RT journal article
T1 Real-time detection of uncalibrated sensors using neural networks
A1 Muñoz Molina, Luis J.
A1 Cazorla Piñar, Ignacio
A1 Domínguez Morales, Juan P.
A1 Lafuente Molinero, Luis
A1 Pérez Peña, Fernando
A2 Ingeniería en AutomáticaElectrónica, Arquitectura y Redes de Computadores
A2 Matemáticas
K1 Neural networks
K1 Sensors
K1 Uncalibrations
K1 Sensor anomalies
K1 Transfer learning
AB Nowadays, sensors play a major role in several fields, such as science, industry and everyday technology. Therefore, theinformation received from the sensors must be reliable. If the sensors present any anomalies, serious problems can arise,such as publishing wrong theories in scientific papers, or causing production delays in industry. One of the most commonanomalies are uncalibrations. An uncalibration occurs when the sensor is not adjusted or standardized by calibrationaccording to a ground truth value. In this work, an online machine-learning based uncalibration detector for temperature,humidity and pressure sensors is presented. This development integrates an artificial neural network as the main componentwhich learns from the behavior of the sensors under calibrated conditions. Then, after being trained and deployed, it detectsuncalibrations once they take place. The obtained results show that the proposed system is able to detect the 100% of thepresented uncalibration events, although the time response in the detection depends on the resolution of the model for thespecific location, i.e., the minimum statistically significant variation in the sensor behavior that the system is able to detect.This architecture can be adapted to different contexts by applying transfer learning, such as adding new sensors or havingdifferent environments by re-training the model with minimum amount of data.
PB Springer
SN 0941-0643
YR 2022
FD 2022-01
LK http://hdl.handle.net/10498/26123
UL http://hdl.handle.net/10498/26123
LA eng
NO Open Access funding provided thanks to the CRUE-CSICagreement with Springer Nature. This work was partially supportedby the Altran Innovation Center for Advance Manufacturing, the EUH2020 project Virtual IoT Maintenance System (VIMS Grant ID:878757), the Spanish grant (with support from the European RegionalDevelopment Fund) MIND-ROB (PID2019-105556GB-C33) and bythe EU H2020 project CHIST-ERA SMALL (PCI2019-111841-2).
DS Repositorio Institucional de la Universidad de Cádiz
RD 09-may-2026