Effective Vibration-Based Anomaly Detection in Water Pump Operation Using Arduino Microcontroller

Abstract

Arduino microcontroller and ADXL accelerometer are commonly paired devices, often considered for creating inexpensive vibration analysers. Many researchers have proven that both pairing devices have good performance in vibration measurement and have the potential for commercialisation. This study evaluates the feasibility of vibration measurement and monitoring using an Arduino microcontroller with an inexpensive accelerometer in detecting anomalies during water pump operation. A dedicated Arduino Mega and an ADXL345 accelerometer were attached to a water pump motor to facilitate continuous monitoring of vibrations. The vibration measurement was set at a sampling rate of 530 Hz. Vibration data in RMS value was sent to the cloud storage for monitoring. Raw data captured during normal and abnormal conditions were collected at the site when anomalies were detected for further analysis. The results showed that the abnormal conditions could be clearly differentiated from normal conditions using the Fast Fourier Transform method and spectrogram analysis. In summary, this study confirms that integrating the Arduino Microcontroller with the ADXL accelerometer effectively detects irregularities in the operating conditions of the water pump.