當太陽能發電系統發生部分遮蔭時，其功率-電壓特性曲線將由單峰變為複雜之多峰曲線，傳統最大功率追蹤(Maximum Power Point Tracking, MPPT)技術應用於部分遮蔭時容易陷入局部最大功率點而使太陽能發電系統無法輸出最大功率，因此能夠追蹤全域最大功率點的技術對太陽能發電系統就顯得格外重要。本文開發一基於類神經網路之混合型太陽能全域最大功率追蹤(Global MPPT, GMPPT)演算法，部分遮蔭發生時先透過第一階段類神經網路及後續計算估測出當時遮蔭樣式下之近似最大功率點電壓，接著利用第二階段自適應步階調變擾動觀察法(Adaptive Variable Step-Size Perturbation and Observation, Adaptive VSS P&O)來正確鎖定全域最大功率點;而於均勻照度下則使用類神經網路估測出第一區間最大功率點電壓後再乘以串聯太陽能電池模組數作為初始操作點，接著執行自適應步階調變擾動觀察法以獲得MPP。
為了驗證本文提出之演算法的可行性及成效，本文利用相同的平台進行模擬，並與近年文獻所提出之8篇基於類神經網路的太陽能全域最大功率追蹤演算法於部分遮蔭情況下針對GMPPT性能進行比較，模擬結果顯示所提方法之平均電能損失僅0.12 W，平均追蹤時間僅1.48 s。為驗證模擬之正確性，本文亦實際完成一1000 W之最大功率追蹤電路，其於隨機挑選的3種遮蔭樣式下之穩態追準確度皆高達99%以上。

When the solar power generation system (SPGS) is partially shaded, its power-voltage characteristic curve will change from a single peak to a complex multi-peak curve. When the traditional maximum power point tracking (MPPT) technology is applied under partially shaded conditions (PSCs), it is easy to fall into a local maximum, and consequently, the SPGS cannot output the maximum power available. Therefore, a technology that can track the global maximum power point (GMPP) is particularly important for the solar power generation system. In this thesis, a neural network-based hybrid global maximum power point tracking (GMPPT) algorithm is developed. When PSC occurs, the first-stage neural network and subsequent calculations are used to estimate the shading pattern and the approximated maximum power point voltage, then the second-stage adaptive variable step-size perturb and observation (VSS P&O) method is used to correctly acquire the GMPP. On the other hand, the neural network is used to estimate the maximum power point voltage of the first interval and then multiplied with the number of solar cell modules in series as the initial operating point under uniform illumination, and then the adaptive VSS P&O method is performed to obtain the exact MPP.
To verify the feasibility and effectiveness of the algorithm proposed in this thesis, this study uses the same platform to perform simulation and compares the obtained results with 8 neural network-based GMPPT algorithms proposed in recent literature in various PSCs. The simulation results show that the average power loss of the proposed method is only 0.12W, and the average tracking time is only 1.48s. To verify the correctness of the proposed technique, a 1000W GMPPT prototyping circuit is also implemented in this thesis, and its GMPP tracking accuracy under three randomly selected shading patterns is all over 99%.

[1] Renewable Energy Policy Network for the 21st Century, Available at: https://www.iea.org/
[2] Kashif Ishaque, Zainal Salam, Amir Shamsudin, Muhammad Amjad, ”A Direct Control Based Maximum Power Point Tracking Method for Photovoltaic System Under Partial Shading Conditions Using Particle Swarm Optimization Algorithm,” Science Direct Applied Energy, Vol. 99, pp. 414-422, (2012).
[3] Kinattingal Sundareswaran, Peddapati Sankar, P. S. R. Nayak, Sishaj P. Simon, and Sankaran Palani, ”Enhanced Energy Output From a PV System Under Partial Shaded Conditions Through Artificial Bee Colony,” IEEE Transactions on Sustainable Energy, Vol. 6, pp. 198-209, (2015).
[4] Karim Kaced, Cherif Larbes, Naeem Ramzan, Moussaab Bounabi, Zine elabadine Dahmane, ”Bat Algorithm Based Maximum Power Point Tracking for Photovoltaic System Under Partial Shading Conditions,” Science Direct Solar Energy, Vol. 158, pp. 490-503, (2017).
[5] M. Yaqoob Javed, Ali Faisal Murtaza, Qiang Ling, Shahid Qamar, M. Majid Gulzar, ”A Novel MPPT Design Using Generalized Pattern Search for Partial Shading,” Science Direct Energy and Buildings, Vol. 133, pp. 59-69, (2016).
[6] Mohamad Amin Ghasemi, Alireza Ramyar, and Hossein Iman-Eini, ”MPPT Method for PV Systems Under Partially haded Conditions by Approximating I–V Curve,” IEEE Transactions on Industrial Electronics, Vol. 65, pp. 3966-3975, (2018).

[7] Seyed Majid Hashemzadeh, ”A New Model-Based Technique for Fast and Accurate Tracking of Global Maximum Power Point in Photovoltaic Arrays Under Partial Shading Conditions,” Science Direct Renewable Energy, Vol. 139, pp. 1061-1076, (2019).
[8] Mustafa Engin Ba¸so˘glu, ”A Fast GMPPT Algorithm Based on PV Characteristic for Partial Shading Conditions,” Multidisciplinary Digital Publishing Institute Electronics, Vol. 8, pp. 1142-1160, (2019).
[9] K. L. Lian, J. H. Jhang, and I. S. Tian, ”A Maximum Power Point Tracking Method Based on Perturb-and-Observe Combined With Particle Swarm Optimization,” IEEE Journal of Photovoltaics, Vol. 4, pp. 626-633, (2014).
[10] Chih-Chiang Hua, Yu-Jun Zhan, ”A Hybrid Maximum Power Point Tracking Method without Oscillations in Steady-State for Photovoltaic Energy Systems,” Multidisciplinary Digital Publishing Institute Energies, Vol. 14, pp. 5590-5605, (2021).
[11] J. Saikrishna Goud, R. Kalpana, and Bhim Singh, ”A Hybrid Global Maximum Power Point Tracking Technique With Fast Convergence Speed for Partial-Shaded PV Systems,” IEEE Transactions on Industrial Applications, Vol. 54, pp. 5367-5376, (2014).
[12] Dhanup S. Pillai, J. Prasanth Ram, A. M. Y. M. Ghias, Md Apel Mahmud, and N. Rajasekar, ”An Accurate, Shade Detection-Based Hybrid Maximum Power Point Tracking Approach for PV Systems,” IEEE Transactions on Power Electronics, Vol. 35, pp. 6594-6608, (2020).

[13] Chakkarapani Manickam, Guru Praanesh Raman, Guru Raghav Raman, Saravana Ilango Ganesan, and Nagamani Chilakapati, ”Fireworks Enriched P&O Algorithm for GMPPT and Detection of Partial Shading in PV Systems,” IEEE Transactions on Power Electronics, Vol. 32, pp. 4432-4443, (2017).
[14] Santi Agatino Rizzo, Giacomo Scelba, ”ANN Based MPPT Method for Rapidly Variable Shading Conditions,” Science Direct Applied Energy, Vol. 145, pp. 124-132, (2015).
[15] Mingxuan Chen, Suliang Ma, Jianwen Wu, Lian Huang, ”Analysis of MPPT Failure and Development of an Augmented Nonlinear Controller for MPPT of Photovoltaic Systems under Partial Shading Conditions,” Multidisciplinary Digital Publishing Institute Applied Sciences, Vol. 7, pp. 95-117, (2017).
[16] K. Punitha, D. Devaraj, S. Sakthivel, ”Artificial Neural Network Based Modified Incremental Conductance Algorithm for Maximum Power Point Tracking in Photovoltaic System Under Partial Shading Conditions,” Science Direct Energy, Vol. 62, pp. 330-340, (2013).
[17] Xiangjian Meng, Feng Gao, Tao Xu, Chenghui Zhang, ”Fast Two-Stage Global Maximum Power Point Tracking for Grid-Tied String PV Inverter Using Characteristics Mapping Principle,” IEEE Journal of Emerging and Selected Topics in Power Electronics, Vol. 10, pp. 564-574, (2022).
[18] Sara Allahabadi, Hossein Iman-Eini, Shahrokh Farhangi, ”Fast Artificial Neural Network Based Method for Estimation of the Global Maximum Power Point in Photovoltaic Systems,” IEEE Transactions on Industrial Electronics, Vol. 69, pp. 5879-5888, (2022).
[19] Hadi M. El-Helw, Ahmed Magdy, Mostafa I. Marei, ”A Hybrid Maximum Power Point Tracking Technique for Partially Shaded Photovoltaic Arrays,” IEEE Access, Vol. 5, pp. 11900-11908, (2017).
[20] Wei Zhang, Guopeng Zhou, Hao Ni, Yunlian Sun, ”A Modified Hybrid Maximum Power Point Tracking Method for Photovoltaic Arrays Under Partially Shading Condition,” IEEE Access, Vol. 7, pp. 160091-160100, (2019).
[21] Syafaruddin, E. Karatepe, T. Hiyama, ”Artificial Neural Network-Polar Coordinated Fuzzy Controller Based Maximum Power Point Tracking Control Under Partially Shaded Conditions,” IET Journals Renewable Power Generation, Vol. 3, pp. 239-253, (2009).
[22] Giovanni Petrone, Carols Andres Ramos-Paja, Giovanni Spagnuolo, ”
Photovoltaic Sources Modeling,” Wiley-IEEE Press, (2017)
[23] Marcelo Gradella Villalva, Jonas Rafael Gazoli, Ernesto Ruppert Filho, ”Comprehensive Approach to Modeling and Simulation of Photovoltaic Arrays,” IEEE Transactions on Power Electronics, Vol. 24, pp. 1198-1208, (2009).
[24] R. Ahmad, Ali F. Murtaza, Hadeed Ahmed Sher, Umar Tabrez Shami, Saheed Olalekan, ”An Analytical Approach to Study Partial Shading Effects on PV Array Supported by literature,” Science Direct Renewable and Sustainable Energy Reviews, Vol. 74, pp. 721-732, (2017).
[25] K. Latha Shenoy, C.Gurudas Nayak, Rajashekar P Mandi, ”Design and Implementation of Interleaved Boost Converter,” International Journal of Engineering and Technology, Vol. 9, pp.496-502, (2017).
[26] R. Buerger, A. Péres, R. Hausmann, R. A. Reiter, A. L. Stankiewicz,
”Ripple Analyze and Design Considerations for An Interleaved Boost Converter (IBC) for A PV Source,” International Conference on Renewable Energies and Power Quality (ICREPQ’14), Cordoba (Spain), pp. 472-477, (2014).

[27] Mehmet Can Kaya, ”Design, Implementation, and Control of a Two–Stage AC/DC Isolated Power Supply with High Input Power Factor and High Efficiency,” Master Dissertation, Department of Electrical and Electronics Engineering, Middle East Technical University, (2008).
[28] Chien-Ming Wang, Chang-Hua Lin, Chien-Min Lu, Jyun-Che Li, ”Analysis, Design, and Realisation of a ZVT Interleaved Boost dc/dc Converter with Single ZVT Auxiliary Circuit,” IET Journals Power Electronics, Vol. 10, pp. 1789-1799, (2017).
[29] 王進德、蕭大全，「類神經網路與模糊控制理論入門」，全華科技圖書，民國96年1月。
[30] Jubaer Ahmed, Zainal Salam, ”An Enhanced Adaptive P&O MPPT for Fast and Efficient Tracking Under Varying Environmental Conditions,” IEEE Transactions on Sustainable Energy, Vol. 9, (2018).
[31] 王俊仁，「適用於太陽能部分遮蔭情形之兩段式最大功率追蹤技術」，國立台灣科技大學電機工程系碩士學位論文，民國108年01月。
[32] Fangrui Liu, Shanxu Duan, Fei Liu, Bangyin Liu, Yong Kang, ”A Variable Step Size INC MPPT Method for PV Systems,” IEEE Transactions on Industrial Electronics, Vol. 55, (2008).
[33] 吳義利，「切換式電源轉換器 : 原理與實用設計技術(實例設計導向)」，文笙總經銷，民國107年10月。
[34] 王順忠，「電力電子學」，臺灣東華書局股份有限公司，民國87年2月。
[35] 呂文隆、張簡士琨、曾國境，「交換式電源設計(第三版)」，全華圖書，民國101年07月。