Guevara, E., Babonneau, F., Homem-de-Mello, T., & Moret, S. (2020). A machine learning and distributionally robust optimization framework for strategic energy planning under uncertainty. Appl. Energy, 271, 18 pp.
Abstract: This paper investigates how the choice of stochastic approaches and distribution assumptions impacts strategic investment decisions in energy planning problems. We formulate a two-stage stochastic programming model assuming different distributions for the input parameters and show that there is significant discrepancy among the associated stochastic solutions and other robust solutions published in the literature. To remedy this sensitivity issue, we propose a combined machine learning and distributionally robust optimization (DRO) approach which produces more robust and stable strategic investment decisions with respect to uncertainty assumptions. DRO is applied to deal with ambiguous probability distributions and Machine Learning is used to restrict the DRO model to a subset of important uncertain parameters ensuring computational tractability. Finally, we perform an out-of-sample simulation process to evaluate solutions performances. The Swiss energy system is used as a case study all along the paper to validate the approach.
|
Inzunza, A., Munoz, F. D., & Moreno, R. (2021). Measuring the effects of environmental policies on electricity markets risk. Energy Econ., 102, 105470.
Abstract: This paper studies how environmental policies, such as renewable portfolio standards (RPS) and carbon taxes, might contribute to reducing risk exposure in the electricity generation sector. We illustrate this effect by first computing long-term market equilibria of the Chilean generation sector for the year 2035 using a risk-averse planning model, considering uncertainty of hydrological scenarios and fossil fuel prices as well as distinct levels of risk aversion, but assuming no environmental policies in place. We then compare these risk-averse equilibria to generation portfolios obtained by imposing several levels of RPS and carbon taxes in a market with risk-neutral firms, separately. Our results show that the implementation of both policies can provide incentives for investments in portfolios of generation technologies that limit the risk exposure of the system, particularly when high levels of RPS (35%) or high carbon taxes (35 $/tonCO2) are applied. However, we find that in the case of a hydrothermal system, the resulting market equilibria under RPS policies yield expected generation cost and risk levels (i.e. standard deviation of costs) that are more similar to the efficient portfolios determined using a risk-averse planning model than the ones we find under the carbon tax.
|
