Adaptive Control of Multivariable Process Using Recurrent Neural Networks

Abstract

Industrial processes are naturally multivariable in nature, which also exhibit non-linear behavior and complex dynamic properties. The multivariable four-tank system has attracted recent attention, as it illustrates many concepts in multivariable control, particularly interaction, transmission zero, and non-minimum phase characteristics that emerge from a simple cascade of tanks. So, the multivariable laboratory process of four interconnected water tanks is considered for modeling and control. For processes which show nonlinear and multivariable characteristics, classical control strategies like PIDs have performance limitations. Hence, intelligent approaches like Neural Networks (NN) is an important term in this juncture. The use of Recurrent Neural Network (RNN) is apt for modeling and control of nonlinear dynamic processes as it contains the past information about the process. The objective of the current study is to design and implement an adaptive control system using RNN for a nonlinear multivariable process.

The proposed adaptive design comprises an estimator based on RNN, which adapts online and predicts one step ahead output. A Recursive Least Square (RLS) based back propagation algorithm is used for training the network. The controller used is also a RNN, which minimizes the difference between the predicted output and reference trajectory. The objective function is minimized using a steepest descent algorithm which gives the optimum control input. Desired performance of the system is ensured by the parallel operation of both. The proposed control strategy is implemented in a laboratory scale four tank system. The trajectory tracking and disturbance rejection response obtained are compared with the response obtained by using a well designed decoupled, decentralized IMC controller.     

Identification and Control of Bioreactor using Recurrent Networks

Abstract

A nonlinear model predictive control (NMPC) strategy based on recurrent neural networks (RNN) is proposed for a single‐input single‐output system (SISO) to control the uncertain nonlinear process. The automatic configuration and modeling of the networks is carried out using a recurrent Elman network using back propagation through time (BPTT) with MATLAB. Identification of the process is performed with a RNN based nonlinear autoregressive with exogenous input (NARX) model and the incorporation of the developed model in the formulation of NMPC is presented. Further, the results of the NMPC is compared with a well tuned IMC based PI controller, which shows a better performance based on the rise time and settling time of the proposed NMPC scheme for the control of an unstable bioreactor.     

Model Based Controller and Linear Cascade Controller Applied to Fed Batch pH Process

Abstract

The present study aims at bringing out the best features of model‐based control, linear cascade control when applied to highly non‐linear systems like pH‐controlled fed‐batch processes. For these processes, control of pH by conventional Proportional‐Integral‐Derivative controller fails to provide satisfactory performance, because of the extreme non‐linearity in the pH dynamics. In the present study, for a fed‐batch neutralization process, a non‐linear control law has been derived for the model‐based Proportional Integral controller. Typical problems in process control like sampling, delay and perturbations in model parameters are addressed in this study using model‐based control. The simulation results show the superior performance and robustness of the model‐based controller and linear cascade controller over that of the conventional Proportional Integral controller.     

Adaptive Control of Neutralization Process using Neural Networks

Abstract

A neural network (NN) based adaptive interaction technique is proposed for controlling highly nonlinear neutralization processes. In this approach, the controller is decomposed into interconnected subsystems and adaptation occurs during the interactions. This approach is adaptive in structure and doesn't use an explicit model of the process in the design. The NN is used to establish the adaptive interaction technique for the development of a nonlinear pH controller, which calculates the necessary change in a manipulated variable to drive the system to the desired value. By applying this adaptive algorithm, the same adaptation as the back‐propagation algorithm is achieved without the need of backward propagating the error throughout a feedback network. This important property makes it possible to adapt the NN controller directly without a process model. This advantage reduces the computational complexity drastically in comparison to the well known back‐propagation algorithm based adaptive NN system and a model based system. The designed model‐free online adaptive controller was implemented to a laboratory scaled pH process in real time by use of a dSPACE 1104 interfacing card. The responses of pH and acid flow rate show good tracking for both the set point and load changes over the entire nonlinear region.     

Wiener model based GMVC design considering sensor noise and delay

Even though there is a plethora of literature available for assessing linear control loop performance, they cannot be applied to the nonlinear control loops. In this paper, a nonlinear generalized minimum variance (NGMV) controller based on a single input–single output (SISO) Wiener model is proposed. The NGMV controller’s performance is used as a benchmark for a class of nonlinear control loops. The advantage of the proposed method is ability of online parameter estimation of the nonlinear model using common recursive least squares (RLS) method. In real-world applications, sensor and measurement tools force noises and extra delay to the control loop which poses limitations on achievable control performance. Hence, the classic control performance assessment techniques, is not attainable anymore. To handle the limitation caused by sensor delay, the k-step ahead prediction method is utilized. Further, the exponential digital filter is used in order to attenuate impact of the measurement noise on the controller. To show the effectiveness of the proposed method, a simulation test on a pH neutralization process is carried out.     

Nonlinear control of a multicomponent distillation process coupled with a binary distillation model as an EKF predictor

The work is devoted to design the globally linearizing control (GLC) strategy for a multicomponent distillation process. The control system is comprised with a nonlinear transformer, a nonlinear closed-loop state estimator [extended Kalman filter (EKF)], and a linear external controller [conventional proportional integral (PI) controller]. The model of a binary distillation column has been used as a state predictor to avoid huge design complexity of the EKF estimator. The binary components are the light key and the heavy key of the multicomponent system. The proposed GLC-EKF (GLC in conjunction with EKF) control algorithm has been compared with the GLC-ROOLE [GLC coupled with reduced-order open-loop estimator (ROOLE)] and the dual-loop PI controller based on set point tracking and disturbance rejection performance. Despite huge process/predictor mismatch, the superiority of the GLC-EKF has been inspected over the GLC-ROOLE control structure.     

Real Time Control of pH in Fed‐Batch Process

Abstract

The objective of this work is to implement the linear, non linear model based and linear cascade controllers to control pH in a fed batch neutralisation process in real time and compare the performance with the simulation results. The control objective here is to make the process output pH to follow the given reference trajectory. This work aims at bringing out the best features of model‐based control and linear cascade control, when applied to highly non‐linear systems like pH‐controlled fed‐batch processes. For these processes, control of pH by a conventional Proportional Integral Derivative controller fails to provide satisfactory performance because of the extreme non‐linearity in the pH dynamics. Typical problems in control, e.g., uncertainty in model parameters, are addressed in this work. These controllers are implemented in real time using a lab scale setup and compared with the simulation. The results show the superior performance of the non linear model‐based and linear cascade controller over that of the conventional Proportional Integral controller.     

Neural networks for tracking of unknown SISO discrete-time nonlinear dynamic systems

This article presents a Lyapunov function based neural network tracking (LNT) strategy for single-input, single-output (SISO) discrete-time nonlinear dynamic systems. The proposed LNT architecture is composed of two feedforward neural networks operating as controller and estimator. A Lyapunov function based back propagation learning algorithm is used for online adjustment of the controller and estimator parameters. The controller and estimator error convergence and closed-loop system stability analysis is performed by Lyapunov stability theory. Moreover, two simulation examples and one real-time experiment are investigated as case studies. The achieved results successfully validate the controller performance.     

Multi-linear model set design based on the nonlinearity measure and H-gap metric

This paper proposes a model bank selection method for a large class of nonlinear systems with wide operating ranges. In particular, nonlinearity measure and H-gap metric are used to provide an effective algorithm to design a model bank for the system. Then, the proposed model bank is accompanied with model predictive controllers to design a high performance advanced process controller. The advantage of this method is the reduction of excessive switch between models and also decrement of the computational complexity in the controller bank that can lead to performance improvement of the control system. The effectiveness of the method is verified by simulations as well as experimental studies on a pH neutralization laboratory apparatus which confirms the efficiency of the proposed algorithm.     

自调整模糊PID及其在pH值控制中的仿真研究

酸碱中和反应中pH值变化的严重非线性及时滞特性给其控制带来了极大的困难。针对这一难题,采用自调整模糊PID控制算法对其进行控制,在传统PID控制器的基础上加入了模糊推理机制,由于模糊控制无需对象的精确模型,因此其具有很好的自适应特性,并且能够克服非线性因素带来的影响,具有较强的鲁棒性。对洗衣粉生产的磺化、分酸工段中的pH值控制的仿真实验结果验证了自调整模糊PID控制器在pH值控制中具有较强的鲁棒性和抗干扰能力。     

Synthesis of nonlinear adaptive controller for a batch distillation

A nonlinear adaptive control strategy is proposed for a binary batch distillation column. The hybrid control algorithm comprises a generic model controller (GMC) and a nonlinear adaptive state estimator (ASE). The adaptive observation scheme mainly estimates the imprecisely known parameters based on the available tray temperature measurements. The sensitivity of the proposed estimator is investigated with respect to the effect of initialization error, unmeasured disturbance and uncertainty. Then, a comparative study is carried out between the derived nonlinear GMC-ASE controller and a traditional proportional integral law in terms of set point tracking and disturbance rejection performance. The study also includes the effect of measurement noise and parametric uncertainty on the closed-loop performance. The proposed adaptive control algorithm is shown to be quite promising due to the exponential error convergence capability of the ASE estimator in addition to the high-quality control action provided by the GMC controller.     

Control System Design for a Neutralization Process using Block Oriented Models

Abstract

This paper focuses on the development of a non‐linear controller for a neutralization process. Block oriented models, namely the Wiener and Hammerstein model structures, are used for the controller design. A neural network architecture that has the capability to model the steady state behavior of a complex non‐linear process is developed. The dynamic behavior is modeled with a linear model. The pH process considered in this study exhibits drastic changes in the gain, even over a small operating range. In this study, the performance of controllers designed using Weiner and Hammerstein models are compared with a PI controller for servo and regulatory changes. The comparison results based on integral square error (ISE) values shows that the Weiner model based controller is suitable for a pH process.     

Real-Time Simulator based hybrid control of DFIG-WES

This paper presents the analysis and design of recurrent neural network (RNN) and proportional and integral(PI) controller based hybrid control(HC) of doubly fed induction generator (DFIG). The proposed HC shows the quick dynamic and good transient response during the sudden changes in wind speed as well as generator speed. Further performance of proposed hybrid controller has been analyzed independently with RNN and PI control as its components. The HC along with RNN gives effective performance compared to conventional DTC (CDTC) and PI DTC in terms of flux ripples,torque ripples and settling time. The proposed technique is implemented in real-time simulator (RTS) based OPAL-RT and MATLAB/SIMULINK environment at a rating of 5.5 KW system under steeply and randomly change in wind velocity.     

Adaptive neural network controller for the molten steel level control of strip casting processes

The twin-roll strip casting process is a steel-strip production method which combines continuous casting and hot rolling processes. The production line from molten liquid steel to the final steel-strip is shortened and the production cost is reduced significantly as compared to conventional continuous casting. The quality of strip casting process depends on many process parameters, such as molten steel level in the pool, solidification position, and roll gap. Their relationships are complex and the strip casting process has the properties of nonlinear uncertainty and time-varying characteristics. It is difficult to establish an accurate process model for designing a model-based controller to monitor the strip quality. In this paper, a model-free adaptive neural network controller is developed to overcome this problem. The proposed control strategy is based on a neural network structure combined with a sliding-mode control scheme. An adaptive rule is employed to on-line adjust the weights of radial basis functions by using the reaching condition of a specified sliding surface. This surface has the on-line learning ability to respond to the system’s nonlinear and time-varying behaviors. Since this model-free controller has a simple control structure and small number of control parameters, it is easy to implement. Simulation results, based on a semiexperimental system dynamic model and parameters, are executed to show the control performance of the proposed intelligent controller. In addition, the control performance is compared with that of a traditional PID controller.     

基于GA的NMPC在污水处理pH值控制中的应用

污水处理pH中和反应过程具有高度非线性和滞后性。文中首先介绍了污水处理pH值控制过程及其机理模型,然后对其进行分析研究,提出了1种基于遗传优化方法的非线性预测控制策略。利用遗传算法的全局寻优、参数调节简单等优点,解决非线性优化难的问题。仿真实验说明,该方法能够对设定的pH值进行很好的跟踪,并且响应时间较快,实现了对pH值的有效控制。     

Design and Implementation of a Flexible Manufacturing Control System Using Neural Network

Design and implementation of a sequential controller based on the concept of artificial neural networks for a flexible manufacturing system are presented. The recurrent neural network (RNN) type is used for such a purpose. Contrary to the programmable controller, an RNN-based sequential controller is based on a definite mathematical model rather than depending on experience and trial and error techniques. The proposed controller is also more flexible because it is not limited by the restrictions of the finite state automata theory. Adequate guidelines of how to construct an RNN-based sequential controller are presented. These guidelines are applied to different case studies. The proposed controller is tested by simulations and real-time experiments. These tests prove the successfulness of the proposed controller performances. Theoretical as well as experimental results are presented and discussed indicating that the proposed design procedure using Elman's RNN can be effective in designing a sequential controller for event-based type manufacturing systems. In addition, the simulation results assure the effectiveness of the proposed controller to outperform the effect of noisy inputs.     

Type-2 fuzzy model based controller design for neutralization processes

In this study, an inverse controller based on a type-2 fuzzy model control design strategy is introduced and this main controller is embedded within an internal model control structure. Then, the overall proposed control structure is implemented in a pH neutralization experimental setup. The inverse fuzzy control signal generation is handled as an optimization problem and solved at each sampling time in an online manner. Although, inverse fuzzy model controllers may produce perfect control in perfect model match case and/or non-existence of disturbances, this open loop control would not be sufficient in the case of modeling mismatches or disturbances. Therefore, an internal model control structure is proposed to compensate these errors in order to overcome this deficiency where the basic controller is an inverse type-2 fuzzy model. This feature improves the closed-loop performance to disturbance rejection as shown through the real-time control of the pH neutralization process. Experimental results demonstrate the superiority of the inverse type-2 fuzzy model controller structure compared to the inverse type-1 fuzzy model controller and conventional control structures.     

非线性状态空间方法辨识电液伺服控制系统

针对回归神经网络辨识和建立非线性动态系统模型的问题,研究非线性状态空间描述的回归神经网络数学模型。讨论极小均方误差网络训练收敛准则,通过研究Kalman 滤波估计公式中的随机变量,提出一种参数增广的回归神经网络非线性状态方程,无导数的Kalman滤波器用于增广参数估计,人工白噪声强迫网络学习,更新网络权值,避免了扩展Kalman滤波器计算Jacobian信息和基于递度学习算法收敛慢的问题。在电液伺服系统辨识建模的应用中表明,回归神经网络较好地跟踪了液压油缸压力变化,与扩展Kalman滤波估计学习算法相比,新的算法具有较快的收敛和精度。     

Synchronization controller design of two coupling permanent magnet synchronous motors system with nonlinear constraints

In this paper, two coupling permanent magnet synchronous motors system with nonlinear constraints is studied. First of all, the mathematical model of the system is established according to the engineering practices, in which the dynamic model of motor and the nonlinear coupling effect between two motors are considered. In order to keep the two motors synchronization, a synchronization controller based on load observer is designed via cross-coupling idea and interval matrix. Moreover, speed, position and current signals of two motor all are taken as self-feedback signal as well as cross-feedback signal in the proposed controller, which is conducive to improving the dynamical performance and the synchronization performance of the system. The proposed control strategy is verified by simulation via Matlab/Simulink program. The simulation results show that the proposed control method has a better control performance, especially synchronization performance, than that of the conventional PI controller.     

Active Vibration Control of SMA Actuated Structures using Fast Output Sampling Based Sliding Mode Control

Abstract

Fast Output Sampling (FOS) based discrete sliding mode control is designed for Shape Memory Alloy (SMA) actuated structures to suppress the structural vibration. For the design of discrete sliding mode control the reaching law, method proposed by W. Gao is used. The simulation results demonstrate the performance of the controller. The proposed control technique achieves good vibration suppression. This methodology is more practical and easier to implement, since the measurement or estimation of the system states is not needed for designing the controller.