Search

To take unit commitment (UC) decisions under uncertain net load, most studies utilize a stochastic UC (SUC) model that adopts a one-size-fits-all representation of uncertainty. Disregarding contextual information such as weather forecasts and temporal information, these models are typically plagued by a poor out-of-sample performance. To effectively exploit contextual information, in this paper, we formulate a conditional SUC problem that is solved given a covariate observation. The presented problem relies on the true conditional distribution of net load and so cannot be solved in practice. To approximate its solution, we put forward a predictive prescription framework, which leverages a machine learning model to derive weights that are used in solving a reweighted sample average approximation problem. In contrast with existing predictive prescription frameworks, we manipulate the weights that the learning model delivers based on the specific dataset, present a method to select pertinent covariates, and tune the hyperparameters of the framework based on the out-of-sample cost of its policies. We conduct extensive numerical studies, which lay out the relative merits of the framework vis-\`a-vis various benchmarks., Comment: 5 pages, 2 figures

The deepening penetration of renewable resources into power systems entails great difficulties that have not been surmounted satisfactorily. An issue that merits special attention is the short-term planning of power systems under net load uncertainty. To this end, we work out a distributionally robust unit commitment methodology that expressly assesses the uncertainty associated with net load. The principal strength of the proposed methodology lies in its ability to represent the probabilistic nature of net load without having to set forth its probability distribution. This strength is brought about by the notion of ambiguity set, for the construction of which the Kullback-Leibler divergence is employed in this paper. We demonstrate the effectiveness of the proposed methodology on real-world data using representative studies. The sensitivity analyses performed provide quantitative answers to a broad array of what if questions on the influence of divergence tolerance and dataset size on optimal solutions., Keywords: data-driven optimization, distributionally robust optimization, uncertainty, unit commitment. arXiv admin note: text overlap with arXiv:2011.05314

The massive scale of the greenhouse gas (GHG) emissions due to the operation of cruise ships creates an acute need to develop cruise ship energy management systems (EMSs) that explicitly assess and mitigate GHG emissions. Renewable resources (RRs)—albeit their ubiquity in recent years—pose key challenges that need to be addressed so as to be efficiently utilized by cruise ship EMSs. To this end, in this paper, we propose a cruise ship EMS that optimizes the operation of controllable generators, battery storage system (BSS), controllable loads, and diesel purchase. The proposed EMS contemplates three objectives: minimization of total costs, mitigation of GHG emissions, and minimization of travel time in the case of an emergency. The proposed EMS harnesses long short-term memory networks (LSTMs) to forecast the generation of integrated PV panels and uncontrollable loads, and utilizes the forecasts to determine the optimal operations. The results illustrate the capabilities and effectiveness of the proposed EMS.

The surge in greenhouse gas emissions creates a dire need for effective energy management systems (EMSs) that enable the greater utilization of renewable resources to meet the power demand of an electric vehicle. In this study, we develop such an EMS for a solar electric vehicle (SEV). The proposed EMS optimizes the power flows and charging schedule of a SEV so as to minimize the total costs based on the PV generation, varying electricity prices, charging infrastructure, and trip characteristics. A key thrust of the proposed EMS is a deep learning model used to predict the uncertain photovoltaic generation of the vehicle. We illustrate the application of the proposed EMS on two case studies that are distinguished by their driving patterns and charging infrastructures. The results show that in both case studies the proposed EMS resulted in lower costs compared to a simple charging model.

This paper proposes a combined day-ahead forecasting and scheduling energy management system (EMS) with multiple complimentary objectives on household level. Our comparative case study, analyses, economical, ecological and grid independence aspects of our model include the utilization of renewable energy source (RES) with a controllable diesel generator and an energy storage system (ESS). The latter includes both stationary and electric vehicle batteries. The forecasting of RES and uncontrollable load of a single household is based on an LSTM approach, which is then used for day-ahead electricity scheduling. In order to be consistent with the forecasting horizon, day-ahead wholesale electricity market prices are used.

The global climate change creates a dire need to mitigate greenhouse gas (GHG) emissions from thermal generation resources (TGRs). While microgrids are instrumental in enabling the deeper penetration of renewable resources, the short-term planning of microgrids needs to explicitly assess the full range of impact of GHG emissions. To this end, we propose a novel unit commitment (UC) approach, which enables the representation of GHG emissions from TGRs, the specification of GHG emission constraints, and the ex-ante evaluation of carbon tax payment with all other costs and benefits. We quantify the relative merits of the proposed approach vis-\`a-vis the classical UC approach via representative studies. The results indicate that the proposed UC approach yields lower costs than does the classical UC approach and achieves a greater reduction in costs as carbon tax rate increases. Further, increasing carbon tax rates can markedly disincentivize TGR generation under the proposed approach., Comment: carbon tax, microgrids, power generation planning, unit commitment