Nonlinear modeling and state estimation for the Tennessee Eastman challenge process

Intelligent state estimation for fault tolerant nonlinear predictive control

“…The model consists of nine differential equations relating bioreactor states (biomass, substrate, product, dissolved oxygen, CO 2 parameters are given in Birol et al [22]. While carrying out the simulations, we assume that the bioreactor operates isothermally and the fermentor temperature is maintained at the optimum value (25°C) by a perfect temperature controller.…”

“…In addition, hard failures, like failures of actuators and sensors can lead to significant structural plant model mismatch (structure mismatch). The conventional approach to deal with the behavior mismatch in the NMPC formulations is through the introduction of additional artificial states in the state observer [2][3][4]. The main limitation of this permanent augmentation approach is that the number of extra states that can be introduced cannot exceed the number of measurements.…”

“…Moreover, the permanent state augmentation approach cannot systematically deal with the difficulties arising out of sensor biases or actuator/sensor failures. The difficulties encountered while selecting such a subset in design of extended Kalman filter (EKF) for a complex large dimensional system (Tennessee Eastman problem) have been highlighted by Ricker and Lee [2].…”

Interior point SQP strategies for structured process optimization problems

“…In this paper the authors describe the process, present a black box simulator, and purposely gave no model. The model of the Tennessee Eastman Plant used in this example was taken from the work of Ricker and Lee [9]. Since the whole Tennessee plant has 26 differential equations, 83 algebraic constraints and 323 variables, this example confines its scope to the control of the unstable reactor.…”

A Plant-wide Case for Control System Study in Teaching of Process Control Engineering

“…The TE process is a simulated large-scale chemical plants and has been widely investigated as a benchmark for both acadamic and educational purposes [2][3][4][5][6][7]. Many researchers have proposed control systems as solutions to operate the plant [2][3][4][5][6][7], among which the one designed by Ricker [6] is very efficient and prevalent. The control system is a decentralized one, which was designed with both quantitive analysis and expert knowledge.…”