{
  "fetched_at": "2026-10-20T04:35:31Z",
  "source_url": "https://www.kaggle.com/competitions/electricity-load-forecasting-challenge-test/data",
  "pages": [
    {
      "name": "data-description",
      "content": "The datasets used in this thesis consist of real-world energy meter readings gathered in conjunction with the nationwide project Flexible Energy Denmark (FED). FED is a datadriven research project that aims to develop digital solutions to accelerate the green transition by increasing flexibility in the energy sector (Flexible-Energy-Denmark, 2019). The danish distribution system operator TREFOR has two test areas of their grid designated as living laboratories for the project, located in Strib, Middelfart, and Nørre Bjert, Kolding. Electricity consumption readings are collected on an hourly basis for most installations within the two areas. However, some meters collect data at a quarter-hourly interval.\n\n## Files\n\n*   **train.csv** - the training set\n*   **example_submission.csv** - a sample submission file in the correct format\n\n## Columns\n\n-  `datetime` - The time when the load was measured, servers as an ID\n-   `value` - the value of the aggregated electricity load at the given timestamp",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "An increasing number of distributed energy resources (DERs) introduces challenges to the energy system, e.g., grid balancing. This challenge is especially crucial to distribution system operators (DSOs) who maintain control of the energy grid. Compared to the large-scale battery, demand response is a better solution. However, demand response might create enormous activities in the grid due to the communication between the electricity markets and the demand sides, which potentially affects the distribution grid stability. Therefore, accurate predictions of the future loads in the distribution grid are necessary. Globally, households make up a large part of electricity. However, most distribution grid designs cannot afford a large number of DERs, especially synchronous charging. Thus, it is critical for DSOs to accurately forecast energy load in the residential areas to improve their operation quality and reduce maintenance costs. \n\nThis competition is hosted as a Test for future competitions.\n###Problem formulation\nIn this competition, you will be developing machine learning models, that can accurately forecast electricity load for Danish residential areas. \n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The metric used in this competition is the **Mean Absolute Percentage Error**.\nThe mean absolute percentage error is often used for forecasting, and it describes how\naccurate a forecast model is as a percentage.\n\n\\\\[\n     \\frac{1}{n} \\sum\\_{i=1} \\Bigg{|}\n     \\frac{Y\\_i - \n          \\widehat{Y}\n           \\_i}\n          {Y\\_i} \\Bigg{|}\n\\\\]\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}