ETT (Electric Transformer Temperature) Time Series Dataset

#Time Series Prediction #Anomaly Detection #Trend Analysis #Power Monitoring #Equipment Maintenance #Fault Warning #Industrial Automation

500
1.3G
CSV
CC-BY-NC-SA 4.0
MOBIUSI INC

Updated:2026-03-15

AI Analysis & Value Prop

In the industrial manufacturing sector, failures in power equipment can lead to significant economic losses and safety issues. Currently, the industry widely applies simplified threshold warning methods, which have low sensitivity to data anomalies and cannot carry out effective prevention in advance. The ETT time series dataset aims to improve the accuracy of electric transformer temperature prediction, catering to the needs of preventive maintenance. The data is collected through high-precision sensors in real industrial environments, simulating complex power load changes. During the data collection process, a strategy combining automated calibration and manual review is adopted, with multiple rounds of annotation and expert team review to ensure data consistency and accuracy. The annotation team consists of 15 people, including power engineers and data scientists. Data preprocessing includes steps such as time series completion, data cleaning, and normalization, and is finally stored in a structured CSV format to facilitate training data models.

Technical Specifications

Field	Type	Description
file_name	string	File name
temperature_min	float	The minimum temperature of the transformer recorded in the file.
temperature_max	float	The maximum temperature of the transformer recorded in the file.
OT	float	The temperature of the transformer's insulating oil.
LULL	float	A supplementary or derived feature paired with LUFL.
LUFL	float	Represents a low-value or low-frequency load on the transformer's low-voltage side, or a related low-usefulness power feature.
MULL	float	A supplementary or derived feature paired with MUFL.
MUFL	float	Represents a medium-value or medium-frequency load on the transformer's medium-voltage side, or a related medium-usefulness power feature.
HULL	float	A supplementary or derived feature often paired with HUFL. The exact interpretation (e.g., reactive power, a different measurement angle) depends on the original dataset documentation.
HUFL	float	Represents a high-value or high-frequency load on the transformer's high-voltage side, or a related high-usefulness power feature.
date	string	The exact date and time of the record.

Compliance Statement

Authorization Type	CC-BY-NC-SA 4.0 (Attribution–NonCommercial–ShareAlike)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

What is the main application of the ETT time series dataset?: The ETT time series dataset is primarily used for predicting and analyzing the temperature of electric transformers to optimize transformer performance and maintenance in industrial manufacturing.

What are the industrial applications of the ETT dataset?: The ETT dataset is mainly applied in the industrial manufacturing sector, particularly in power and energy management.

How does the ETT time series dataset enhance transformer performance in industrial manufacturing?: The ETT dataset helps enhance transformer performance by providing accurate temperature prediction data, preventing overheating issues and improving the efficiency and safety of industrial equipment.

What are the advantages of the ETT dataset in temperature prediction?: The ETT dataset is known for its high quality and accuracy, providing a reliable data foundation for precise temperature predictions of electric transformers.

Why choose the ETT time series dataset for research?: The ETT time series dataset is chosen for research because it provides crucial data support for transformer temperature management and fault prevention in the power industry.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={ETT (Electric Transformer Temperature) Time Series Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/a7664eb013237340a3b12b9fc0ea6be2?dataset_scene_cate_type=5},
  urldate={2026-02-04},
  keywords={Transformer Temperature Prediction, Industrial Time Series Dataset, Power Equipment Monitoring},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Products

Scene

Domain

Modality

Task