隨著當代用電量的需求逐漸地提高，能源消耗過度的議題與提升能源使用效
率的研究因而開始受到重視。然而，若僅依靠人力進行能源的管理與配置，將大幅降
低能源使用效率與決策的即時性。近年來，有鑑於物聯網技術發展的蓬勃發展，產、
官、學、業界開始將物聯網技術應用在民眾的生活中，如智慧農場、智慧家庭與智慧
電表等。透過物聯網技術監控環境的變化與影響，不僅可以有效地使用能源，同時亦
降低能源管理的成本。因此，物聯網技術與能源管理的結合，儼然成為提升能源使用
效率的議題上重要的解決方法之一。在本論文中，我們針對智慧農場的能源管理情
境，設計一套結合物聯網技術的智慧調光系統，此系統由三個子系統所組成，分別為
物聯網設備端、使用者端以及雲端。為了達到增加增加農場管理的效率以及加強調光
的精確性及能源效率之目的，此系統同時導入及整合低功耗通訊技術傳輸、能源監控
裝置、雲端整合性服務平台以及類神經網路演算法等四項功能模組。
在本論文中設計的三個子系統中，首先，物聯網設備端透過多種異質感測器(如：
色彩感測器、溫溼度感測器、電子計量器等等)感測農場內環境的變化後，閘道器
藉由低功耗通訊傳輸技術MQTT將感測資料上傳至雲端平台進行資料圖形化呈現與
資料儲存，並且可以接收由使用者端與雲端控制裝置的開關以及調整LED的亮度；
再者，在雲端方面，我們選擇Microsoft開發的Azure雲端整合性平台，於虛擬機中建
構Node-RED平台作為本系統的開發工具，連接三個子系統。最後，在使用者端中，使
用者能夠即時透過Node-RED平台中設計的圖形化界面存取需要的資訊並進行遠端的
控制與管理，進而實現了物聯網雙向傳輸架構以及垂直整合的服務。此外，本論文導
入GRNN神經網路以協助計算室內光與陽光互補的最佳配置策略。透過GRNN神經網
路，系統將判斷當前環境下是否需要調整光源，並針對個別區域計算出最佳的PWM調
光值，達到於調光的即時性、能源消耗以及維持光源供給的穩定性等三者間的平衡。
最後，透過本論文的系統設計，我們期望以基於物聯網技術的整合性服務提供使用者
隨時隨地能遠端監控與管理農場內的情況，並依照外部光源變化能即時、精準的進行
互補調光，以最低的能源消耗成本實現亮度供給的最佳化。在未來的研究中，本論文
期能在智慧家庭的環境下，將本文設計的系統結合能源管理系統(HEMS)的應用，達到
同時減少家用照明設備的能源消耗以及優化光源供應的目的。

With the increasing demand of electricity consumption, the issue of excessive energy
consumption and the study of improving energy eciency have been receive more atten-
tion. However, relying solely on human resources for energy management and allocation
will signicantly reduce the eciency of energy use and the immediacy of decision-making.
In recent years, in view of the rapid development of Internet of things technology, indus-
try, government, academic and industrial sectors have begun to apply IoT technology
in people's life, such as smart farms, smart homes and smart meters,etc. Monitoring
environmental changes and impacts through IoT technology can not only eectively use
energy, but also reduce the cost of energy management. Therefore, the combination of
Internet of things technology and energy management has become one of the important
solutions to improve energy eciency . In this paper, we design a smart dimming system
that combines IoT technology for the energy management scenario of smart farms. This
system consists of three subsystems, namely the IoT device end, the user end and the
cloud. In order to increase the eciency of farm management and enhance the accuracy
of light regulation and energy eciency, this system simultaneously introduces and inte-
grates four functional modules, including low-power communication technology , energy
monitoring device, cloud integrated service platform and neural network algorithm.
In this thesis design of three subsystems, rst of all, the Internet of things devices
through a variety of heterogeneous sensors (such as: colour sensor, the temperature and
humidity sensors, electronic meter, etc.) after sensing changes in the environment on the
farm, gateway will upload sensing data to the cloud platform by low power communication
transmission technology MQTT for graphical representations and data storage, and it
can receive by the end user and the cloud control switch and adjust the LED brightness ;
Furthermore, in the cloud-end, we choose the Azure cloud platform developed by Microsoft
and build Node-RED platform in the virtual machine as the development tool of the
system, which connect three subsystems. Finally, in the user -end, users can access the
required information, control and manage IoT devices remotely through the graphical
interface designed in Node-RED platform, thus realizing the bidirectional transmission
architecture of the IoT and vertically integrated services. In addition, GRNN neural
network is introduced in this paper to help calculate the optimal allocation strategy of
light and sunlight complementing each other. Through GRNN neural network, the system
will judge whether the light source needs to be adjusted in the current environment, and
calculate the best PWM dimming value for certain regions, so as to achieve the balance
among the immediacy of dimming, energy consumption and stability of the light source
supply. Finally, through the system design of this paper, we expect to provide users
with integrated services based on the Internet of things technology to remotely monitor
and manage the situation in the farm at any time and anywhere, and make timely and
accurate complementary dimming according to the change of external light source, so as
to optimize the brightness supply with the lowest energy consumption cost. In the future
research, this paper is expected to combine the system designed in this paper with the
application of energy management system (HEMS) in the smart home environment, so as
to achieve the purpose of simultaneously reducing the energy consumption of household
lighting equipment and optimizing the supply of light source.

1 簡介1
1.1 前言. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 研究動機. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 研究背景與相關文獻探討5
2.1 智慧農場以及精準農業. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 遠端照明控制系統介紹. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.3 調光系統演算法. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4 深度學習. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4.1 多層感知器Multilayer Perceprton (MLP) . . . . . . . . . . . . . . . 10
2.4.2 放射狀基函數網路Radial Basis Function Neuron Network(RBFNN) 12
2.4.3 機率類神經網路Probabilistic Neural Network(PNN) . . . . . . . . . 13
2.4.4 廣義迴歸神經網路Generalized regression neural network(GRNN) . 14
3 研究方法15
3.1 問題描述與系統需求. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.1.1 物聯網設備端. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.1.2 雲端. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.3 使用者端. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2 基於GRNN之調光系統演算法. . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2.1 GRNN數學理論背景. . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2.2 GRNN調光演算法運算架構. . . . . . . . . . . . . . . . . . . . . . 23
3.2.3 調光演算法. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4 實作與結果分析27
4.1 實驗環境. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.1.1 實驗硬體模組. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.1.2 系統平台. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.2 實驗設計. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.3 實驗結果分析. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.3.1 GRNN調光演算法中寬度係數效益比較. . . . . . . . . . . . . . . . 35
4.3.2 不同亮度誤差區間效能比較. . . . . . . . . . . . . . . . . . . . . . 35
4.3.3 PNN調光演算法與GRNN調光演算法效益比較. . . . . . . . . . . . 39
4.3.4 GRNN調光演算法在不同天氣下調光效益比較. . . . . . . . . . . . 44
5 結論與未來研究 56

[1] L. Atzori, A. Iera, and G. Morabito, \The internet of things: A survey," Comput.
Netw., vol. 54, no. 15, pp. 2787{2805, Oct. 2010.
[2] J. Lin, W. Yu, N. Zhang, X. Yang, H. Zhang, and W. Zhao, \A survey on internet of
things: Architecture, enabling technologies, security and privacy, and applications,"
IEEE Internet of Things Journal, vol. 4, no. 5, pp. 1125{1142, Oct 2017.
[3] O. Elijah, T. A. Rahman, I. Orikumhi, C. Y. Leow, and M. N. Hindia, \An overview
of internet of things (iot) and data analytics in agriculture: Benets and challenges,"
IEEE Internet of Things Journal, vol. 5, no. 5, pp. 3758{3773, Oct 2018.
[4] C. Yoon, M. Huh, S. Kang, J. Park, and C. Lee, \Implement smart farm with
iot technology," in Proceedings of 2018 20th International Conference on Advanced
Communication Technology (ICACT), Chuncheon-si Gangwon-do,Korea, Feb 2018,
pp. 749{752.
[5] A. Somov, D. Shadrin, I. Fastovets, A. Nikitin, S. Matveev, I. seledets, and
O. Hrinchuk, \Pervasive agriculture: Iot-enabled greenhouse for plant growth con-
trol," IEEE Pervasive Computing, vol. 17, no. 4, pp. 65{75, Oct 2018.
[6] T. Namgyel, C. Khunarak, S. Siyang, T. Pobkrut, J. Norbu, and T. Kerdcharoen,
\Eects of supplementary led light on the growth of lettuce in a smart hydroponic
system," in Proceedings of 2018 10th International Conference on Knowledge and
Smart Technology (KST), Chiangmail,Tailand, Jan 2018, pp. 216{220.
[7] T. Namgyel, S. Siyang, C. Khunarak, T. Pobkrut, J. Norbu, T. Chaiyasit, and
T. Kerdcharoen, \Iot based hydroponic system with supplementary led light for smart home farming of lettuce," in Proceedings of The 15th International Conference
on Electrical Engineering/Electronics, Computer, Telecommunications and Informa-
tion Technology (ECTI-CON), Chiang Rai Tailand, July 2018, pp. 221{224.
[8] M. Roopaei, P. Rad, and K. R. Choo, \Cloud of things in smart agriculture: In-
telligent irrigation monitoring by thermal imaging," IEEE Cloud Computing, vol. 4,
no. 1, pp. 10{15, Jan 2017.
[9] A. L. Diedrichs, F. Bromberg, D. Dujovne, K. Brun-Laguna, and T. Watteyne, \Pre-
diction of frost events using machine learning and iot sensing devices," IEEE Internet
of Things Journal, vol. 5, no. 6, pp. 4589{4597, Dec 2018.
[10] A. M. Abuleil, G. W. Taylor, and M. Moussa, \An integrated system for mapping
red clover ground cover using unmanned aerial vehicles: A case study in precision
agriculture," in Proceedings of The 12th Conference on Computer and Robot Vision,
Halifax,Canada, June 2015, pp. 277{284.
[11] N. Verma and A. Jain, \Optimized automatic lighting control in a hotel building for
energy eciency," in Proceedings of The International Conference on Power Energy,
Environment and Intelligent Control (PEEIC), Greater Noida,India, April 2018, pp.
168{172.
[12] Y. Tang, Q. Chen, P. Ju, Y. Jin, F. Shen, B. Qi, and Z. Xu, \Research on load char-
acteristics of energy-saving lamp and led lamp," in Proceedings of IEEE International
Conference on Power System Technology (POWERCON), Wollongong,Australia,
Sep. 2016, pp. 1{5.
[13] R. Zhang and H. S. Chung, \A triac-dimmable led lamp driver with wide dimming
range," in Proceedings of The Twenty-Eighth Annual IEEE Applied Power Electronics
Conference and Exposition, Long Beach,CA,USA, March 2013, pp. 840-847.
[14] S. Mahadeokar and M. Sardeshmukh, \Energy ecient pwm dimmable smart digital
led driver," in Proceedings of The 2015 International Conference on Energy Systems
and Applications, Pune,India, Oct 2015, pp. 306{311.
[15] H. Jung, J. Kim, B. Lee, and D. Yoo, \A new pwm dimmer using two active switches
for ac led lamp," in Proceedings of The 2010 International Power Electronics Con-
ference - ECCE ASIA -, Sapporo, Japan, June 2010, pp. 1547-1551.
[16] J. Erdelyi and P. Cicak, \Survey on communication in internet of things environment," in Proceedings of The 16th International Conference on Emerging eLearning
Technologies and Applications (ICETA), Stary Smokovec, Slovakia, Nov 2018, pp.
149{156.
[17] Z. Shelby, K. Hartke, and C. Bormann, \The Constrained Application
Protocol (CoAP)," RFC 7252, Jun. 2014. [Online]. Available: https://rfc-
editor.org/rfc/rfc7252.txt
[18] E. Rescorla and N. Modadugu, \Datagram Transport Layer Security Version 1.2,"
RFC 6347, Jan. 2012. [Online]. Available: https://rfc-editor.org/rfc/rfc6347.txt
[19] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, and M. Ayyash, \Internet
of things: A survey on enabling technologies, protocols, and applications," IEEE
Communications Surveys Tutorials, vol. 17, no. 4, pp. 2347{2376, Fourthquarter
2015.
[20] G. Boscarino and M. Moallem, \Daylighting control and simulation for led-based
energy-ecient lighting systems," IEEE Transactions on Industrial Informatics,
vol. 12, no. 1, pp. 301{309, Feb 2016.
[21] S. Attarchi and M. Moallem, \Set-point control of led luminaires for daylight har-
vesting," in Proceedings of The 5th International Conference on Control, Instrumen-
tation, and Automation, Shiraz, Iran, Nov 2017, pp. 244{248.
[22] J. Jiang and M. Moallem, \Development of greenhouse led system with redlblue
mixing ratio and daylight control," in Proceedings of IEEE Conference on Control
Technology and Applications, Copenhagen, Denmark, Aug 2018, pp. 1197{1202.
[23] A. G. Parlos, K. T. Chong, and A. F. Atiya, \Application of the recurrent multilayer
perceptron in modeling complex process dynamics," IEEE Transactions on Neural
Networks, vol. 5, no. 2, pp. 255{266, March 1994.
[24] Lu Yingwei, N. Sundararajan, and P. Saratchandran, \Performance evaluation of a
sequential minimal radial basis function (RBF) neural network learning algorithm,"
IEEE Transactions on Neural Networks, vol. 9, no. 2, pp. 308{318, March 1998.
[25] D. F. Specht, \Probabilistic neural networks," Neural Networks, vol. 3, no. 1, pp.
109 { 118, 1990.
[26] D. F. Specht, \A general regression neural network," IEEE Transactions on Neural
Networks, vol. 2, no. 6, pp. 568{576, Nov 1991.
[27] G. Dudek, \Ensembles of general regression neural networks for short-term electricity
demand forecasting," in Proceedings of The 18th International Scientic Conference
on Electric Power Engineering (EPE), Kouty nad Desnou,Czech Republic, May 2017,
pp. 1{5.
[28] L. Rutkowski, \Generalized regression neural networks in time-varying environment,"
IEEE Transactions on Neural Networks, vol. 15, no. 3, pp. 576{596, May 2004.