Accurate and reliable information about the current state of a dynamic process is paramount for plant monitoring, control, optimization, and management. Unfortunately, plant data are often corrupted by measurement errors that reduce the efficiency of strategies devised by plant operators, engineers, and managers to improve plant operations. This book contributes to the development of dynamic data reconciliation (DDR) techniques, namely, optimization-based DDR, predictor-corrector-based DDR, and auto-associative-neural-network-based DDR implemented in real time and closed loop to attenuate the measurement errors. Examples clearly illustrate how measurement errors affect plant monitoring and controller performance, and how the embedded DDR can greatly compensate for this loss of plant monitoring and controller performance. The developed DDR techniques are applied and tested in numerous processes, storage tanks and distillation column, to demonstrate methodologies for practical applications. This book is of interest not only for academics whose research is in process state estimation and signal processing, but also for process control engineers working at plants.
