{
  "fetched_at": "2026-10-19T07:24:33Z",
  "source_url": "https://www.kaggle.com/competitions/classification-of-defects-in-a-rotary-machine/data",
  "pages": [
    {
      "name": "data-description",
      "content": "Here you can download the dataset files of the experiment to detect defects in fluid friction bearing units. There are two types of files. Files named \"data...\" are for training networks. The first digit in the file name indicates the experiment number, and the second one indicates the defect class. In total, there were 4 types of the state of the experimental stand. Correspondence of the defect number and its type: 1 - no defects, 2 - misalignment in the coupling, 3 - imbalance, 4 - bushing with defects.\nThe second type of files called \"Series...\" are intended for testing the algorithms used. The final fault labeling file should be prepared based on the data of these experiments. The true type of defect is hidden, do not use the data from these files when training.\n\nInformation on the uploaded participant file is located in the Overview => Evaluetion.\nDue to the small amount of data, a test sample is available to participants. Therefore, we hope for your honesty.\n\nThe experiment time was 10 min. The frequency of obtaining data on the vibrations of the rotor is 1000 sig/sec. The frequency of temperature data acquisition is 1 sig/sec.\nData Description:\nEach column corresponds to a certain parameter, the rows contain the readings from the sensors in each time period. Temperature sensor data is duplicated for every second to simplify data correlation.\n1 column of the array - horizontal oscillations in the left bearing, in Volts,\n2 column of the array - vertical oscillations in the left bearing, in Volts,\n3 column of the array - horizontal oscillations in the right bearing, in Volts,\n4 column of the array - vertical oscillations in the right bearing bearing, in Volts,\n5 column of the array - horizontal oscillations in the left bearing, taking into account smoothing, in Volts,\n6 column of the array - vertical oscillations in the left bearing, taking into account smoothing, in Volts,\n7 column of the array - horizontal vibrations in the right bearing, taking into account smoothing, in Volts,\n8 column of the array - vertical oscillations in the right bearing bearing, taking into account smoothing, in Volts,\n9 column of the array - temperature in the left bearing, in Ohms,\n10 column of the array - temperature in the right bearing, in Ohms.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Hello Dear colleagues,\n\nWe are launching a competition for the classification of malfunctions of rotary machines. We want to invite you to participate in it with the possibility of further publication of your results. We decided to use this competition to identify advanced classification algorithms that can effectively detect faults. If you have advanced ideas and enthusiasm, then you should not think any more. \n\nGood luck!!\n\n## Acknowledgements\nThe work has been carried out at the Oryol State University named after I.S. Turgenev” with the financial support of the Ministry of Science and Higher Education of the Russian Federation within the project “Creation of a digital system for monitoring, diagnosing and predicting the state of technical equipment using artificial intelligence technology based on domestic hardware and software”, Agreement No. 075-11-2021-043 from 25.06.2021. \n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "In order to check the accuracy of your model, you need to create a CSV file. The two cells in the first row should contain the column names. Names should be for the first - id, for the second - class. Then, in the cells of the first column, the number of the experiment (see \"Data\" files \"Series...\") is indicated from 1 to 12. The second in the second column is the number indicating the class of the state of the experimental setup. The accuracy of your model will be measured by the absolute deviation.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "Our main rules are:\n\n - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}