Este archivo ha sido creado el 10-11-2024 por Juan Manuel Montes Sánchez GENERAL INFORMATION ------------------ 1. Dataset title: Peristaltic pump aging detection dataset 2. Authorship: Name: Juan Manuel Montes Sánchez Institution: Universidad de Sevilla Email: juanmanuelmontes@us.es ORCID: 0000-0002-0983-2386 Name: Yoko Uwate Institution: Tokushima University Email: uwate@ee.tokushima-u.ac.jp ORCID: 0000-0002-2992-8852 Name: Yoshifumi Nishio Institution: Tokushima University Email: nishio@ee.tokushima-u.ac.jp ORCID: 0000-0002-0247-0001 Name: Saturnino Vicente Díaz Institution: Universidad de Sevilla Email: satur@us.es ORCID: 0000-0001-9466-485X Name: Ángel Francisco Jiménez Fernández Institution: Universidad de Sevilla Email: angel@us.es ORCID: 0000-0003-3061-5922 DESCRIPTION ---------- 1. Dataset language: English 2. Abstract: This dataset contains samples coming from a hydraulic block from a biomedical equipment. The block mounts 3 Thomas SR10/30 DC standard perisltaltic pumps, which were filled with distilled water. Only one pump was running at the same time during these recordings. The cassettes of the pumps were changed before each recording. We used cassettes with 2 different levels of degradation: NEW (unused) and OLD (lifetime already expired). We defined 3 different classes: Class 1 is STOP (no pump running), class 2 is NEW (one pump running with a new cassette), and class 3 is OLD (one pump running with an old cassette). The classified samples were recorded using several sensors: 3 accelerometers, 1 gyroscope, 1 magnetometer and 1 microphone. All data were recorded at the same time at the maximum available frequency using the device "ST SensorTile.box". The raw data were then processed into different .csv files (.wav files for audio), which are the ones present in this dataset. Este dataset contiene muestras de un bloque hidráulico de un equipo biomédico. El bloque monta 3 bombas peristálticas estándar de corriente continua modelo Thomas SR10/30, que fueron llenadas con agua destilada. Solo una bomba estuvo funcionando a la vez durante la recogida de datos. Los cassettes de las bombas fueron cambiados antes de cada recogida de datos. Usamos cassettes con dos niveles distintos de degradación: NEW (sin previo uso) y OLD (vida útil agotada). Definimos 3 clases diferentes: la clase 1 es STOP (ninguna bomba funcionando), la clase 2 es NEW (una bomba funcionando con un cassette nuevo) y la clase 3 es OLD (una bomba funcionando con un cassete viejo). Las muestras fueron recogidas usando varios sensores: 3 acelerómetros, un giróscopo, un magnetómetro y un micrófono. Todos los datos fueron recogidos al mismo tiempo a la máxima frecuencia de muestreo disponible usando el dispositivo "ST SensorTile.box". Los datos en bruto fueron procesados en diferentes archivos .csv (archivos .wav para el audio), que están presentes en este dataset. 3. Keywords: Accelerometer, Acelerómetro, Gyroscope, Giróscopo, Magnetometer, Magnetómetro, Microphone, Micrófono, Audio, IMU, IIoT, peristaltic pump, bomba peristáltica, predictive maintenance, mantenimiento predictivo, Edge AI 4. Date of data collection: 28-06-2023 5. Publication Date: 25-09-2024 6. Grant information: Grant Agency: Agencia Estatal de Investigación - Ministerio de Ciencia, Innovación y Universidades (Gobierno de España) Grant Number: Proyecto PREDICAR (ref. PID2023-149777OB-I00) 7. Geographical location/s of data collection: Tokushima University, Tokushima, JAPAN ACCESS INFORMATION ------------------------ 1. Creative Commons License of the dataset: CC BY-NC 4.0 2. Dataset DOI: 10.12795/11441/162880 3. Related publication: Title: Predictive Maintenance Edge Artificial Intelligence Application Study Using Recurrent Neural Networks for Early Aging Detection in Peristaltic Pumps DOI: 10.1109/TR.2024.3488963 4. Link to related datasets: N/A DOI/URL: VERSIONING AND PROVENANCE --------------- 1. Last modification date: 05-05-2024 for data, 10-11-2024 for this README 2. Were data derived from another source?: No 3. Additional related data not included in this dataset: N/A METHODOLOGICAL INFORMATION ----------------------- 1. Description of the methods used to collect and generate the data: This dataset contains samples coming from a hydraulic block from a biomedical equipment. The block mounts 3 Thomas SR10/30 DC standard perisltaltic pumps, which were filled with distilled water. Only one pump was running at the same time during these recordings, always at maximum constant speed. The cassettes of the pumps were changed before each recording. We used cassettes with 2 different levels of degradation: NEW (unused) and OLD (lifetime already expired). We defined 3 different classes: Class 1 is STOP (no pump running), class 2 is NEW (one pump running with a new cassette), and class 3 is OLD (one pump running with an old cassette). The classified samples were recorded using several sensors: 3 accelerometers, 1 gyroscope, 1 magnetometer and 1 microphone. All data were recorded at the same time at the maximum available frequency using the device "ST SensorTile.box". 2. Data processing methods: The raw data has already been processed into sepparate different .csv files (.wav files for audio) using python code. 3. Software or instruments needed to interpret the data: any text reader for .csv format, any audio player for .wav format. 4. Information about instruments, calibration and standards: See vendor information for recording device "ST SensorTile.box" 5. Environmental or experimental conditions: Closed room. No external equipment running besides lighting, power supply and PC. No external audible noises like human voices. 6. Quality-assurance procedures performed on the data: Empty or non valid values removed, if any. FILE OVERVIEW ---------------------- 1. Explain the file naming conversion: Each sensor has its own folder. The sensor folder name starts with ACC for accelerometer data, GYRO for gyroscope data, MAG for magnetometer data or MIC for microphone data. After that, the name of the sensor. At the end, sampling frequency in Hz. For example: ACC_LISDW12_1600Hz folder contains data for the accelerometer named LISDW12 at a sampling rate of 1600Hz. Each sensor folder contains 17 .csv files except for the microphone folder, which contains 17 .wav files. Each file is a unique sample tagged with one class (1, 2 or 3). At the end of each filename this class is also included. For example, 0008_03.csv is the sample number 8, which corresponds to class 3 tagged data. 2. File list: File name: 0000_02.csv / 0000_02.wav Description: Sample number 0000, corresponding class is 2 (NEW). Recommended for the holdout dataset. File name: 0001_01.csv / 0001_01.wav Description: Sample number 0001, corresponding class is 1 (STOP). Recommended for the holdout dataset. File name: 0002_02.csv / 0002_02.wav Description: Sample number 0002, corresponding class is 2 (NEW). Recommended for the holdout dataset. File name: 0003_01.csv / 0003_01.wav Description: Sample number 0003, corresponding class is 1 (STOP). File name: 0004_02.csv / 0004_02.wav Description: Sample number 0004, corresponding class is 2 (NEW). File name: 0005_01.csv / 0005_01.wav Description: Sample number 0005, corresponding class is 1 (STOP). File name: 0006_02.csv / 0006_02.wav Description: Sample number 0006, corresponding class is 2 (NEW). File name: 0007_01.csv / 0007_01.wav Description: Sample number 0007, corresponding class is 1 (STOP). File name: 0008_03.csv / 0008_03.wav Description: Sample number 0008, corresponding class is 3 (OLD). Recommended for the holdout dataset. File name: 0009_01.csv / 0009_01.wav Description: Sample number 0009, corresponding class is 1 (STOP). Recommended for the holdout dataset. File name: 0010_03.csv / 0010_03.wav Description: Sample number 0010, corresponding class is 3 (OLD). Recommended for the holdout dataset. File name: 0011_03.csv / 0011_03.wav Description: Sample number 0011, corresponding class is 3 (OLD). File name: 0012_01.csv / 0012_01.wav Description: Sample number 0012, corresponding class is 1 (STOP). File name: 0013_03.csv / 0013_03.wav Description: Sample number 0013, corresponding class is 3 (OLD). File name: 0014_03.csv / 0014_03.wav Description: Sample number 0014, corresponding class is 3 (OLD). File name: 0015_01.csv / 0015_01.wav Description: Sample number 0015, corresponding class is 1 (STOP). File name: 0016_03.csv / 0016_03.wav Description: Sample number 0016, corresponding class is 3 (OLD). File name: 0017_01.csv / 0017_01.wav Description: Sample number 0017, corresponding class is 1 (STOP). 3. Relationship between files: all samples with the same sample number were recorded at the same time, but come from different sensors. Each sensor folder contains the same amount of .csv or .wav files (17). 4. File format: .csv format for all sensor data except microphone data, which is in .wav format. 5. If the dataset includes multiple files, specify the directory structure and relationships between the files: Each folder contains data from a different sensor. There are 3 accelerometers, 1 gyroscope, 1 magnetometer and 1 microphone. Each folder is named as described in 1. The structure is as follows: -DATA | |-ACC_LIS2DW12_1600Hz | | | |- files from 0000_02.csv to 0017_01.csv | |-ACC_LIS3DHH_1100Hz | | | |- files from 0000_02.csv to 0017_01.csv | |-ACC_LSM6DSOX_6667Hz | | | |- files from 0000_02.csv to 0017_01.csv | |-GYRO_LSM6DSOX_6667Hz | | | |- files from 0000_02.csv to 0017_01.csv | |-MAG_LIS2MDL_100Hz | | | |- files from 0000_02.csv to 0017_01.csv | |-MIC_MP23ABS1_192000Hz | |- files from 0000_02.wav to 0017_01.wav SPECIFIC INFORMATION FOR TABULAR DATA ------------------------------------------- 1. Name file: all .csv files 2. Number of rows and columns: 6 columns, variable number of rows 3. Variables list: Variable name: index Description: index column, non relevant Units of measure or value labels | Unidades de medida o etiquetas de valor: units Variable name: Time Description: time mark for the sample Units of measure or value labels | Unidades de medida o etiquetas de valor: seconds Variable name: A_x[g], A_y[g] and A_z[g]. Description: Only pressent in accelerometer data. G force in each axxis (x, y, z). Units of measure or value labels | Unidades de medida o etiquetas de valor: standard gravity units. Variable name: G_x[mdps], G_y[mdps] and G_z[mdps]. Description: Only pressent in gyroscope data. Degrees per second in each axxis (x, y, z). Units of measure or value labels | Unidades de medida o etiquetas de valor: millidegrees per second. Variable name: M_x[gauss], M_y[gauss] and M_z[gauss]. Description: Only pressent in magnetometer data. Magnetic field measurements in each axxis (x, y, z). Units of measure or value labels | Unidades de medida o etiquetas de valor: Gauss. Variable name: Class. Description: Corresponding class for that sample. 1 is STOP, 2 is NEW and 3 is OLD. Units of measure or value labels | Unidades de medida o etiquetas de valor: 1, 2 or 3. Each number indicates a different class. 4. Codes or symbols for missing data: There are no missing data. All empty or non valid values have been removed, if any. Code or symbol: Definition: 5. Special formats or abbreviations used: See variable names above. MORE INFORMATION -------------- The following sample numbers are recommended for the holdout dataset when performing holdout validation: 0000, 0001, 0002, 0008, 0009 and 0010. This data has been taken using unique cassetes that are not present in the rest of the samples.