🤖 AI Summary
This paper addresses the cost–reliability co-optimization challenge in renewable energy scheduling under dual uncertainties—stochastic wind/solar generation and load demand. To this end, it proposes a prediction–optimization joint modeling paradigm. Leveraging Monash microgrid operational data, meteorological records, and electricity market prices, the authors develop a joint time-series forecasting model (using gradient boosting trees or random forests) for wind/solar output and load. They further design a multi-scenario mixed-integer linear/quadric programming (MILP/MIQP) dispatch framework integrating scenario generation and sample average approximation (SAA), jointly optimizing generator commitment and battery charge/discharge scheduling. The work establishes, for the first time, a standardized Predict+Optimize benchmark task to advance end-to-end co-design research. Evaluated on real-world microgrid data, the proposed SAA-based joint optimization method achieved first place among seven top-tier teams, significantly reducing energy procurement costs.
📝 Abstract
Algorithms that involve both forecasting and optimization are at the core of solutions to many difficult real-world problems, such as in supply chains (inventory optimization), traffic, and in the transition towards carbon-free energy generation in battery/load/production scheduling in sustainable energy systems. Typically, in these scenarios we want to solve an optimization problem that depends on unknown future values, which therefore need to be forecast. As both forecasting and optimization are difficult problems in their own right, relatively few research has been done in this area. This paper presents the findings of the ``IEEE-CIS Technical Challenge on Predict+Optimize for Renewable Energy Scheduling,"held in 2021. We present a comparison and evaluation of the seven highest-ranked solutions in the competition, to provide researchers with a benchmark problem and to establish the state of the art for this benchmark, with the aim to foster and facilitate research in this area. The competition used data from the Monash Microgrid, as well as weather data and energy market data. It then focused on two main challenges: forecasting renewable energy production and demand, and obtaining an optimal schedule for the activities (lectures) and on-site batteries that lead to the lowest cost of energy. The most accurate forecasts were obtained by gradient-boosted tree and random forest models, and optimization was mostly performed using mixed integer linear and quadratic programming. The winning method predicted different scenarios and optimized over all scenarios jointly using a sample average approximation method.
