Advanced Warehouse Energy Storage System Control Using Deep Supervised and Reinforcement Learning
Doctoral thesis
Published version
Permanent lenke
https://hdl.handle.net/11250/3099396Utgivelsesdato
2023Metadata
Vis full innførselSamlinger
Originalversjon
Opalic, S. M. (2023). Advanced Warehouse Energy Storage System Control Using Deep Supervised and Reinforcement Learning. [Doctoral Dissertation]. University of Agder.Sammendrag
The world is undergoing a shift from fossil fuels to renewable energy sources due to the threat of global warming, which has led to a substantial increase in complex buildingintegrated energy systems. These systems increasingly feature local renewable energy production and energy storage systems that require intelligent control algorithms. Traditional approaches, such as rule-based algorithms, are dependent upon timeconsuming human expert design and maintenance to control the energy systems efficiently. Although machine learning has gained increasing amounts of research attention in recent years, its application to energy cost optimization in warehouses still remains in a relatively early stage. Suggested newer approaches are often too complex to implement efficiently, very computationally expensive, or lacking in performance. This Ph.D. thesis explores, designs, and verifies the use of deep learning and reinforcement learning approaches to solve the bottleneck of human expert resource dependency with respect to efficient control of complex building-integrated energy systems. A technologically advanced smart warehouse for food storage and distribution is utilized as acase study. The warehouse has a commercially available Intelligent Energy ManagementSystem (IEMS).
Består av
Paper I: Opalic, S. M., Goodwin, M., Jiao, L., Nielsen, H. K. & Kolhe, M. L. (2019). Modelling of compressors in an industrial CO2-based operational cooling system using ANN for energy management purposes. 20th International Conference on Engineering Applications of Neural Networks (EANN), 43-54. doi: 10.1007/978-3-030-20257-64. Accepted manuscript. Full-text is not available in AURA as a separate file.Paper II: Opalic, S. M., Goodwin, M., Jiao, L., Nielsen, H. K., Pardiñas, Á. Á., Hafner, A., Kolhe, M. L. (2020). ANN modelling of CO2 refrigerant cooling system COP in a smart warehouse. Journal of Cleaner Production, 260, 1-10. doi: 10.1016/j.jclepro.2020.120887. Accepted manuscript. Full-text is available in AURA as a separate file: https://hdl.handle.net/11250/3058108
Paper III: Opalic, S. M., Goodwin, M., Jiao, L., Nielsen, H. K. & Kolhe, M. L. (2020). A Deep Reinforcement Learning scheme for Battery Energy Management. 5th International Conference on Smart and Sustainable Technologies (SpliTech), 1-6. doi: 10.23919/SpliTech49282.2020.9243797. Accepted manuscript. Full-text is not available in AURA as a separate file.
Paper VI: Opalic, S. M., Goodwin, M., Jiao, L., Nielsen, H. K. & Kolhe, M. L. (2022). Augmented Random Search with Artificial Neural Networks for energy cost optimization with battery control. Journal of Cleaner Production, 380, 1-11. doi: 10.1016/j.jclepro.2022.134676. Accepted manuscript. Full-text is not available in AURA as a separate file.
Paper V: Opalic, S. M., Palumbo, F., Goodwin, M., Jiao, L., Nielsen, H. K. & Kolhe, M. L. (2023). COST-WINNERS: COST reduction WIth ANN-ARS for simultaneous thermal and electrical energy storage control. Journal of Energy Storage, 72, 1-9. doi: 10.1016/j.est.2023.108202. Accepted manuscript. Full-text is not available in AURA as a separate file.