A simplified optimization model to short-term electricity planning

• Published in: Energy (Oxford) 2015-12, Vol.93, p.2126-2135
• Main Authors: Pereira, Sérgio, Ferreira, Paula, Vaz, A.I.F.
• Format: Article
• Language: English
• Subjects: Electricity planning; Optimization model; Quadratic penalty function; Renewable power plants; Thermal power plants
• ISSN: 0360-5442
• DOI: 10.1016/j.energy.2015.10.040

Short-term optimization models, usually applied to traditional problems like UC (unit commitment) and economic dispatch problem, are essential tools for the planning and operation of power systems. However, the large number of variables and restrictions, necessary for a good and more accurate representation of any electricity system, require high computational resources, frequently resulting in high computation times. This study proposes a simplified approach of a model for the electricity planning of power plants allocation based on the available resources. The model resources to quadratic penalty functions and avoid on/off binary variables. The approach is then supported on a non-linear optimization model able to solve this electricity planning problem in shorter computation times, with solutions close to the ones obtained with more complex models. The model is fully described and tested under different scenarios of an electricity system comprising thermal, wind, and hydropower plants. The results were compared to the ones obtained with a more complex model, analysing the main differences obtained for cost, CO2 emissions and of wind power impacts on this electricity system. The most remarkable advantage of the simplified model comes from the significant reduction on computational time needed for state-of-the-art optimization solvers to provide an optimal solution, comparatively to mixed integer models. •A simplified unit commitment model with quadratic penalty functions is proposed.•The model is applied to a system comprising thermal, wind, and hydropower plants.•The model and results are compared to the ones obtained with a more complex model.•The simplified model reduces significantly the required computational time.•The simplified model is an effective approach for the analysis of complex systems.