Dark Mode
Liquid Phase Binary Classification Data
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Digital signals derived from a capacitive sensor array are captured to enable the binary classification of liquids, specifically distinguishing between water and oil phases. The data represents intensity values obtained from sensor electrodes distributed vertically along a Printed Circuit Board (PCB). These electrodes are excited sequentially, measuring the digital voltage which is then converted into intensity values based on a base voltage recorded in air. The acquisition process involves a linear actuator moving the sensor vertically at a constant speed, pausing at specific positions to record the electrode's immersion in oil (upper position) or water (bottom position). This collection supports the development of machine learning models aimed at inferring the material type surrounding an electrode based on short, fixed-size signal samples.
Columns
- col1 - col10: Consecutive intensity values representing the digital signal captured from the capacitive sensor electrode. These values are calculated using the magnitude of the difference between the measured voltage and the base voltage (air), normalised by the base voltage.
- target: The classification label indicating the medium in which the electrode was immersed. A value of +1 denotes water, while -1 denotes oil.
Distribution
The data is structured in a CSV format with a file size of approximately 259.24 kB. It contains 11 columns in total (10 feature columns and 1 target label). Metadata indicates a volume of 4,475 total values per column, suggesting the dataset comprises approximately 4,475 records. Each record represents a signal composed of 10 consecutive values captured over a 1-second interval (100 msec sampling rate).
Usage
- Binary Classification Training: developing algorithms to automatically differentiate between water and oil based on capacitive readings.
- Sensor Signal Analysis: investigating the behaviour of capacitive electrodes in different dielectric phases.
- Automated Quality Control: applying material detection logic in industrial liquid management systems.
- Robotic Sensing: enhancing feedback loops for actuators interacting with liquid environments.
Coverage
- Environment: The data was collected under controlled indoor laboratory conditions.
- Temperature: Experiments were conducted at a room temperature of approximately 23 degrees Celsius.
- Phases: The scope is limited to three phases: Air (for calibration), Oil (labeled -1), and Water (labeled +1).
- Equipment: Data was generated using a capacitive sensor array moved by a linear actuator at a fixed speed of 20 mm/second.
License
Attribution 4.0 International (CC BY 4.0)
Who Can Use It
- Data Scientists working on signal processing and classification problems.
- Electrical Engineers analysing capacitive sensor performance.
- Machine Learning Researchers looking for simple binary datasets for benchmarking algorithms.
- Robotics Engineers developing liquid detection systems.
Dataset Name Suggestions
- Capacitive Sensor Water vs Oil Signals
- Liquid Phase Binary Classification Data
- Sensor Electrode Intensity Values
- Water and Oil Capacitive Readings
Attributes
Original Data Source: Liquid Phase Binary Classification Data
QUALITYLoading...