EnergyTS TimeSeries Dataset with Price

We provide an evaluation benchmark for the following scenario: Day-ahead electricity price forecasting - Prediction of electricity market prices for the next day based on datetime information. This scenario contains multiple sub-datasets. The datasets are mainly generated based on a Structural Causal Model (SCM) technical framework. Centered on the operational logic of power systems, a directed acyclic graph (DAG) is used to characterize the causal relationships among load, renewable power generation, unit status, grid constraints, and locational marginal prices. Samples are then generated based on the electricity market clearing mechanism. The evaluation script corresponding to the current dataset can be accessed at https://github.com/antgroup/Energy-EVA for assessment.我们针对如下场景提供了一款评估基准测试集： 日前电价预测——基于时间信息预测次日电力市场电价。该场景包含多个子数据集。本系列数据集主要基于结构因果模型（Structural Causal Model, SCM）技术框架生成，以电力系统运行逻辑为核心，采用有向无环图（directed acyclic graph, DAG）刻画负荷、可再生能源发电、机组状态、电网约束以及节点边际电价之间的因果关联关系，随后基于电力市场出清机制生成样本数据。 当前数据集对应的评估脚本可通过https://github.com/antgroup/Energy-EVA获取，用于开展评估工作。