveya, "AI and Web Application in Medicine Delivery Drones," in 2022 International Conference on Applied Artificial Intelligence and Computing (ICAAIC), Salem, India, 2022.
[11] N. Molina, F. Cabrera, V. Araña, M. Tichavska, B. Dorta and J. Godoy, "A Wireless Method for Drone Identification and Monitoring Using AIS Technology," in 2018 2nd URSI Atlantic Radio Science Meeting (AT-RASC), Gran Canaria, Spain, 2018.
[12] J.-T. Zou and R. G. V, "Drone-based solar panel inspection with 5G and AI Technologies," in 2022 8th International Conference on Applied System Innovation (ICASI), Nantou, Taiwan, 2022.
[13] W. Tang, Q. Yang, K. Xiong and W. Yan, "Deep learning based automatic defect identification of photovoltaic module using electroluminescence images," Solar Energy, vol. 201, pp. 453-460, 2020.
[14] K. Simonyan and A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition," arxiv Computer Science, 2015.
[15] O. P. Olawale, K. Dimililer and F. A. Turjman, "Chapter Six - AI simulations and programming environments for drones: an overview," Drones in Smart-Cities, pp. 93-106, 2020.
[16] J. Schwalb, V. Menon, N. Tenhundfeld, K. Weger, B. Mesmer and S. Gholston, "A Study of Drone-based AI for Enhanced Human-AI Trust and Informed Decision Making in Human-AI Interactive Virtual Environments," in 2022 IEEE 3rd International Conference on Human-Machine Systems (ICHMS), Orlando, FL, USA, 2022.
[17] M.-F. R. Lee, C.-Y. Chung, A. S. M. Espinola, M. J. G. Vera and G. F. P. Caballero, "Deep Learning based Gesture Recognition for Drones," in 2022 18th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA), Taipei, Taiwan, 2022.
[18] M. Z. Shanti, C. S. Cho, B. G. d. Soto, Y. J. Byon, C. Y. Yeun and T. Y. Kim , "Real-time monitoring of work-at-height safety hazards in construction sites using drones and deep learning," Journal of Safety Research, vol. 83, pp. 364-370, 2022.
[19] O. G. Ajayi, J. Ashi and B. Guda, "Performance evaluation of YOLO v5 model for automatic crop and weed classification on UAV images," Smart Agricultural Technology, vol. 5, 2023.
[20] M. Harvey, J. Rowland and K. Luketina, "Drone with thermal infrared camera provides high resolution georeferenced imagery of the Waikite geothermal area, New Zealand," Journal of Volcanology and Geothermal Research, vol. 325, pp. 61-69, 2016.
[21] M. Ji, L. Zhang and Q. Wu, "Automatic grape leaf diseases identification via UnitedModel based on multiple convolutional neural networks," Information Processing in Agriculture, vol. 7, no. 3, pp. 418-426, 2022.
[22] A. A. Maashri, A. Saleem, H. Bourdoucen, O. Eldirdiry and A. A. Ghadani, "A novel drone-based system for accurate human temperature measurement and disease symptoms detection using thermography and AI," Remote Sensing Applications: Society and Environment, vol. 27, 2022.
[23] M.-F. R. Lee, C.-Y. Chung, A. S. M. Espinola, M. J. G. Vera and G. Federi, "Deep Learning based Gesture Recognition for Drones," in 2022 18th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA), Taipei, Taiwan, 2022.
[24] M. N. A. Kharim , A. Wayayok, A. F. Abdullah , A. R. M. Shariff , E. M. Husin and M. R. Mahadi, "Predictive zoning of pest and disease infestations in rice field based on UAV aerial imagery," The Egyptian Journal of Remote Sensing and Space Sciences, vol. 25, no. 3, pp. 831-840, 2022.
[25] T. Thaker, "Medium," 8 2021. [Online]. Available: https://medium.com/nerd-for-tech/vgg-16-easiest-explanation-12453b599526.
[26] G. Jocher, "GitHub," ultralytics, 10 1 2023. [Online]. Available: https://github.com/ultralytics/ultralytics.
[27] O. G. Ajayi, J. Ashi and B. Guda, "Performance evaluation of YOLO v5 model for automatic crop and weed classification on UAV images," Smart Agricultural Technology, vol. 5, 2023.
[28] J.-T. Zou and R. G. V, "Drone-based solar panel inspection with 5G and AI Technologies," in 2022 8th International Conference on Applied System Innovation (ICASI), Nantou, Taiwan, 2022.

電子全文 電子全文(網際網路公開日期:20290416)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊