Title | Optimal design, operational controls, and data-driven machine learning in sustainable borehole heat exchanger coupled heat pumps: Key implementation challenges and advancement opportunities |
---|---|
ID_Doc | 33785 |
Authors | Ahmed, N; Assadi, M; Ahmed, AA; Banihabib, R |
Title | Optimal design, operational controls, and data-driven machine learning in sustainable borehole heat exchanger coupled heat pumps: Key implementation challenges and advancement opportunities |
Year | 2023 |
Published | |
Abstract | The integration of technologies has made it possible to develop optimal operating conditions at reduced costs, which results in a more sustainable energy transition away from fossil fuels and a step closer towards net-zero emission buildings (NZEB) for sustainable development. In recent years, ground source heat pump has gained recognition as an established thermal technology that can be integrated into smart energy systems to support the sudden rise in energy demand, flatten the quick changes in the supply side, and lower energy costs. On the user side, low-temperature heating and high-temperature cooling borehole coupled heat pumps (BCHP) have gained popularity due to its excellent performance in terms of energy efficiency, sustainability, and simplicity of inte-gration with renewable resources. As a green solution for building space heating/cooling, BCHP systems have the potential to considerably contribute to the CO2 reduction milestones, but they are still underutilized, mostly because of their high initial investment costs. The applications of automation, data retrieval, smart decision making, control optimization, modeling and monitoring, are recent areas where data driven AI algorithms are becoming increasingly significant. AI approaches can help BCHP become more intelligent and offer new op-portunities for studying heating and cooling systems. While much research is conducted to improve the design of borehole heat exchanger (BHEx) based heat pump, an efficient control approach is equally essential to achieving long-term performance and a shorter payback period. The objective of the current study is to identify the po-tential of most recent innovations in the field of data driven machine learning techniques to enhance BCHP operations and performance predictions to meet NZEBs. The explicit implementation challenges linked with BCHPs modeling are pointed out and the requirements needed for setting BCHP control algorithms are presented. Various methods found in the literature studies to come up with more accurate modeling and optimized control for BCHPs, with a special interest for the ones based on data driven machine learning algorithms such as artificial neural networks (ANN), are reviewed, categorized, and their advantages along with limitations are addressed. The latest developments in machine learning algorithms and how they have been utilized in heating/cooling applications are reviewed critically and their significance for Hybrid-BCHP control optimization is presented. Opportunities and limitations associated with their physical implementation on real-time heating/cooling sys-tems are also discussed. In summary, data driven machine learning algorithms can not only be implemented for modeling of BHEx performance, but also can be applied to design and optimization of operational control of Hybrid-BCHP system, while each component's aspect might be intricately related. |
http://manuscript.elsevier.com/S0973082623000753/pdf/S0973082623000753.pdf |
No similar articles found.