مدیریت هوشمند مصرف انرژی در گرههای اینترنت اشیاء با دوقلوی دیجیتال و هوش مصنوعی مولد در مواجهه با چالشهای شروع سرد
مریم نورائی آباده
1
(
گروه مهندسی کامپیوتر، واحد بینالمللی اروند، دانشگاه آزاد اسلامی، آبادان، ایران
)
شهره آجودانیان
2
(
دانشکده مهندسي كامپيوتر، واحد نجف آباد، دانشگاه آزاد اسلامي، نجفآباد، ايران
)
کلید واژه: هوش مصنوعی مولد, اینترنت اشیاء, مدیریت هوشمند مصرف انرژی, یادگیری ماشین, دوقلوی دیجیتال,
چکیده مقاله :
مدیریت هوشمند مصرف انرژی در گرههای IoT، با توجه به گسترش سریع این فناوری در حوزههایی مثل خانهها و شهرهای هوشمند، صنعت و سلامت، چالشی بزرگ است. گرههای IoT معمولاً با محدودیت منابع مواجهاند و بهینهسازی مصرف انرژی در آنها حیاتی است. یکی از مشکلات اصلی، بهویژه در شرایط شروع سرد، کمبود دادههای اولیه برای پیشبینی و بهینهسازی مصرف انرژی است که تصمیمگیری دقیق را دشوار میکند. این مقاله با معرفی دوقلوی دیجیتال و هوش مصنوعی مولد راهحلهایی نوآورانه ارائه میدهد. دوقلوی دیجیتال با شبیهسازی رفتار گرههای، عملکرد سیستم را پیشبینی میکند و هوش مصنوعی مولد با تولید دادههای مصنوعی، کمبود دادهها در شروع سرد را جبران میکند. این روشها به کاهش مصرف انرژی و افزایش عمر باتری گرهها کمک میکنند. همچنین، مدلهای یادگیری ماشین و عمیق برای پیشبینی و بهینهسازی مصرف انرژی بررسی و معیارهایی چون دقت، زمان آموزش و پیشبینی مقایسه شدهاند. نتایج آزمایشها نشان میدهند که ترکیب دادههای تولیدشده با مدلهای سنتی باعث کاهش خطاهای پیشبینی و افزایش دقت مدلها میشود. همچنین، تأثیر روش پیشنهادی بر چالش شروع سرد بررسی شده و مشاهده گردید که دادههای مصنوعی میتوانند دقت اولیه مدلهای یادگیری را بهبود بخشند.
چکیده انگلیسی :
Smart energy management in IoT nodes, considering the rapid expansion of this technology in areas such as smart homes, smart cities, industry, and healthcare, presents a significant challenge. IoT nodes are typically resource-constrained, and optimizing energy consumption in them is crucial. One of the main issues, especially in cold start scenarios, is the lack of initial data for predicting and optimizing energy usage, which makes accurate decision-making difficult. This paper introduces innovative solutions by leveraging Digital Twin technology and Generative AI. The Digital Twin simulates the behavior of nodes and predicts system performance, while Generative AI generates synthetic data to compensate for the lack of data in cold start situations. These methods help reduce energy consumption and extend the battery life of nodes. Furthermore, machine learning and deep learning models for energy consumption prediction and optimization are examined, and metrics such as accuracy, training time, and prediction time are compared. Experimental results show that combining generated data with traditional models reduces prediction errors and enhances model accuracy. Additionally, the impact of the proposed approach on the cold start challenge is explored, and it is observed that synthetic data can improve the initial accuracy of learning models.
[1] Zeadally, S., Shaikh, F. K., Talpur, A., and Sheng, Q. Z., "Design architectures for energy harvesting in the Internet of Things," Renewable and Sustainable Energy Reviews, vol. 128, p. 109901, 2020/08/01/ 2020, doi: https://doi.org/10.1016/j.rser.2020.109901.
[2] Mohammadi, R. and Shirmohammadi, Z., "DRDC: Deep reinforcement learning based duty cycle for energy harvesting body sensor node," Energy Reports, vol. 9, pp. 1707-1719, 2023/12/01/ 2023, doi: https://doi.org/10.1016/j.egyr.2022.12.138.
[3] Ahn, J., Kim, D., Ha, R., and Cha, H., "Controlling Action Space of Reinforcement-Learning-Based Energy Management in Batteryless Applications," IEEE Internet of Things Journal, vol. 10, no. 11, pp. 9928-9941, 2023, doi: 10.1109/JIOT.2023.3234905.
[4] Prauzek, M., Paterova, T., Konecny, J., and Martinek, R., "Data-Driven Self-Learning Controller for Power-Aware Mobile Monitoring IoT Devices," Computers, Materials \& Continua, vol. 70, no. 2, pp. 2601--2618, 2022. [Online]. Available: http://www.techscience.com/cmc/v70n2/44649.
[5] Adila, A. S., Husam, A., and Husi, G., "Towards the self-powered Internet of Things (IoT) by energy harvesting: Trends and technologies for green IoT," in 2018 2nd International Symposium on Small-scale Intelligent Manufacturing Systems (SIMS), 16-18 April 2018 2018, pp. 1-5, doi: 10.1109/SIMS.2018.8355305.
[6] Alhasnawi, B. N., Jasim, B. H., Esteban, M. D., and Guerrero, J. M., "A Novel Smart Energy Management as a Service over a Cloud Computing Platform for Nanogrid Appliances," Sustainability, vol. 12, no. 22, p. 9686, 2020. [Online]. Available: https://www.mdpi.com/2071-1050/12/22/9686.
[7] Abd Eldjalil CD, L. K., "Optimal priority-queuing for EV charging-discharging service based on cloud computing. ," presented at the In2017 IEEE International Conference on Communications (ICC) 2017 May 21, 2017.
[8] Shewale, A., Mokhade, A., Funde, N., and Bokde, N. D., "A Survey of Efficient Demand-Side Management Techniques for the Residential Appliance Scheduling Problem in Smart Homes," Energies, vol. 15, no. 8, p. 2863, 2022. [Online]. Available: https://www.mdpi.com/1996-1073/15/8/2863.
[9] Dabbaghjamanesh, M., Mehraeen, S., Kavousi-Fard, A., and Ferdowsi, F., "A New Efficient Stochastic Energy Management Technique for Interconnected AC Microgrids," in 2018 IEEE Power & Energy Society General Meeting (PESGM), 5-10 Aug. 2018 2018, pp. 1-5, doi: 10.1109/PESGM.2018.8586362.
[10] Prauzek, M., Krömer, P., Rodway, J., and Musilek, P., "Differential evolution of fuzzy controller for environmentally-powered wireless sensors," Applied Soft Computing, vol. 48, pp. 193-206, 2016/11/01/ 2016, doi: https://doi.org/10.1016/j.asoc.2016.06.040.
[11] Grieves, M. and Vickers, J., "Digital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex Systems," in Transdisciplinary Perspectives on Complex Systems: New Findings and Approaches, F.-J. Kahlen, S. Flumerfelt, and A. Alves Eds. Cham: Springer International Publishing, 2017, pp. 85-113.
[12] Vacca, J. R., Solving Urban Infrastructure Problems Using Smart City Technologies: Handbook on Planning, Design, Development, and Regulation. Elsevier, 2020.
[13] Ojha, T., Misra, S., Raghuwanshi, N. S., and Poddar, H., "DVSP: Dynamic virtual sensor provisioning in sensor–cloud-based Internet of Things," IEEE Internet of Things Journal, vol. 6, no. 3, pp. 5265-5272, 2019.
[14] Xuezhou Fan a, Y. L. b., "Energy management of renewable based power grids using artificial intelligence: Digital twin of renewables," Solar Energy, 2023.
[15] Wang, Y. and Cai, Z., "Combining Multiobjective Optimization With Differential Evolution to Solve Constrained Optimization Problems," IEEE Transactions on Evolutionary Computation, vol. 16, no. 1, pp. 117-134, 2012, doi: 10.1109/TEVC.2010.2093582.
[16] Prauzek, M., Paterova, T., Konecny, J., and Martinek, R., "Data-Driven Self-Learning Controller for Power-Aware Mobile Monitoring IoT Devices," Computers, Materials & Continua, vol. 70, no. 2, 2022.
[17] Hsu, R. C., Lin, T.-H., and Su, P.-C., "Dynamic energy management for perpetual operation of energy harvesting wireless sensor node using fuzzy Q-learning," Energies, vol. 15, no. 9, p. 3117, 2022.
[18] Zhang, T., Lv, C., Ma, F., Zhao, K., Wang, H., and O'Hare, G. M. P., "A photovoltaic power forecasting model based on dendritic neuron networks with the aid of wavelet transform," Neurocomputing, vol. 397, pp. 438-446, 2020/07/15/ 2020, doi: https://doi.org/10.1016/j.neucom.2019.08.105.