Abstract
NeuraEngineDx is a multi-sensor device to monitor condition of small-scale fishing vessel engines and is intended for IoT-based condition maintenance applications. The system integrates temperature sensor, vibration, sound and an infrared sensor, with an RTC DS3231 module for precise time-stamping. Data are collected in real time, logged to a MicroSD card, and can be transmitted to a cloud server via internet connectivity for remote analysis and condition monitoring. The enclosure is 3D-printed using Nylon Carbon (PA6-CF) to withstand high temperatures and harsh field conditions, while the circuit board is manufactured using FR4 material to ensure electrical stability. Performance validation against reference instruments demonstrated high metrological integrity, particularly for temperature and rotational speed, with R(2) values of 0.997 and 0.989, respectively. Quantitative metrics revealed a Mean Absolute Percentage Error (MAPE) of 1.11% for temperature and 0.36% for RPM, with a relative expanded uncertainty (Rel. U) below 1% for both parameters. While the sound and vibration sensors exhibited higher variances due to stochastic ambient noise, the system effectively identifies engine parameter deviations under simulated fault conditions. Limitations include susceptibility of the sound sensor to ambient noise, dependency on internet signal quality for cloud data transmission, and increased power consumption during extended modem activity. NeuraEngineDx provides an efficient, robust, and flexible platform for research and condition monitoring applications, particularly for small-scale diesel engines used by artisanal fisheries.