隨著 LED 燈具的普及，智慧照明系統的引入已經成為必然。此
類系統利用先進的感應技術、調光控制、物聯網和自動化等功能，
根據實際需求智能調整照明亮度以實現節能效果。智慧照明系統的
目的是減少能源消耗和電力浪費，降低能源成本並減少對環境的不
良影響。它能夠充分利用 LED 燈具的優勢，最大程度地提升節能效
益。
本研究旨在建立一個自動調光系統，透過配合日光和物聯網技
術，確保照明條件符合各種工作場所需求。在此過程中，本文採用
了前饋神經網路模型加速粒子群演算法的收斂以實現個別調光，從
而達到最佳的照明效果。本系統模擬具有三個照度觀測點和八顆
LED 燈泡的場地進行實驗，研究結果表明，透過本系統最高能夠節
省 43.22 %的功耗。同時，相較於傳統粒子群演算法，所提結合前饋
神經網路之粒子群演算法在收斂速度可改善 43.58 %，精準度部分則
可增加 3.21 %。

With the widespread adoption of LED lighting, the introduction of
smart lighting systems has become inevitable. These systems utilize
advanced sensing technology, dimming controls, internet of things (IoT),
and automation technology to intelligently adjust lighting brightness based
on actual requirements, achieving energy-saving effects. The purpose of
smart lighting systems is to reduce energy consumption, minimize power
waste, lower energy costs, and mitigate adverse environmental impacts.
These systems leverage the advantages of LED lighting to maximize
energy efficiency.
This study aims to establish an automatic dimming system that ensures
lighting conditions meet various operational requirements through the
integration of daylight and IoT technologies. In this process, a Feedforward
Neural Network(FFNN) model is incorporated to accelerate the
convergence of Particle Swarm Optimization(PSO) for individual
dimming, thereby achieving optimal lighting effects. The system is
simulated in a venue with three illuminance measuring points and eight
LED bulbs. The research findings demonstrate that the system can achieve
a maximum power saving of 43.22 %. Moreover, compared to traditional
PSO, the proposed FFNN-integrated PSO method exhibits improved
convergence speed by 43.58 % and enhanced accuracy by 3.21 %.

[1] I. E. Agency, "Net Zero by 2050 A Roadmap for the Global Energy Sector," 2021. [Online]. Available: https://iea.blob.core.windows.net/assets/deebef5d-0c34-4539-9d0c-10b13d840027/NetZeroby2050-ARoadmapfortheGlobalEnergySector_CORR.pdf.
[2] 工業技術研究院, "2021家庭用電資訊百科," 2022. [Online]. Available: https://www.energypark.org.tw/save-energy-resource/books.html?task=convert.getpdf&id=78&filename=123.pdf.
[3] S. Yoo, J. Kim, C.-Y. Jang, and H. Jeong, "A sensor-less LED dimming system based on daylight harvesting with BIPV systems," Optics Express, vol. 22, no. 101, pp. A132-A143, 2014.
[4] 工業技術研究院，李麗玲, "LED智慧照明技術及應用," 2021. [Online]. Available: https://www.taesco.org.tw/wp-content/uploads/2022/04/LED%E6%99%BA%E6%85%A7%E7%85%A7%E6%98%8E%E6%8A%80%E8%A1%93%E5%8F%8A%E6%87%89%E7%94%A8.pdf.
[5] 經濟部能源局, "室內照明燈具節能標章能源效率基準與標示方法修正規定," 2020. [Online]. Available: https://www.energylabel.org.tw/applying/efficiency/upt.aspx?cid=27&Con=1&uid=0&year=&month=&day=&key=&subID=217.
[6] S. H. Lee and J. K. Kwon, "Distributed dimming control for LED lighting," Optics express, vol. 21, no. 106, pp. A917-A932, 2013.
[7] X. Wang and J.-P. Linnartz, "Intelligent illuminance control in a dimmable LED lighting system," Lighting Research & Technology, vol. 49, no. 5, pp. 603-617, 2017.
[8] A. Cziker, M. Chindris, and A. Miron, "Implementation of fuzzy logic in daylighting control," in 2007 11th International Conference on Intelligent Engineering Systems, 2007: IEEE, pp. 195-200.
[9] M. Fakhari, R. Fayaz, and R. Lollini, "The Impact of Evaluated Daylight to the Total Light Ratio on the Comfort Level in Office Buildings," Buildings, vol. 12, no. 12, p. 2161, 2022.
[10] M. Petkovic et al., "Smart dimmable LED lighting systems," Sensors, vol. 22, no. 21, p. 8523, 2022.
[11] C.-H. Kim, K.-H. Lee, and K.-S. Kim, "Evaluation of Illuminance Measurement Data through Integrated Automated Blinds and LED Dimming Controls in a Full-Scale Mock-up," Energies, vol. 13, no. 12, p. 3238, 2020.
[12] R. Mistrick, "Analysis of daylight responsive dimming system performance," Building and Environment, vol. 34, no. 3, pp. 231-243, 1998.
[13] A. Choi and M. Sung, "Development of a daylight responsive dimming system and preliminary evaluation of system performance," Building and environment, vol. 35, no. 7, pp. 663-676, 2000.
[14] I.-T. Kim, Y.-S. Kim, H. Nam, and T. Hwang, "Advanced Dimming Control Algorithm for Sustainable Buildings by Daylight Responsive Dimming System," Sustainability, vol. 10, no. 11, p. 4087, 2018.
[15] I.-T. Kim, Y.-S. Kim, M. Cho, H. Nam, A. Choi, and T. Hwang, "High-performance accuracy of daylight-responsive dimming systems with illuminance by distant luminaires for energy-saving buildings," Energies, vol. 12, no. 4, p. 731, 2019.
[16] W. Si, H. Ogai, T. Li, and K. Hirai, "A novel energy saving system for office lighting control by using RBFNN and PSO," in IEEE 2013 Tencon-Spring, 2013: IEEE, pp. 347-351.
[17] W. Si, H. Ogai, K. Hirai, H. Takahashi, and M. Ogawa, "An improved PSO method for energy saving system of office lighting," in SICE Annual Conference 2011, 2011: IEEE, pp. 1533-1536.
[18] L. Mu, Z. Wang, D. Wu, L. Zhao, and H. Yin, "Prediction and evaluation of fuel properties of hydrochar from waste solid biomass: Machine learning algorithm based on proposed PSO–NN model," Fuel, vol. 318, p. 123644, 2022.
[19] 中央氣象局, "觀測資料查詢系統." [Online]. Available: https://e-service.cwb.gov.tw/HistoryDataQuery/.
[20] 中央氣象局, "觀測資料查詢系統說明." [Online]. Available: https://e-service.cwb.gov.tw/HistoryDataQuery/downloads/Readme.pdf.
[21] 郭俊聲, "照度與日射量轉換模式探討." [Online]. Available: https://tpl.ncl.edu.tw/NclService/pdfdownload?filePath=lV8OirTfsslWcCxIpLbUfhqD9W-i3vsGsh2y0wCRuAm5Z5ALYuM6rGCiokqpURMz&imgType=Bn5sH4BGpJw=&key=uEUjXERbGu3M53l7H4cJIq9L8VeQe7jhGsf26XH3kW4eVVU9OyINO4qBZJhLTxWd&xmlId=0005342111.
[22] UPRtek, "MK350S Premium," 2020. [Online]. Available: https://www.uprtek.com/downloads/MK350S_Premium_Specification_TCH_202007.pdf.
[23] IKEA, "TRÅDFRI Led智慧燈泡e27 1055流明." [Online]. Available: https://www.ikea.com.tw/zh/products/light-sources-and-smart-lighting/smart-lighting/tradfri-art-20489749.
[24] IKEA, "TRÅDFRI 閘道器." [Online]. Available: https://www.ikea.com.tw/zh/products/connectivity-and-control/connectivity-and-control/tradfri-art-30445260.
[25] R. P. Foundation, "Raspberry Pi 4 Model B." [Online]. Available: https://www.raspberrypi.com/products/raspberry-pi-4-model-b/.
[26] J. Kennedy and R. Eberhart, "Particle swarm optimization," in Proceedings of ICNN'95-international conference on neural networks, 1995, vol. 4: IEEE, pp. 1942-1948.
[27] S. Agatonovic-Kustrin and R. Beresford, "Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research," Journal of pharmaceutical and biomedical analysis, vol. 22, no. 5, pp. 717-727, 2000.
[28] C. Lv et al., "Levenberg–Marquardt backpropagation training of multilayer neural networks for state estimation of a safety-critical cyber-physical system," IEEE Transactions on Industrial Informatics, vol. 14, no. 8, pp. 3436-3446, 2017.

[29] H. Yonaba, F. Anctil, and V. Fortin, "Comparing sigmoid transfer functions for neural network multistep ahead streamflow forecasting," Journal of hydrologic engineering, vol. 15, no. 4, pp. 275-283, 2010.
[30] R. Tibshirani, "Regression shrinkage and selection via the lasso," Journal of the Royal Statistical Society: Series B (Methodological), vol. 58, no. 1, pp. 267-288, 1996.
[31] MATLAB, "LASSO." [Online]. Available: https://www.mathworks.com/help/stats/lasso.html.
[32] S. Ruder, "An overview of gradient descent optimization algorithms," arXiv preprint arXiv:1609.04747, 2016.
[33] A. Ranganathan, "The levenberg-marquardt algorithm," Tutoral on LM algorithm, vol. 11, no. 1, pp. 101-110, 2004.
[34] M. T. Hagan and M. B. Menhaj, "Training feedforward networks with the Marquardt algorithm," IEEE transactions on Neural Networks, vol. 5, no. 6, pp. 989-993, 1994.
[35] D. W. Marquardt, "An algorithm for least-squares estimation of nonlinear parameters," Journal of the society for Industrial and Applied Mathematics, vol. 11, no. 2, pp. 431-441, 1963.
[36] 中華民國經濟部, "CNS12112 Z1044照度標準." [Online]. Available: https://www.prodigital.com.tw/DOC/CNS12112.pdf.