Phase Transition Analysis Based Quality Prediction for Multi-phase Batch Processes

摘要: Batch processes are usually involved with multiple phases in the time domain and many researches on process monitoring as well quality prediction have been done using phase information. However, few of them consider transitions, though they exit widely batch non-ignorable impacts product qualities. In present work, a phase-based partial least squares (PLS) method utilizing transition information is proposed to give both online offline predictions. First, divided into several regression parameters other than prior knowledge. Then steady transitions which great influences qualities identified critical-to-quality statistical methods. Finally, based analysis different characteristics phases, an integrated algorithm developed for prediction. The application injection molding shows effectiveness comparison traditional MPLS PLS method.