Autoassociative neural networks for robust dynamic data reconciliation

Abstract

in Wiley InterScience (www.interscience.wiley.com).Reliable estimation of process variables for plant monitoring and control is an important topic that has been studied extensively. The Kalman filter has often been used and has acquired an enviable reputation. However, use of the Kalman filter suffers from two restrictive conditions: (1) it requires state-space models and (2) it has to be tuned online to achieve its best performance. Recently, an alternative methodology based on dynamic data reconciliation has been proposed to overcome the first restriction. Although the approach of dynamic data reconciliation can incorporate any form of model, it involves online optimization that may require long computation time for complex systems. This report explores a new methodology based on a combination of autoassociative neural networks (AANNs) and dynamic data reconciliation that overcomes the need for online tuning, as required by the Kalman filter, as well as online optimization as required by conventional dynamic data reconciliation methods. Simulation examples of a distillation column demonstrate that the AANN-based dynamic data reconciliation approach is capable of effectively attenuating measurement noise and is robust to changes of the noise level in plant measurements and the loss of measurements.

Fuzzy supervisory controller tuning and analysis for optimal steam distribution in industry

“…To incorporate process dynamics into AANN structure a recurrent structure can be used with network outputs taken back to the input layer. [9] This has for the consequence that the neural network is not fully auto-associative because additional past inputs of the process are required, and that vector of inputs is not necessary identical to the vector of outputs. However the neural network is auto-associative with respect to the output variables.…”

Non‐linear steady‐state data reconciliation: Theoretical perspective and practical scenarios

“…[ 12 ] Even neural networks have been widely used to solve the problems. [ 13 ] Detailed discussions of the DR, gross error detection, and similar techniques are provided in the literature. [ 14–20 ]…”

“…[12] Even neural networks have been widely used to solve the problems. [13] Detailed discussions of the DR, gross error detection, and similar techniques are provided in the literature. [14][15][16][17][18][19][20] DR techniques are significantly useful in the accurate estimation of plant production/consumption, quality control, and improvements.…”