More Like this

1. Price-Aware Deep Learning for Electricity Markets

   Dvorkin, Vladimir; Fioretto, Ferdinando (February 2024, NeurIPS 2023 Workshop: Tackling Climate Change with Machine Learning)

   While deep learning gradually penetrates operational planning, its inherent prediction errors may significantly affect electricity prices. This letter examines how prediction errors propagate into electricity prices, revealing notable pricing errors and their spatial disparity in congested power systems. To improve fairness, we propose to embed electricity market-clearing optimization as a deep learning layer. Differentiating through this layer allows for balancing between prediction and pricing errors, as oppose to minimizing prediction errors alone. This layer implicitly optimizes fairness and controls the spatial distribution of price errors across the system. We showcase the price-aware deep learning in the nexus of wind power forecasting and short-term electricity market clearing. 

   more » « less
2. Electric Load Forecasting Using Multiple Output Gaussian Processes and Multiple Kernel Learning

   https://doi.org/10.1109/isiea58478.2023.10212291  

   Ghasempour, Alireza; Martínez-Ramón, Manel (July 2023, 2023 IEEE Symposium on Industrial Electronics & Applications (ISIEA))

   Electric load forecasting refers to forecasting the electricity demand at aggregated levels. Utilities use the predictions of this technique to keep a balance between electricity generation and consumption at each time and make accurate decision for power system planning, operations, and maintenance, etc. Based on prediction time horizon, electric load forecasting is classified to very short-term, short-term, medium-term, and long-term. In this paper, a multiple output Gaussian processes with multiple kernel learning is proposed to predict short-term electric load forecasting (predicting 24 load values for the next day) based on load, temperature, and dew point values of previous days. Mean absolute percentage error (MAPE) is used as a measure of prediction accuracy. By comparing MAPE values of the proposed method with the persistence method, it can been seen that the proposed method improves the persistence method MAPE up to 4%. 

   more » « less
3. Long-Term Multi-Resolution Probabilistic Load Forecasting Using Temporal Hierarchies

   https://doi.org/10.3390/en18112908  

   Bahman, Shafie; Zareipour, Hamidreza (June 2025, Energies)

   Accurate long-term electricity load forecasting is critical for energy planning, infrastructure development, and risk management, especially under increasing uncertainty from climate and economic shifts. This study proposes a multi-resolution probabilistic load forecasting framework that leverages temporal hierarchies to generate coherent forecasts at hourly, daily, monthly, and yearly levels. The model integrates climate and economic indicators and employs tailored forecasting techniques at each resolution, including XGBoost and ARIMAX. Initially incoherent forecasts across time scales are reconciled using advanced methods such as Ordinary Least Squares (OLS), Weighted Least Squares with Series Variance Scaling (WLS_V), and Structural Scaling (WLS_S) to ensure consistency. Using historical data from Alberta, Canada, the proposed approach improves the accuracy of deterministic forecasts and enhances the reliability of probabilistic forecasts, particularly when using the OLS reconciliation method. These results highlight the value of temporal hierarchy structures in producing high-resolution long-horizon load forecasts, providing actionable insights for utilities and policymakers involved in long-term energy planning and system optimization. 

   more » « less
4. Dynamic Model Predictive Shielding for Provably Safe Reinforcement Learning

   Banerjee, Arko; Rahmani, Kia; Biswas, Joydeep; Dillig, Isil (December 2024, Neurips 2024)

   Among approaches for provably safe reinforcement learning, Model Predictive Shielding (MPS) has proven effective at complex tasks in continuous, high-dimensional state spaces, by leveraging a backup policy to ensure safety when the learned policy attempts to take risky actions. However, while MPS can ensure safety both during and after training, it often hinders task progress due to the conservative and task-oblivious nature of backup policies. This paper introduces Dynamic Model Predictive Shielding (DMPS), which optimizes reinforcement learning objectives while maintaining provable safety. DMPS employs a local planner to dynamically select safe recovery actions that maximize both short-term progress as well as long-term rewards. Crucially, the planner and the neural policy play a synergistic role in DMPS. When planning recovery actions for ensuring safety, the planner utilizes the neural policy to estimate long-term rewards, allowing it to observe beyond its short-term planning horizon. Conversely, the neural policy under training learns from the recovery plans proposed by the planner, converging to policies that are both high-performing and safe in practice. This approach guarantees safety during and after training, with bounded recovery regret that decreases exponentially with planning horizon depth. Experimental results demonstrate that DMPS converges to policies that rarely require shield interventions after training and achieve higher rewards compared to several state-of-the-art baselines 

   more » « less
5. Integrated Hurricane Relief Logistics and Evacuation Planning under Forecast Uncertainty: A Case Study for Hurricane Florence

   Bhattarai, Sudhan; Song, Yongjia (July 2023, Proceedings of the IISE Annual Conference & Expo 2023)

   Babski-Reeves, K.; Eksioglu, B.; Hampton, D. (Ed.)

   In this paper, we study an integrated hurricane relief logistics and evacuation planning (HRLEP) problem. We propose stochastic optimization models and methods that integrate the hurricane relief item pre-positioning problem and the hurricane evacuation planning problem, which are often treated as separate problems in the literature, by incorporating the forecast information as well as the forecast errors (FE). Specifically, we fit historical FE data into an auto-regressive model of order one (AR-1), from which we generate FE realizations to create evacuation demand scenarios. We compare a static decision policy based on the proposed stochastic optimization model with a dynamic policy obtained by applying this model in a rolling-horizon (RH) procedure. We conduct a preliminary numerical experiment based on real-world data to validate the value of stochastic optimization and the value of the dynamic policy based on the RH procedure. 

   more » « less