Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Electrical and Computer Engineering

Supervisor

Shami, Abdallah

Abstract

Vibration-Based Condition Monitoring (VBCM) is commonly utilized in Prognostics and Health Management (PHM) due to its non-destructive nature and inherent advantages over alternative forms of condition monitoring. Furthermore, the rapid evolution of sensor fabrication and the rise of the Internet of Things (IoT) have facilitated large-scale VBCM systems across diverse domains, including industry, transportation, healthcare, agriculture, and wildlife monitoring. The recent advancements in computing technologies have significantly expanded the potential for VBCM by leveraging the synergy between signal processing and Machine Learning (ML). Accordingly, data-driven VBCM has emerged as a paradigm shift, improving the performance and reliability of VBCM systems. To this end, addressing various attributes of data-driven VBCM becomes increasingly important since it represents the core of current and future VBCM systems. The work presented in this thesis addresses the main aspects of VBCM, including signal processing fundamentals, feature extraction, availability of labeled data, computational complexity, and power efficiency. The methods employed in this thesis span the fields of Digital Signal Processing (DSP) and ML techniques (supervised, Deep Learning (DL)), including signal preprocessing, signal denoising, signal frequency-domain analysis, signal time-frequency domain analysis, feature extraction, signal companding (compression-expansion), and 1-dimensional (1D) convolutional reconstruction autoencoders. These methods address extraction of effective condition-related features, limited availability of labeled data, noise removal, complexity considerations in VBCM systems, and power efficiency of power-constrained sensor nodes in remote VBCM. By addressing the aforementioned problems, the end-to-end performance of VBCM systems can be improved in terms of the size of training data, the reliability of the monitoring process, system delay, memory requirements, and power consumption. To ensure the explainability of the extracted features, the developed methods for the extraction of condition-related features are based on signal processing since feature engineering using signal processing creates explainable features that link meaningfully to signal conditions or classes compared to DL-based features. The thesis also contributes to the VBCM literature by providing a comprehensive tutorial on signal processing fundamentals, an overview of a typical signal-based ML pipeline, and an application-independent review of feature extraction techniques. The work presented in this thesis presents efficient solutions to the main challenges that face the practical deployment of real-world VBCM systems

Summary for Lay Audience

The ongoing technological transformation of data-driven Vibration-Based Condition Monitoring (VBCM) has enormous potential for consumers and businesses. This transformation is centered around integrating Digital Signal Processing (DSP) with Machine Learning (ML) models to facilitate reliable and efficient automated VBCM applications. VBCM utilizes vibration signals generated by various systems to monitor their integrity and predict any abnormal behavior within these systems. For instance, VBCM is commonly adopted in industrial environments where maintenance requirements are predicted based on the machine's or equipment's current condition, ensuring a safe working environment, enhancing productivity, and eliminating costly corrective and preventive maintenance actions. Data-driven VBCM is achieved by training ML models on historical vibration measurement data to learn healthy and abnormal operational conditions. However, despite its advantages, the practical deployment of data-driven VBCM systems faces significant challenges, such as the availability of proper historical data, computational complexity, presence of measurement noise, and high power consumption in vibration sensor nodes. Motivated by the need to develop effective solutions to overcome these challenges, the work presented in this thesis addresses each of the above-mentioned challenges in an effort to accelerate the deployment of data-driven VBCM across various fields and enhance their performance. Accordingly, the work presented in this thesis leverages the advancements of DSP and ML to facilitate reliable and practical data-driven VBCM applications. Specifically, the thesis introduces a similarity-based algorithm along with its open-source software implementation that performs VBCM of rotating machinery using very limited labeled historical vibration data. The software offers a practical solution compatible with other open-source libraries, making it ready for integration within various applications. Additionally, the thesis addresses computational burden and monitoring delay in VBCM systems. These two aspects are crucial for the real-world deployment of VBCM and directly impact safety and financial costs. Specifically, a higher computation burden increases the memory requirements, and a long delay in condition prediction could not prevent costly catastrophic failures. Furthermore, the thesis introduces a framework to facilitate power-efficient VBCM in power-constrained Wireless Sensor Networks (WSN). The framework proposes an innovative method to realize various signal processing operations in the framework using lightweight ML models that can be efficiently implemented on microcontrollers. Additionally, the thesis introduces a comprehensive tutorial on the fundamentals of signal processing, as well as a review of signal-based ML pipeline and feature extraction techniques. The aim is to highlight the crucial role of signal processing in VBCM and to bridge the gap between the two interdisciplinary fields of signal processing and ML by enhancing existing knowledge

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Download

Share

COinS