Study on data compression feasibility of sparse coding for machine condition-based monitoring

• Main Author: Nicholas Sadjoli
• Other Authors: School of Electrical and Electronic Engineering
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Simulation and modeling
• Online Access: http://hdl.handle.net/10356/75191
• Institution: Nanyang Technological University

On-time milling machine maintenance and upkeep are critical to ensure on-time production in fabrication cells. In recent years, Condition-Based Monitoring (CBM) has been regarded as one of the most state-of-the-art machine maintenance techniques that can significantly lower unscheduled maintenance cost and provide greater efficiency. This however means that more data needs to be streamed and stored for CBM to work properly, which might prove costly in the long run in terms of data storage. This project therefore focuses on the feasibility of using sparse coding as a method for signal compression for CBM purposes. Specifically, the feasibility of the method will be measured and compared with other traditional compression methods for vibration signals coming from the spindle of machines. Literature review on the sparse coding and other traditional compression methods was first conducted to evaluate and identify the main concepts behind each compression method. Following which, a number of performance tests were designed and conducted to determine the best algorithm for use in the signal approximation process of sparse coding. A second series of tests would then be designed and conducted using sparse coding and the traditional discrete cosine transform (DCT) and discrete wavelet transform (DWT) to compress samples of vibration signals of an actual running machine’s spindle, from which parameters such as compression speed, accuracy, and memory usage of each methods will be measured. Finally, based on the collected measurements, a comparison amongst the compression performances of sparse coding against the more traditional methods such as DCT and DWT reviews that sparse coding is indeed the most suitable method for compressing the spindle vibration signals due to it obtaining much better compression accuracy while still maintaining sufficient compression speed. This result then validates the feasibility of the sparse coding for use as a compression method in a CBM process.