Temperature and concentration control of exothermic chemical processes in continuous stirred tank reactors

Abstract

Exothermic chemical reaction taking place in continuous stirred tank reactor is considered. Heat release from the chemical reaction, non-linear dynamic behavior of the process and uncertainty in parameters are the main factors motivating the use of robust control design. Viewing temperature and molar concentration as variables both accessible in real time, PI and optimal state-feedback controllers driven by temperature and concentration error signals are proposed to regulate the system over reactor’s steady-state working points by counteracting undesired disturbances. Since access to concentration value has proved beneficial for the reactor’s performance, estimation techniques are examined to compensate for the problematic nature of the concentration’s measurement. A linear reduced-order observer is first proposed to estimate the concentration value using temperature measurements. In addition, assuming concentration measurement is available with a relatively short delay via sample analysis, a linear and non-linear discrete-time predictor is constructed to estimate the concentration’s real-time value. A linear combination of the two estimation schemes (observer, predictor) is proposed resulting in a combined estimator, in which the emphasis between the two individual schemes can be controlled via a scalar parameter. The work presented in this paper was supported by the GLOW project – New weather-stable low gloss powder coatings based on bifunctional acrylic solid resins and nanoadditives – as part of the development of novel and efficient processing technologies regarding the production of new families of powder coatings, responding to industrial requirements for quality improvement at lower cost and shorter development cycles.

Residual Neural Network-Based Observer Design for Continuous Stirred Tank Reactor Systems

“…For CSTR reaction systems, the factors that have the greatest impact on the quality and yield of the product are the temperature in the vessel and the concentration of the product [15]. Accurate monitoring of these two factors is critical to the CSTR reaction process.…”

Markov Chain Monte Carlo Analysis of the Variable-Volume Exothermic Model for a Continuously Stirred Tank Reactor

“…Authors in [14] investigated the performance of CSTR as bioreactor for producing biohydrogen from water melon waste in the anaerobic digester. The Lyapunov-based stochastic non-linear model predictive control was used in [15] to shape the state probability density functions in the CSTR with the exothermic reaction k A B → , where ݇ is the reaction rate, ‫ܣ‬ is the reactant and ‫ܤ‬ is the product. The Luenberger fuzzy observer with and without sliding modes, the Walcott-Zak fuzzy observer, and the Utkin fuzzy observer were adopted and used as fault detection sensors of the CSTR in [16].…”

A practical robust yaw servo architecture of ROVs by multi-vector propulsion and nonlinear controller

“…This is very challenging, because ROVs typically travel in complicated ocean environments with time-varying and uneven currents (Fan and Woolsey, 2013). Within this extreme underwater nonlinear environment, it is critical to design a vehicle that adapts to a nonlinear environment (Vlahakis and Halikias, 2019).…”