Fractal component analysis: an integrated approach for autocorrelated signal separation and health monitoring of feedback control system

ABSTRACT

In a feedback control system like the manufacturing equipment or physiological system, its observed data usually manifests a specific autocorrelation structure and mixes with two external signals: disturbance and noise. The external disturbance and noise need to be estimated and compensated by the feedback control mechanism. However, the autocorrelation structure could mask problem signals and affect the monitoring of the system’s inner health. Without separating the mixed signals, neither the evaluation of inner health nor the compensation of external disturbance can be achieved. A new signal separation approach, fractal component analysis (FCA), is developed to identify the autocorrelation structure for improving the health of the feedback control system. As a result, our proposed method outperformed FastICA and principal component analysis (PCA).

KEYWORDS:

• Process monitoring
• fractal structure
• blind source separation
• principle component analysis
• independent component analysis

Additional information

Notes on contributors

Chih-Min Fan

Chih-Min Fan received his Ph.D. degree in electrical engineering from National Taiwan University in 2000. He started his career as an assistant professor at Yuan Ze University for 7 years. After he left academia, he worked in industries. He is now the Chief Data & Analytics Officer at Compal Electronic.

Shao-Jen Weng

Shao-Jen Weng is a Professor of the Department of Industrial Engineering and Enterprise Information at Tunghai University. His research interests are Healthcare Management, Operations Research, Decision Support Systems, Simulation, Quality and Statistics Control, Six Sigma Methodology, and Data Envelopment Analysis (DEA).

Yao-Te Tsai

Yao-Te Tsai received his Ph.D. degree in industrial and systems engineering from Auburn University in 2015. He is currently an Assistant Professor with the Department of International Business, Feng Chia University. His research interests include the Internet of Things, Data Analytics, Supply Chain Management, and Operations Management.

Wei-Hsuan Wu

Wei-Hsuan Wu received his master's degree in industrial engineering and management at Yuan Ze University. He is now a quality engineer at China Engine Corporation.

Chih-Hao Chen

Chih-Hao Chen is pursuing a Ph.D. degree in the Department of Industrial Engineering and Enterprise Information at Tunghai University. His research interests include system simulation and healthcare management.