论文信息 - Quality Monitoring Using Principal Component Analysis and Fuzzy Logic Application in Continuous Casting Process

Quality Monitoring Using Principal Component Analysis and Fuzzy Logic Application in Continuous Casting Process

This paper deals with non linear system monitoring, based on a combined use of Principal Components Analysis (PCA) and fuzzy logic to process and quality monitoring. PCA coupled to fuzzy logic was used to estimate the fault or defect according to the dynamic changes in the process inputsâoutputs characterized by T2 Hoteling and Squared Prediction Error (SPE). Correlation between the relevant process variables and the importance of defects/faults was obtained by a reliable selection of a reduced set of relevant descriptors. The effectiveness of the computing procedure based on fuzzy rule proved by its application to quality estimation of the solidification process in continuous casting.

[1] Duc Truong Pham,et al. State-space identification of dynamic systems using neural networks , 1990 .

[2] Julian Morris,et al. A procedure for determining the topology of multilayer feedforward neural networks , 1994, Neural Networks.

[3] Ai Poh Loh,et al. Neural network modelling and control strategies for a pH process , 1995 .

[4] Dukman Lee,et al. Adaptive fuzzy control of the molten steel level in a strip-casting process , 1996 .

[5] Rolf Isermann,et al. Trends in the Application of Model Based Fault Detection and Diagnosis of Technical Processes , 1996 .

[6] Soundar Kumara,et al. BOUNDARY DEFECT RECOGNITION USING NEURAL NETWORKS , 1997 .

[7] Richard D. Braatz,et al. Fault Detection and Diagnosis in Industrial Systems , 2001 .

[8] Salah Bouhouche,et al. Fault Detection and Monitoring of Length Loop Control System in Pickling Process , 2005, International Conference on Computational Intelligence for Modelling, Control and Automation and International Conference on Intelligent Agents, Web Technologies and Internet Commerce (CIMCA-IAWTIC'06).