PROCESS/MODEL MISMATCH COMPENSATION FOR MODEL-BASED CONTROLLERS

Process model-based control algorithms that employ a process model directly in the controller, have been shown to produce good control performance and robust behaviour, despite process modelling errors. However, when the process/model mismatch is large, the closed-loop response, while still being better than responses obtained by conventional controllers, will be degraded. This paper presents a new approach to compensate for process/model mismatch errors, and is based upon the Generic Model Control (GMC) algorithm. This approach is applicable to both linear and nonlinear model-based algorithms. Simulation results are presented to illustrate the efficiency of the approach     

Model-based control strategies for a chemical batch reactor with exothermic reactions

Batch reactor control provides a very challenging problem for the process control engineer. This is because a characteristic of its dynamic behavior shows a high nonlinearity. Since applicability of the batch reactor is quite limited to the effectiveness of an applied control strategy, the use of advanced control techniques is often beneficial. This work presents the implementation and comparison of two advanced nonlinear control strategies, model predictive control (MPC) and generic model control (GMC), for controlling the temperature of a batch reactor involving a complex exothermic reaction scheme. An extended Kalman filter is incorporated in both controllers as an on-line estimator. Simulation studies demonstrate that the performance of the MPC is slightly better than that of the GMC control in nominal case. For model mismatch cases, the MPC still gives better control performance than the GMC does in the presence of plant/model mismatch in reaction rate and heat transfer coefficient.     

Fuzzy logic controller implementation on a microbial electrolysis cell for biohydrogen production and storage

This work presents the implementation of fuzzy logic control (FLC) on a microbial electrolysis cell (MEC). Hydrogen has been touted as a potential alternative source of energy to the depleting fossil fuels. MEC is one of the most extensively studied method of hydrogen production. The utilization of biowaste as its substrate by MEC promotes the waste to energy initiative. The hydrogen production within the MEC system, which involves microbial interaction contributes to the system’s nonlinearity. Taking into account of the high complexity of MEC system, a precise process control system is required to ensure a well-controlled biohydrogen production flow rate and storage application inside a tank. Proportional-derivative-integral (PID) controller has been one of the pioneer control loop mechanism. However, it lacks the capability to adapt properly in the presence of disturbance. An advanced process control mechanism such as the FLC has proven to be a better solution to be implemented on a nonlinear system due to its similarity in human-natured thinking. The performance of the FLC has been evaluated based on its implementation on the MEC system through various control schemes progressively. Similar evaluations include the performance of Proportional-Integral (PI) and PID controller for comparison purposes. The tracking capability of FLC is also accessed against another advanced controller that is the model predictive controller (MPC). One of the key findings in this work is that the FLC resulted in a desirable hydrogen output via MEC over the PI and PID controller in terms of shorter settling time and lesser overshoot.     

Fuzzy logic control of an unstable biological reactor

In the present work, we employ a fuzzy logic controller (FLC) to control the unstable state of a nonlinear biological reaction. The state variable vectors consist of cell density and substrate concentration. The dilution rate is used as a manipulated variable to control the reaction dynamics. An analytic form of FLC employing Zadeh AND logic along with Center of Mass defuzzification method is considered. Simulations reveal that for servo response test, the FLC shows satisfactory performance for natural unsteady states for which a conventional PI controller is known to fail. Further simulations also show that the FLC gives satisfactory regulatory response and is relatively insensitive to the deviations in model parameters.     

Control of milk pasteurization process using model predictive approach

A milk pasteurization process, a nonlinear process and multivariable interacting system, is difficult to control by the conventional on–off controllers since the on–off controller can handled the temperature profiles for milk and water oscillating over the plant requirements. The multi-variable control approach with model predictive control (MPC) is proposed in this study. The proposed algorithm was tested for control of a milk pasteurization process in four cases of simulation such as set point tracking, model mismatch, difference control and prediction horizons, and time sample. The results for the proposed algorithm show the well performance in keeping both the milk and water temperatures at the desired set points without any oscillation and overshoot and giving less drastic control action compared to the cascade generic model control (GMC) strategy.     

The design of experiments,training and implementation of nonlinear controllers based on neural networks

In the area of nonlinear predictive control, several control schemes using artificial neural networks have been proposed. In this work, the issues relating to the information contents of the data used to train the neural network components of these nonlinear predictive control schemes are considered. This raises questions about the design of experiments. A class of feedback-feedforward nonlinear controller based on the model predictive structure (also known as Internal Model Control, IMC, structure) is investigated. The implementation and performance of these neural network based controllers, together with comparisons to other nonlinear and linear controllers, are illustrated on two nonlinear continuous-stirred-tank-reactor simulations.     

Generic model controller tuning for chemical processes with input saturation

Control in the face of process input constraints is very common and of great practical importance in the processing industries. Generic Model Control (GMC) is a model‐based control framework for both linear and nonlinear systems. In this paper, a constrained GMC controller tuning approach using a nonlinear least squares technique is proposed. This tuning approach is simple to apply. For a SISO GMC control system with input saturation, the tracking performance is significantly improved by adding a simple heuristic switching strategy. The effectiveness of the proposed controller tuning approach is demonstrated using dynamic simulations and MIMO real‐time experiments.     

HYBRID INTELLIGENT CONTROL SCHEME FOR AIR H£ATING SYSTEM USING FUZZY LOGIC AND GENETIC ALGORITHM

Fuzzy logic provides a means for converting a linguistic control strategy, based on expert knowledge, into an automatic control strategy. Its performance depends on membership function and rule sets. In the traditional Fuzzy Logic Control (FLC) approach, the optimal membership is formed by trial-and-error method. In this paper, Genetic Algorithm (GA) is applied to generate the optimal membership function of FLC. The membership function thus obtained is utilized in the design of the Hybrid Intelligent Control (HIC) scheme. The investigation is carried out for an Air Heat System ( AHS) an important component of drying process. The knowledge of the optimum PID controller designed, is used to develop the traditional FLC scheme. The computational difficulties in finding optimal memebership function of traditional FLC is allieviated using GA in the design of HIC scheme. The qualitative performance indicies are evaluated for the three control strategies, namely, PID, FLC and HIC. The comparison reveals that the HIC schene designed based on the hybridisation of FLC with GA performs better. Moreover, GA is found to be an effective tool for designing the FLC, eliminating the human interface required to generate the membership functions.     

HYBRID INTELLIGENT CONTROL SCHEME FOR AIR H£ATING SYSTEM USING FUZZY LOGIC AND GENETIC ALGORITHM

ABSTRACT

Fuzzy logic provides a means for converting a linguistic control strategy, based on expert knowledge, into an automatic control strategy. Its performance depends on membership function and rule sets. In the traditional Fuzzy Logic Control (FLC) approach, the optimal membership is formed by trial-and-error method. In this paper, Genetic Algorithm (GA) is applied to generate the optimal membership function of FLC. The membership function thus obtained is utilized in the design of the Hybrid Intelligent Control (HIC) scheme. The investigation is carried out for an Air Heat System ( AHS) an important component of drying process. The knowledge of the optimum PID controller designed, is used to develop the traditional FLC scheme. The computational difficulties in finding optimal memebership function of traditional FLC is allieviated using GA in the design of HIC scheme. The qualitative performance indicies are evaluated for the three control strategies, namely, PID, FLC and HIC. The comparison reveals that the HIC schene designed based on the hybridisation of FLC with GA performs better. Moreover, GA is found to be an effective tool for designing the FLC, eliminating the human interface required to generate the membership functions.     

MODELLING AND CONTROL OF AN ALFALFA ROTARY DRYER

A dynamic model of an alfalfa rotary dryer was developed and used to test the performance of two different feedback controllers. One controller is a conventional PI (Proportional-Integral) controller with fixed tuning parameters whereas the other is a gain-scheduled PI controller with automatically adjusted tuning parameters. The performance of the two controllers was compared with the performance of the dryer under manual control. The gain-scheduled PI controller was found to be superior in the sense that it used less control action and achieved the same control performance as the fixed tuning parameter PI controller. The use of the gain-scheduled controller was shown to reduce energy consumption, increase dryer throughput and had an estimated pay-back time of nine months.     

A model predictive functional control based on proportional-integral-derivative (PID) and proportional-integral-proportional-derivative (PIPD) using extended non-minimal state space: Application to a molten carbonate fuel cell process

The performance of most controllers, including proportional-integral-derivative (PID) and proportional-integral-proportional-derivative (PIPD) controllers, depends upon tuning of control parameters. In this study, we propose a novel tuning strategy for PID and PIPD controllers whose control parameters are tuned using the extended non-minimal state space model predictive functional control (ENMSSPFC) scheme based on the auto-regressive moving average (ARMA) model. The proposed control method is applied numerically in the operation of the MCFC process with the parameters of PID and PIPD controllers being optimized by ENMSSPFC based on the ARMA model for the MCFC process. Numerical simulations were carried out to assess the set-point tracking performance and disturbance rejection performance both for the perfect plant model, which represents the ideal case, and for the imperfect plant model, which is usual in practical applications. When there exists uncertainty in the plant model, the PIPD controller exhibits better overall control performance compared to the PID controller.     

Model based control of a high purity distillation column

In this paper, the problem of dual product composition control of an industrial high purity distillation column, a deisohexanizer (DIH), is addressed using a Generic Model Control framework. A dynamic simulation of the DIH was performed for preliminary studies of the performance of different controller strategies/algorithms. The performance of Generic Model Control incorporating different process models was studied. Process models are presented ranging from simple first order approximations to mechanistic short cut distillation models where a tradeoff between model complexity and model adaptivity is investigated. The different controllers were implemented and compared using a dynamic simulation of an industrial deisohexanizer (DIH) to select the best condidate controller. A controller using a nonlinear process model emerged as the best controller and was implemented on the actual process, resulting in improved performance over the original controller. Simulation results and industrial plant data are presented.     

Toward self-driving processes: A deep reinforcement learning approach to control

Advanced model-based controllers are well established in process industries. However, such controllers require regular maintenance to maintain acceptable performance. It is a common practice to monitor controller performance continuously and to initiate a remedial model re-identification procedure in the event of performance degradation. Such procedures are typically complicated and resource intensive, and they often cause costly interruptions to normal operations. In this article, we exploit recent developments in reinforcement learning and deep learning to develop a novel adaptive, model-free controller for general discrete-time processes. The deep reinforcement learning (DRL) controller we propose is a data-based controller that learns the control policy in real time by merely interacting with the process. The effectiveness and benefits of the DRL controller are demonstrated through many simulations.     

Nonlinear adaptive control algorithm for a multicomponent batch distillation column

In this article, a nonlinear adaptive control strategy is proposed for a multicomponent batch distillation column. The hybrid control scheme consists of a generic model controller (GMC) and a nonlinear adaptive state estimator (ASE). In the first part of the study, an adaptive observer is designed aiming to estimate the partially known parameters based on the measured compositions in the presence of process/predictor mismatch. The open-loop dynamic behavior of the developed ASE estimator is investigated under initialization error, disturbance, and uncertain parameters. In the subsequent part, the adaptive GMC-ASE controller (GMC control structure in conjunction with ASE estimator) has been synthesized for the example distillation column. A simulation-based comparative study has been conducted between the derived nonlinear GMC-ASE control algorithm and a gain-scheduled proportional integral (GSPI) law in terms of constant composition control. The proposed adaptive control scheme is shown to be quite promising due to the exponential error convergence capability of the ASE estimator in addition to the high-quality performance of the GMC controller.     

MODELLING AND CONTROL OF AN ALFALFA ROTARY DRYER

ABSTRACT

A dynamic model of an alfalfa rotary dryer was developed and used to test the performance of two different feedback controllers. One controller is a conventional PI (Proportional-Integral) controller with fixed tuning parameters whereas the other is a gain-scheduled PI controller with automatically adjusted tuning parameters. The performance of the two controllers was compared with the performance of the dryer under manual control. The gain-scheduled PI controller was found to be superior in the sense that it used less control action and achieved the same control performance as the fixed tuning parameter PI controller. The use of the gain-scheduled controller was shown to reduce energy consumption, increase dryer throughput and had an estimated pay-back time of nine months.     

Nitrate control strategies in an activated sludge wastewater treatment process

We studied nitrate control strategies in an activated sludge wastewater treatment process (WWTP) based on the activated sludge model. Two control strategies, back propagation for proportional-integral-derivative (BP-PID) and adaptive-network based fuzzy inference systems (ANFIS), are applied in the WWTP. The simulation results show that the simple local constant setpoint control has poor control effects on the nitrate concentration control. However, the ANFIS (4*1) controller, which considers not only the local constant setpoint control of the nitrate concentration, but also three important indices in the effluent—ammonia concentration, total suspended sludge concentration and total nitrogen concentration—demonstrates good control performance. The results also prove that ANFIS (4*1) controller has better control performance than that of the controllers PI, BP-PID and ANFIS (2*1), and that the ANFIS (4*1) controller is effective in improving the effluent quality and maintaining the stability of the effluent quality.     

FUZZY MODELING AND CONTROL OF BATCH BEER FERMENTATION+

Fuzzy reasoning based modeling of heuristic control rules are employed for control of batch beer fermentation. The effect of different types of membership functions, viz., line, triangular and phi membership functions is evaluated for the fuzzy subset. Various fuzzy model based controllers are presented using two approaches, namely simple fuzzy controller of few rules (FCFR) and rigorous fuzzy controller of many rules (FCM R), and also applied for the temperature control of fermenter. Zadeh's logic and Lukasiewicz's logic are adopted for computing the compositional rule of fuzzy logic inference. The results demonstrate that the proposed fuzzy controllers show better performance than the conventional controllers. FCFR approach provides better control performance, but needs optimum tuning or selection of gains for the fuzzy input and output variables, whereas FCMR approach is preferred due to flexibility in the operation of many control rules. Further, FCMR approach is free from optimum tuning or selection of gains for the fuzzy input and output variables.     

Comparison of control algorithms for the blood glucose concentration in a virtual patient with an artificial pancreas

To obtain the most suitable control algorithm for a wearable artificial pancreas, different control algorithms were compared and tested using a Hovorka model. Model predictive control (MPC), linear and nonlinear model forms, proportional integral derivative control (PID), neural-network-based model predictive control (NN-MPC), nonlinear autoregressive moving average (NARMA-L2) and sequential quadratic programming (SQP) were evaluated using the Hovorka model. Due to the fact that modeling of biomedical processes are very complex, to present the most effective control algorithm, various control strategies were needed to application. In the control algorithms, set point tracking and disturbance rejection were performed. With respect to the rise times of the control algorithms, SQP with optimal control had the shortest time, and NARMA-L2 had the longest time. Because the control algorithm connects the glucose meter and the insulin pump in an artificial pancreas, the rise time is the most important parameter. We propose that optimal control with SQP is the most suitable control algorithm to connect the glucose meter and the insulin pump.     

Tuning PI and fractional order PI controllers with an additional fractional order Pole

The PI controller with an additional pole (PI?+?P) already has been proposed to decrease the noise effect on the control signal. In this paper, a fractional order pole is employed to increase the PI?+?P controller performance. The fractional order is obtained by adjusting the Nyquist plot slope in gain crossover frequency. This condition as well as gain crossover frequency and phase margin specifications are utilized to design the PI controller augmented with an additional fractional order pole (PI?+?FO[P]). To design these two controllers, a first-order plus delay time (FOPDT) model is utilized. For plants that could not be described by this model, its fractional order version (FFOPDT) could be utilized. In this case, a FOPI?+?FO[P] controller is obtained that could improve the transient response of the closed-loop system. The numerical simulations accomplished on various plant models (including chemical plants) demonstrate the effectiveness of the proposed controller comparing with the PI?+?P, PID, FOPI, and fractional order proportional-integral-derivative controllers.     

An adaptive neuro-fuzzy sliding mode controller for MIMO systems with disturbance

This paper introduces the mathematical model of ammonia and urea reactors and suggested three methods for designing a special purpose controller. The first proposed method is Adaptive model predictive controller, the second is Adaptive Neural Network Model Predictive Control, and the third is Adaptive neuro-fuzzy sliding mode controller. These methods are applied to a multivariable nonlinear system as an ammonia–urea reactor system. The main target of these controllers is to achieve stabilization of the outlet concentration of ammonia and urea, a stable reaction rate, an increase in the conversion of carbon monoxide(CO) into carbon dioxide(CO_2) to reduce the pollution effect, and an increase in the ammonia and urea productions, keeping the NH_3/CO_2 ratio equal to 3 to reduce the unreacted CO_2 and NH_3, and the two reactors' temperature in the suitable operating ranges due to the change in reactor parameters or external disturbance. Simulation results of the three controllers are compared. Comparative analysis proves the effectiveness of the suggested Adaptive neurofuzzy sliding mode controller than the two other controllers according to external disturbance and the change of parameters. Moreover, the suggested methods when compared with other controllers in the literature show great success in overcoming the external disturbance and the change of parameters.