xpertsystems/enr003-sample

ENR003是一个合成电力需求与负荷预测数据集（样本预览版），由XpertSystems.ai的合成数据工厂（能源与气候垂直领域）创建。它是一个单表、负载研究校准的数据集，覆盖8个公用事业需求区域，跨越多样化的IECC气候区（从湿热到极冷），具有15分钟间隔分辨率。每一行数据整合了区域级负荷组成（如住宅、商业、工业、农业和电动汽车需求）、天气数据、日前/小时前/周前概率预测、峰值事件标志、ERCOT 4CP检测、需求响应激活、分时电价（TOU）层级、节点边际电价（LMP）以及表后分布式能源资源（DER，包括屋顶太阳能、电池储能系统、电动汽车和车网互动）。数据集校准基于行业标准，包括EPRI PRISM负载研究、PJM负荷预测存档、EIA-861、FERC电力报告和DOE EV充电研究2023。此样本预览包含8个区域×1周（2022-01-01至2022-01-08）×15分钟节奏的数据，共约5,376行和99列。完整产品则覆盖50个区域×完整年度周期（约175万行），包括多季节负荷因子、50年一遇峰值超限建模和N-1电网压力场景。数据集适用于负荷预测、概率预测、峰值需求预测、需求响应目标、电动汽车充电负荷分解、车网互动调度优化、表后DER聚合、气候区迁移学习、负荷持续时间曲线构建、价格弹性估计和节点边际电价预测等用例。ENR003 is a synthetic electricity demand and load forecasting dataset (sample preview) created by XpertSystems.ais Synthetic Data Factory in the Energy & Climate vertical. It is a single-table, load-research-calibrated dataset spanning 8 utility demand zones across diverse IECC climate zones (from Hot-Humid to Very Cold), with 15-minute interval resolution. Each row joins zone-level load composition (e.g., residential, commercial, industrial, agricultural, and EV demand), weather data, day-ahead/hour-ahead/week-ahead probabilistic forecasts, peak event flags, ERCOT 4CP detection, demand response activations, time-of-use (TOU) pricing tiers, locational marginal prices (LMPs), and behind-the-meter DER (including rooftop solar, battery energy storage systems, electric vehicles, and vehicle-to-grid). The dataset is calibrated against industry benchmarks such as EPRI PRISM load research, PJM Load Forecast Archive, EIA-861, FERC Electric Power Reports, and DOE EV Charging Study 2023. This sample preview covers 8 zones × 1 week (from 2022-01-01 to 2022-01-08) at a 15-minute cadence, with approximately 5,376 rows and 99 columns. The full product includes 50 zones × a full annual cycle (around 1.75 million rows) with multi-seasonal load factors, 1-in-50-year peak exceedance modeling, and N-1 grid stress scenarios. It is suitable for use cases like load forecasting, probabilistic forecasting, peak demand prediction, demand response targeting, EV charging load disaggregation, V2G dispatch optimization, behind-the-meter DER aggregation, climate zone transfer learning, load duration curve construction, price elasticity estimation, and LMP forecasting.