IEEE Transactions on Sustainable Energy · 2023 · 38 citations · 34 references
ComputingData ScienceForecastingMachine LearningRenewable Energy SystemsElectricity Load ManagementEnergy StorageEnergy System ModelingGenerative ModelingLoad ForecastingRenewable EnergyBattery Health AssessmentEnergy PlanningEnergy Storage ModelingGenerative Adversarial NetworksPredictionRenewable Energy ForecastingScenario GenerationScenario Reduction
The historical measured data of renewable energy sources and loads can be processed in various ways to generate scenarios for energy storage planning. With the development of advanced forecast technology, the valuable reference of massive forecast data accumulated by the prediction platform in scenario generation is ignored. To this end, we propose a new paradigm based on scenario generation for energy storage planning considering source-load uncertainties. First, a novel generative adversarial network (GAN) is constructed under weakly supervised learning. The network can extract the prediction errors between predicted data and measured data to generate source-load profiles. The loss function of the network is redesigned to adapt to the data generation task. To match the planning process, scenario generation and reduction algorithms are embedded in the GAN. Then, an energy storage sizing model considering battery health constraints is established. Case studies show that the proposed method can accurately portray the local fluctuation characteristics of source-load power and properly reduce conservatism compared with the model-driven method.
34
Data-driven chance constrained stochastic program
Ruiwei Jiang, Yongpei Guan · Mathematical Programming · 2015 · 502 citations
Pilar Meneses de Quevedo, Gregorio Muñoz-Delgado, Javier Contreras · IEEE Transactions on Smart Grid · 2017 · 234 citations