Smart supply-side management of optimal hydro reservoirs using the water/energy nexus concept: A hydropower pinch analysis

• Published in: Applied energy Vol. 281; p. 116136
• Main Authors: Tayerani Charmchi, Amir Saman, Ifaei, Pouya, Yoo, ChangKyoo
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-01-2021
• Subjects: Climate change; Deep learning; Hydropower pinch analysis; Optimal water management; Renewable energy recovery; Water/energy nexus; Deep learning; Climate change; Water/energy nexus; Hydropower pinch analysis; Optimal water management; Renewable energy recovery
• ISSN: 0306-2619; 1872-9118
• DOI: 10.1016/j.apenergy.2020.116136

[Display omitted]
•A novel smart framework was detailed for optimal management of hydro reservoirs.•New graphical tools were accompanied by an iterative numerical method.•The advanced model predicted sources of Karkheh hydro reservoir with R2 = 97.3%.•37.2 GWh extra hydroelectricity was produced in Karkheh hydro reservoir in 2018.•1.66 Bm3 water was saved in Karkheh reservoir after meeting all demands in 2018.
A novel, smart, supply-side management approach is proposed for optimal operation of multi-purpose hydro reservoirs using the water/energy nexus concept and introducing a hydropower pinch analysis (HyPoPA). The nexus among water and energy sources and sinks are considered to develop hydropower composite curves, grand composite curves, and continuous composite curves. These graphical tools are accompanied by hydropower cascade tables to facilitate numerical analysis of conservation and recovery of water resources at high resolution. The minimum hydro storage is targeted, and three management cases are determined by obtaining the surplus (or deficit) head of a hydro reservoir in successive operational years in unreliable, reliable, and self-sufficient cases. The effects of climate change are predicted using smart algorithms to manage varying downstream energy and water sinks under optimal conditions by HyPoPA. Two prediction scenarios are developed to mimic annual operation and online monitoring cases using advanced neural networks. Karkheh hydro reservoir serves as a case study to verify smart HyPoPA. The results showed that the sources were successfully predicted employing a hybrid long short-term memory and gated recurrent unit network in 2018 (R2 = 97.3%, MAPE = 15.9%), which was a dry year when reservoir water levels fell in a non-reliable case with deficit head. The Karkheh reservoir produced 37.2 GWh more hydroelectricity and saved 1.66 billion m3 of water after meeting all water requirements using the smart HyPoPA in the target year.