Clustering based assessment of cost, security and environmental tradeoffs with possible future electricity generation portfolios. (15th July 2020)

• Record Type: Journal Article
• Authors: Tanoto, Yusak; Haghdadi, Navid; Bruce, Anna; MacGill, Iain
• Is Part Of: Applied energy. Volume 270(2020)
• Publication Date: 2020-07-15
• Subjects: Clustering analysis -- Electricity industry -- Energy trilemma -- Generation portfolios -- Reliability -- Tradeoffs; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2020.115219
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: Near-optimal future generation portfolios considering cost uncertainties are mapped. Clustering techniques are applied to group technology mixes with similar patterns. Reliability-cost-emissions outcomes of the clustered technology mixes are compared. A wide range of technology mixes show reliable, low cost options to lower emissions. Abstract: The electricity sector has a key role to play in the sustainable energy transition. The falling costs of wind and solar PV have added to both the opportunities yet also challenges of balancing sometimes competing industry objectives of costs, security, and environmental impacts. This paper presents novel techniques for assessing possible future industry generation portfolios in three ways: (1) incorporating explicit metrics for energy trilemma objectives into modelling, (2) using the optimization process of evolutionary programming to map the solution space of 'high performing', near least-cost, portfolio solutions, and (3) applying boundary min–max cases and clustering to categorize these varied portfolios to better facilitate planning and policy making. We use an open-source evolutionary programming tool, National Electricity Market Optimiser, to assess possible future generation portfolios for Indonesia's Java-Bali interconnected power system. Our findings highlight the wide range of possible portfolios that might potentially deliver similar total industry costs, and their different security and environmental implications. In particular, additional solar photovoltaic deployment appears a low-risk opportunity to reduce costs and emissions compared to more fossil-fuel oriented mixes. Our novel techniques may be useful for the energy modelling community seeking to better understand and communicate complex, uncertain, and multi-dimensional choices for electricity industry planning.