Electricity Consumption Data

该数据集包含了从澳大利亚五个州收集的每30分钟一次的电力消耗数据，用于预测电力消耗。数据通过最小最大归一化方法被标准化至[0, 1]的范围。每个任务包含395个样本，其中316个用于训练，79个用于测试。规模上分为两组任务：一步提前预测（5个任务）和多步提前预测（20个任务）。任务的目的是进行电力消耗预测。

This dataset contains 30-minute interval electricity consumption data collected from five Australian states, designed for electricity consumption prediction tasks. The raw data has been standardized to the range [0, 1] using the Min-Max normalization method. Each task comprises 395 samples, among which 316 are utilized for model training and 79 for model testing. The tasks are divided into two groups by prediction scale: one-step-ahead prediction (5 tasks) and multi-step-ahead prediction (20 tasks). The core objective of all tasks is electricity consumption forecasting.

搜集汇总

背景与挑战

背景概述

该数据集包含2025年3月至2026年4月期间的月度电力消费历史数据，以压缩文件形式提供，每个文件对应一个月的电力需求记录。