As the technology in mechanical engineering advances, increasing complex mechanical systems are developed to achieve better quality and accuracy with prolonged useful life. However, with more components and processes included in the system, the relationship between the input process parameter and output process parameter is difficult to define precisely. Additionally, with more component added to the system, the probability of a breakdown of an individual part increases. Although the quality of the components increases, it does not offset the complication of added mechanical components.
Traditional scheduled maintenance cannot accommodate to the change of configuration of the system in a timely manner. Therefore, condition-based-maintenance (CBM) is implemented to reduce cost and improve reliability. The CBM monitors the actual running condition of the mechanical systems by using data acquired from different sensors. Then, various signal processing techniques are implemented to denoise and extract critical features from the acquired data. After the analysis of the data, prognostics algorithms are used to predicate the remaining useful life (RUL) of the system. Statistical methods are generally added in the prognosis to address the stochastic nature of the mechanical system.
The goal of this study is to implement intelligent algorithms with adaptation to assist the prediction of remaining useful life in the mechanical system and achieve optimized process parameter. More specifically, various signal processing techniques and optimization algorithms will be implemented. The first part of this study focuses on various signal processing techniques to obtain critical feature in the acquired signal from the mechanical system. After the feature extraction is completed, intelligent algorithms are implemented to evaluate the health condition of the current system. Case studies using the vibrational signal of rolling element bearings will be demonstrated to prove the developed algorithm. The second part of this study concentrates on optimization of machining process parameter using machine learning. This part of the study includes a case study of optimization of process parameter of the electro-chemical machining (ECM) process.