Research and Development of Hardware-Based Noise Reduction Strategies for Predictive Maintenance

Seah, Jia Qi (2026) Research and Development of Hardware-Based Noise Reduction Strategies for Predictive Maintenance. Final Year Project (Diploma), Tunku Abdul Rahman University of Management and Technology.

Text Seah Jia Qi_Thesis.pdf Restricted to Registered users only Download (3MB)

Abstract

Machine condition monitoring plays a crucial role in predictive maintenance for industrial systems. However, the accuracy of acoustic-based monitoring is significantly affected by environmental and machinery noise in factory settings, which can obscure early fault indicators. This project aims to design and implement a hardware-based real-time adaptive noise cancellation system to enhance the clarity of machine condition signals for predictive maintenance applications. The proposed system utilises the Least Mean Square (LMS) adaptive filtering algorithm, chosen for its computational simplicity, hardware compatibility, and ability to cancel stationary and quasi-stationary industrial noise such as fan hums and conveyor sounds. The algorithm is implemented on an embedded platform to ensure low latency and real-time performance. The proposed system successfully developed an effective noise reduction mechanism that enhances the quality of acoustic signals for condition monitoring. Experimental results demonstrate an improvement in signal-to-noise ratio (SNR) and fault detection accuracy under different operating conditions, contributing to reduced unexpected downtime and maintenance costs in industrial environments