Dark Mode
Motor Thermal Characteristics Dataset
Retail & Consumer Behavior
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset contains sensor recordings from a permanent magnet synchronous motor (PMSM) undergoing test bench measurements, collected by the LEA department at Paderborn University. It includes 185 hours of data, expanded from an initial 138 hours, with all data now deanonymised. The motor is a prototype model from a German OEM. The data was gathered by exciting the motor with hand-designed driving cycles that simulate real-world conditions by denoting reference motor speed and torque. This dataset is particularly valuable for developing predictive models for rotor temperature and torque, as these are features that are often difficult and costly to measure reliably in commercial vehicles.
Columns
The dataset consists of 13 columns, sampled at a frequency of 2 Hz:
- u_q: Voltage q-component measurement in dq-coordinates (in Volts).
- coolant: Coolant temperature (in degrees Celsius).
- stator_winding: Stator winding temperature (in degrees Celsius), measured using thermocouples.
- u_d: Voltage d-component measurement in dq-coordinates (in Volts).
- stator_tooth: Stator tooth temperature (in degrees Celsius), measured using thermocouples.
- motor_speed: Motor speed (in revolutions per minute).
- i_d: Current d-component measurement in dq-coordinates.
- i_q: Current q-component measurement in dq-coordinates.
- pm: Permanent magnet temperature (in degrees Celsius), measured with thermocouples and transmitted wirelessly via a thermography unit.
- stator_yoke: Stator yoke temperature (in degrees Celsius), measured using thermocouples.
- ambient: Ambient temperature (in degrees Celsius).
- torque: Motor torque (in Newton-metres).
- profile_id: An integer identifier for each distinct measurement session.
Distribution
The dataset is provided as a CSV file,
measures_v2.csv, and has a file size of 300.06 MB. It comprises approximately 1.33 million records, with no mismatched or missing values in the detailed column statistics provided. The recordings are sampled at 2 Hz and span a total of 185 hours. Each measurement session, identifiable by its profile_id, can vary in duration from one to six hours.Usage
This dataset is ideal for various applications, including:
- Temperature prediction for electric motors, especially the challenging-to-measure rotor temperature.
- Developing deep learning models like Deep Residual Machine Learning for estimating motor temperatures.
- Enhancing motor control strategies by providing accurate torque estimations, which can lead to reduced power losses and less heat build-up.
- Research and development in the automotive industry for manufacturing more material-efficient motors and optimising motor utilisation.
- Studying the thermal behaviour of Permanent Magnet Synchronous Motors.
Coverage
The dataset covers sensor data from a permanent magnet synchronous motor prototype from a German OEM. All measurements were collected on a test bench at Paderborn University. The time range of the data spans 185 hours of recordings, with individual measurement sessions lasting between one and six hours. There is no specific demographic scope for this type of technical data. All data has been deanonymised as of an update on 26th April 2021.
License
CC BY-SA 4.0
Who Can Use It
This dataset is particularly useful for:
- Automotive industry professionals focused on electric motor design, manufacturing, and performance optimisation.
- Researchers and academics in fields such as electrical engineering, machine learning, and thermal management.
- Data scientists and engineers working on predictive maintenance, sensor data analysis, and control systems for electric motors.
Dataset Name Suggestions
- Electric Motor Temperature & Performance Data
- PMSM Test Bench Sensor Readings
- Automotive Permanent Magnet Synchronous Motor Data
- Motor Thermal Characteristics Dataset
- Electric Drive System Performance
Attributes
Original Data Source: Motor Thermal Characteristics Dataset
QUALITY