Dynamic Simulation and Optimization of Two-Stage Extractive Alcoholic Fermentation Process: Design Impact on Controllability

Abstract

In this paper, we address dynamic controllability of the two different designs of extractive fermentation process, namely one-stage and two-stage designs. The operating conditions that maximize yields and productivity for both designs are determined by optimization using the method of factorial design and response surface analysis. The results show that in terms of the achievable yield and productivity, the performance of the two-stage design is comparable to that of the single-stage, but the former design leads to a significant reduction in the fermentor size required. Furthermore, we analyze the dynamic controllability of the two designs of extractive fermentation process using a so-called control relevant metrics to examine their closed-loop dynamic performance in the face of uncertainty. This analysis reveals that the single-stage design has more favorable dynamic controllability than the two-stage design.

Inter-Communicative Decentralized Multi-Scale Control (ICD-MSC) Scheme: A New Approach to Overcome MIMO Process Interactions

“…Here, we demonstrate the applicability of both I-I and O-I structures of ICD-MSC scheme to the ISP process (27). We choose a 1-way communication of I-I structure.…”

“…The final tunings for the sub-controllers involved are given as follows: Additionally, the following setpoint pre-filters are used for both I-I and O-I structures of ICD-MSC scheme. To evaluate the performance robustness of the different control schemes, we consider a perturbed condition for the model (27), which is based on 50% modeling errors in gains of the diagonal elements; -50% modeling errors in gains of the off-diagonal elements; and 50% modeling errors in time-delays of diagonal and off-diagonal elements. The performance of the ICD-MSC scheme is compared with that of the decentralized PID control scheme of Rajapandiyan and Chidambaram [21] with complete (2-way) decoupling.…”

“…This example further highlights that the proposed scheme based on the feedback control action to mitigate the loop disturbance is indeed more robust against plant uncertainties than the conventional decoupling control scheme, which uses feedforward control action to mitigate the loop disturbance. Note that the TSCEF system is difficult to control not only due to the severe process interactions but also because of its nonlinear dynamic behaviors [27]. A number of researchers have studied the control and optimization strategies for the alcoholic extractive fermentation process, for example, Costa et al [28] studied the control structures for the single-stage alcoholic extractive fermentation process; Sharma and Rangaiah [29] compared the performance of the three-stage fermentation integrated with pervaporation with that of the three-stage fermentation combined with extraction; they found the former to be better; Nandong et al [30] proposed a PCA-based method for the selection of control structure for the two-stage extractive fermentation process.…”

Heuristic-based multi-scale control procedure of simultaneous multi-loop PID tuning for multivariable processes

“…The RGA (21) suggests the use of direct controller pairings and shows significant process interactions in the system. In the past, several researchers used the WB column as an example for comparing the effectiveness of different multi-loop PID design methods; e.g., see [7,10,18].…”

“…The last example is a 5 × 5 MIMO system shown in (26), which is obtained from the two-stage extractive alcoholic fermentation (EAF) process reported in Nandong et al [21]: …”