ARTIFICIAL NEURAL NETWORKS: PRINCIPLE AND APPLICATION TO MODEL BASED CONTROL OF DRYING SYSTEMS - A REVIEW

This paper reviews the developments in the model based control of drying systems using Artificial Neural Networks (ANNs). Survey of current research works reveals the growing interest in the application of ANN in modeling and control of non-linear, dynamic and time-variant systems. Over 115 articles published in this area are reviewed. All landmark papers are systematically classified in chronological order, in three distinct categories; namely, conventional feedback controllers, model based controllers using conventional methods and model based controllers using ANN for drying process. The principles of ANN are presented in detail. The problems and issues of the drying system and the features of various ANN models are dealt with upto-date. ANN based controllers lead to smoother controller outputs, which would increase actuator life. The paper concludes with suggestions for improving the existing modeling techniques as applied to predicting the performance characteristics of dryers. The hybridization techniques, namely, neural with fuzzy logic and genetic algorithms, presented, provide, directions for pursuing further research for the implementation of appropriate control strategies. The authors opine that the information presented here would be highly beneficial for pursuing research in modeling and control of drying process using ANN     

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.     

Modelling of A Fluidized Bed Drier Using Artificial Neural Network

ABSTRACT

Proper modelling of a fluidized bed drier (FBD) is important to design model based control strategies. A FBD is a non-linear multivariable system with non-minimum phase characteristics. Due to the complexities in FBD conventional modelling techniques are cumbersome. Artificial neural network (ANN) with its inherent ability to “learn” and “absorb” non-linearities, presents itself as a convenient tool for modelling such systems.

In this work, an ANN model for continuous drying FBD is presented. A three layer fully connected feedfordward network with three inputs and two outputs is used. Backpropagation learning algorithm is employed to train the network. The training data is obtained from computer simulation of a FBD model from published literature. The trained network is evaluated using randomly generated data as input and observed to predict the behaviour of FBD adequately.     

Modeling Intermittent Drying Using an Adaptive Neuro-Fuzzy Inference System

ABSTRACT

Artificial intelligence systems such as artificial neural networks (ANN) and fuzzy inference systems (FIS) are widely accepted as a technology offering an alternative way to tackle complex and ill-defined problems. The advantages of a combination of ANN and FIS are obvious. This article presents the application of a hybrid neuro-fuzzy system called adaptive-network-based fuzzy inference system (ANFIS) to time dependent drying processes and is illustrated by an application to model intermittent drying of grains in a spouted bed. An introduction to the ANFIS modeling approach is also presented. The model showed good performance in terms of various statistical indices.     

Prediction of Timber Kiln Drying Rates by Neural Networks

The purpose of this exploratory work was to apply artificial neural network (ANN) modeling to the prediction of timber kiln drying rates based on species and basic density information for the hem-fir mix that grows along the local coastal areas. The ANN models with three inputs (initial moisture content, basic density, and drying time) were developed to predict one output, namely, average final moisture content. The back-propagation algorithm, the most common neural network learning method, was implemented for testing, training, and validation. Optimal configuration of the network model was obtained by varying its main parameters, such as transfer function, learning rule, number of neurons and layers, and learning runs. Accurate prediction of the experimental drying rate data by the ANN model was achieved with a mean absolute relative error less than 2%, thus supporting the powerful predictive capacity of this modeling method.     

Prediction of Timber Kiln Drying Rates by Neural Networks

The purpose of this exploratory work was to apply artificial neural network (ANN) modeling to the prediction of timber kiln drying rates based on species and basic density information for the hem-fir mix that grows along the local coastal areas. The ANN models with three inputs (initial moisture content, basic density, and drying time) were developed to predict one output, namely, average final moisture content. The back-propagation algorithm, the most common neural network learning method, was implemented for testing, training, and validation. Optimal configuration of the network model was obtained by varying its main parameters, such as transfer function, learning rule, number of neurons and layers, and learning runs. Accurate prediction of the experimental drying rate data by the ANN model was achieved with a mean absolute relative error less than 2%, thus supporting the powerful predictive capacity of this modeling method.     

Infrared Drying Process of an Experimental Water Painting: Model Predictive Control

Abstract

This article deals with the experimental control of an infrared drying process of a water based epoxy-amine painting. This approach is based on a unidirectional diffusional modeling of infrared drying phenomena where both heat and mass transfers under shrinkage conditions are accounted for. The control problem is concerned with the tracking of any given trajectory for one of the characteristics (i.e., the temperature or the mean water content) during the drying cycle. This is solved using the well-known model predictive control framework where the nonlinear diffusional model is directly used in the control formulation. Experimental results show the efficiency of the trajectory tracking. This method can be extended for more general constrained control problem.     

Optimal Control of the Primary Drying Stage of Freeze Drying of Solutions in Vials Using Variational Calculus

Abstract

The problem of operating freeze drying of pharmaceutical products in vials placed in trays of a freeze dryer to remove free water (in frozen state) at a minimum time was formulated as an optimal control problem. Two different types of freeze dryer designs were considered. In type I freeze dryer design, upper and lower plate temperatures were controlled together, while in type II freeze dryer design, upper and lower plate temperatures were controlled independently. The heat input to the material being dried and the drying chamber pressure were considered as control variables. Constraints were placed on the system state variables by the melting and scorch temperatures during primary drying stage. Necessary conditions of optimality for the primary drying stage of freeze drying process in vials are derived and presented. Furthermore, an approach for constructing the optimal control policies that would minimize the drying time for the primary drying stage was given. In order to analyze optimal control policy for the primary drying stage of the freeze-drying process in vials, a rigorous multi-dimensional unsteady state mathematical model was used. The theoretical approach presented in this work was applied in the freeze drying of skim milk. Significant reductions in the drying times of primary drying stage of freeze drying process in vials were obtained, as compared to the drying times obtained from conventional operational policies.     

A Simulation Tool for the Optimization of Lumber Drying Schedules

Abstract

A two-dimensional wood drying model based on the water potential concept is used to simulate the convection batch drying of lumber at conventional temperature. The model computes the average drying curve, the internal temperature and moisture content profiles, and the maximum effective moisture content gradient through board thickness. Various scenarios of conventional kiln-drying schedules are tested and their effects on drying time, maximum effective moisture content gradient, final moisture content distribution within and between boards, and energy consumption are analyzed. Simulations are performed for two softwood species, black spruce (Picea mariana (Mill.) B.S.P.) and balsam fir (Abies balsamea (L.) Mill.). The simulation results indicate that the predictive model can be a very useful tool to optimize kiln schedules in terms of drying time, energy consumption, and wood quality. Such a model could be readily combined with intelligent adaptive kiln controllers for on-line optimization of the drying schedules.     

Application of Artificial Neural Networks (ANNs) in Drying Technology: A Comprehensive Review

Inspired by the functional behavior of the biological nervous system of the human brain, the artificial neural network (ANN) has found many applications as a superior tool to model complex, dynamic, highly nonlinear, and ill-defined scientific and engineering problems. For this reason, ANNs are employed extensively in drying applications because of their favorable characteristics, such as efficiency, generalization, and simplicity. This article presents a comprehensive review of numerous significant applications of the ANN technique to solve problems of nonlinear function approximation, pattern detection, data interpretation, optimization, simulation, diagnosis, control, data sorting, clustering, and noise reduction in drying technology. We summarize the use of the ANN approach in modeling various dehydration methods; e.g., batch convective thin-layer drying, fluidized bed drying, osmotic dehydration, osmotic-convective drying, infrared, microwave, infrared- and microwave-assisted drying processes, spray drying, freeze drying, rotary drying, renewable drying, deep bed drying, spout bed drying, industrial drying, and several miscellaneous applications. Generally, ANNs have been used in drying technology for modeling, predicting, and optimization of heat and mass transfer, thermodynamic performance parameters, and quality indicators as well as physiochemical properties of dried products. Moreover, a limited number of researchers have focused on control of drying systems to achieve desired product quality by online manipulating of the drying conditions using previously trained ANNs. Opportunities and limitations of the ANN technique for drying process simulation, optimization, and control are outlined to guide future R&D in this area.     

DEVELOPMENT OF ANN MODEL FOR NON-LINEAR DRYING PROCESS

ABSTRACT

This paper presents an application of artificial neural network (ANN) technique to develop a model representing the non-linear drying process. The air heat plant (AHP), an important component in drying process is fabricated and used for building the ANN model. An optimal feed forward neural network topology is identified for the air heating system set-up. The training sets are obtained from experimental data. Back propogation algorithm with momentum factor is used for training. The results show that the back propogation ANN can learn the functional mapping between input and output. The advantages of ANN model developed for AHP is highlighted. The developed model can be used for control purposes.     

An Artificial Neural Network Model for Prediction of Drying Rates

Abstract

Drying rate data were generated for training of an ANN model using a liquid diffusion model for potato slices of different thicknesses using air at different velocities, humidities and temperatures. Moisture content and temperature dependence of the liquid diffusivity as well as the heat of wetting for bound moisture were included in the diffusion model making it a highly nonlinear system. An ANN model was developed for rapid prediction of the drying rates using the Page equation fitted to the drying rate curves. The ANN model is verified to provide accurate interpolation of the drying rates and times within the ranges of parameters investigated.     

MULTIVARIABLE PROCESS IDENTIFICATION AND CONTROL OF CONTINUOUS FLUIDIZED BED DRYERS

ABSTRACT

Dynamic models that rigorously describe fluidized bed dryers based on the fundamental principles of the process are usually so complex to be employed in control system design. To obtain simple reduced-order models for such systems, a sequence of step changes in the manipulated and load variables is introduced into the rigorous model. The obtained input–output dynamic response data are used for off-line model identification. Different types of linear models are generated, which are shown to be adequately representing the fluidized bed drying dynamics. The derived models are useful to develop model-based control algorithms such as Internal Model Control (IMC) and Model Predictive Control (MPC). Performance and robustness properties of these controllers are analyzed. Simulation results demonstrate a good performance in terms of tracking and load rejection capabilities.     

Control of periodically operated reactors

Control of periodically operated reactors has in common with control of reactors operating at steady state the objective of minimizing the effect of disturbances on an objective function such as the cost of a product or the deviation of an outlet concentration of a pollutant from a statuary target. Simple feedback control employing feedback PID regulators, however, is not “adequate for most disturbances because of problems with tracking a time-varying output and the necessarily non-linear character of the reactors with respect to controlled as well as uncontrolled inputs. This contribution is a review of the literature and a discussion of research needs. The literature on the control of periodically operated reactors is not voluminous. Nevertheless this literature clearly indicates that model based predictive controllers can be used for this type of reactor”. Further research on the limitations, maintenance and implementation costs of model based controllers in this application would be worthwhile. Experimental studies are wholly absent. Unique regulators for other periodic operations, such as adaptive control or forcing the output toward a reference trajectory using an open loop model based control strategy, certainly warrant study of their application to periodically operated reactors. Additionally, proper design of the reactor may lead to configurations that are simpler to control and that may not require complex control strategies for efficient operation.     

DYNAMIC AND HYBRID NEURAL MODEL OF THERMAL DRYING IN A FLUTDIZED BED

ABSTRACT

A preliminary study aimed at comparing Classical Dynamic Neural Modelling (CDNM) and Hybrid Neural Modelling (HNM) to describe thermal dewatering process in a fluidized bed is presented Two schemes of HN modelling were developed to find the most efficient way of combining a classical mathematical model of the process and Artificial Neural Network (ANN). CDN model was developed using “moving window” technique. In the first scheme of HNM a feed-forward ANN was trained to predict evaporation rate and heat flux in the drying process. In the second scheme of the HN model, ANN was used to determine heat transfer coefficient only. Excellent prediction of drying process by HNM is proved.     

An Investigation of Drying Process of Shelled Pistachios in a Newly Designed Fixed Bed Dryer System by Using Artificial Neural Network

In this paper, the drying of Siirt pistachios (SSPs) in a newly designed fixed bed dryer system and the prediction of drying characteristics using artificial neural network (ANN) are presented. Drying characteristics of SSPs with initial moisture content (MC) of 42.3% dry basis (db) were studied at different air temperatures (60, 80, and 100 °C) and air velocities (0.065, 0.1, and 0.13 m/s) in a newly designed fixed bed dryer system. Obtained results of experiments were used for ANN modeling and compared with experimental data. Falling rate period was observed during all the drying experiments; constant rate period was not observed. Furthermore, in the presented study, the application of ANN for predicting the drying time (DT) for a good quality product (output parameter for ANN modeling) was investigated. In order to train the ANN, experimental measurements were used as training data and test data. The back propagation learning algorithm with two different variants, so-called Levenberg–Marguardt (LM) and scaled conjugate gradient (SCG), and tangent sigmoid transfer function were used in the network so that the best approach can be determined. The most suitable algorithm and neuron number in the hidden layer are found out as LM with 15 neurons. For this number level, after the training, it is found that Root-mean squared (RMS) value is 0.3692, and absolute fraction of variance (R²) value is 99.99%. It is concluded that ANNs can be used for prediction of drying SSPs as an accurate method in similar systems.     

Modeling Intermittent Drying Using an Adaptive Neuro-Fuzzy Inference System

Artificial intelligence systems such as artificial neural networks (ANN) and fuzzy inference systems (FIS) are widely accepted as a technology offering an alternative way to tackle complex and ill-defined problems. The advantages of a combination of ANN and FIS are obvious. This article presents the application of a hybrid neuro-fuzzy system called adaptive-network-based fuzzy inference system (ANFIS) to time dependent drying processes and is illustrated by an application to model intermittent drying of grains in a spouted bed. An introduction to the ANFIS modeling approach is also presented. The model showed good performance in terms of various statistical indices.     

On-Farm Drying and Storage Systems

ABSTRACT

Rotary dryers are widely used for the continuous drying of minerals and chemicals on a large scale. Hot gases are passed parallel to the flowing solid to achieve the desired product moisture content. Because these dryers are energy intensive, it is mandatory to operate them as efficiently as possible to respond to economic pressures. Using a dynamic rotary dryer simulator for mineral concentrate, five control strategies are evaluated and compared. Two control strategies are based on PI controllers and the others use neural network models. Results clearly show that a feedforward action, in conjunction with a PI controller or incorporated within the structure of a neural network model, led to the best performances provided an accurate measurement of the feed moisture content is available.     

OPTIMAL CONTROL OF THE PRIMARY AND SECONDARY DRYING STAGES OF BULK SOLUTION FREEZE DRYING IN TRAYS

Abstract

The problem of operating a tray freeze dryer to obtain a desired final bound water content in minimum time is formulated as an optimal control problem with the use of the rigorous unsteady state mathematical model of Sadikoglu and Liapis [9] that has been found to describe satisfactorily the experimental dynamic behavior of the primary and secondary drying stages of bulk solution freeze drying of pharmaceuticals in trays. The heat input to the material being dried and the drying chamber pressure are considered to be control variables. Constraints are placed on the system state variables by the melting and scorch temperatures during primary drying, and by the scorch temperature during secondary drying. Necessary conditions of optimality for both the primary and secondary drying stages are derived and presented, and an approach for constructing the optimal control policies that would minimize the drying times for both the primary and secondary drying stages, is presented. The theoretical approach presented in this work was applied in the freeze drying of skim milk, and significant reductions in the drying times of primary and secondary drying were obtained, when compared with the drying times obtained using the operational policies reported in the literature, by using the optimal control policies constructed from the theory presented in this work. Furthermore, it is shown that the optimal control policy leads to the desired in practice result of having at the end of secondary drying temperature and bound water concentration profiles (in the dried layer) whose gradients are very small. It is also shown that by using the optimal control policy and an excipient capable of increasing the melting temperature without affecting product quality, one can significantly reduce the drying time of the primary drying stage.     

Rotary Dryer Control Using a Grey-Box Neural Model Scheme

The application of a Grey-box Neural Model (GNM) in a nonlinear model predictive control scheme (NMPC) of a direct rotary dyer is presented in this work. The GNM, which is based on the combination of phenomenological models and empirical artificial neural network (ANN) models, was properly developed and validated by using experimental fish-meal rotary drying information. The GNM was created by combining the rotary dryer mass and energy balances and a feed forward neural network (FFNN), trained off-line to estimate the drying rate and the volumetric heat transfer coefficient. The GNM results allowed us to obtain the relation between the controlled variable (solid moisture content) and the manipulated variable (gas phase entrance temperature) used in the predictive control strategy. Two NMPC control strategies, one with a fixed extended prediction horizon and another with an extended range prediction horizon, were applied to a simulated industrial fish-meal drying process. The results showed that a correct rotary dryer representation can be obtained by using a GNM approach. Due to the representation capability of the GNM approach, excellent control performances of the NMPCs were observed when the process variables were subject to disturbances. As analyzed in this work, the fixed extended prediction horizon MPC surpassed recognized control methodologies (quadratic dynamic matrix control).