Multi-Objective optimization of solar park design under climatic uncertainty

• Published in: Solar energy Vol. 231; pp. 958 - 969
• Main Authors: Barros, E.G.D., Van Aken, B.B., Burgers, A.R., Slooff-Hoek, L.H., Fonseca, R.M.
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.01.2022; Pergamon Press Inc
• Subjects: Agricultural land; Climate change; Design; Design optimization; Economic analysis; Economic models; Electricity pricing; Energy demand; Land use; Multiple objective analysis; Optimization; Parks; Parks & recreation areas; Performance assessment; Performance evaluation; Solar energy; Solar farms; Uncertainty; Workflow
• ISSN: 0038-092X; 1471-1257
• DOI: 10.1016/j.solener.2021.12.026

•Computer-assisted model-based optimization framework for solar park design.•Multi-objective optimization of solar park designs enables multiple land use.•Ground irradiance is used as proxy for agricultural performance indicator.•Randomized combinations of real weather data accounts for meteorological variation.•Farming area is significantly doubled without impact on power yield. The scarcity of land near energy demand poses the challenge of designing multi-functional solar parks in terms of land use in some countries. This requires solutions accounting for multiple conflicting objectives, e.g., power generation and multi-functional use of the land (agricultural, construction, ecological). Moreover, the performance of solar park projects in terms of these criteria is subject to uncertainties, e.g., meteorological aspects impacted by climate change, electricity prices, grid infrastructure availability. In this work we present a framework for multi-objective optimization under uncertainty to aid in the development of smart solar park configurations accounting for multi-purpose land use. A solar park simulator and a techno-economic model are combined to evaluate key performance indicators serving as objective functions. Meteorological uncertainty throughout the park lifetime is characterized through an ensemble of scenarios generated based on the variability of historical data, instead of the current practice of assessing the performance of solar park using a deterministic profile of an average meteorological year. The developed workflow is demonstrated through a case study where power yield and agricultural land use are two conflicting objectives being optimized with the orientation, tilt angle, spacing and height of the modules as the optimization variables. A series of optimization experiments with varying importance weights between the objectives is performed. Obtained solutions show solar park designs which double the available farming area without compromising the levelized cost of energy. These results showcase the value generated through an integrated framework for multi-objective optimization under uncertainty leading to optimized solar parks.