David Samek

Elman Neural Networks In Model Predictive Control.

The goal of this paper is to present interesting way how to model and predict nonlinear systems using recurrent neural network. This type of artificial neural networks is underestimated and marginalized. Nevertheless, it offers superior modelling features at reasonable computational costs. This contribution is focused on Elman Neural Network, two-layered recurrent neural network. The abilities of this network are presented in the nonlinear system control. The task of the controller is to control the liquid level in the second of two interconnected cylindrical tanks. The mathematical model of the realtime system was derived in order to test predictor and consequently the controller in Matlab/Simulink simulations. INTRODUCTION Model predictive control (MPC) (Camacho and Bordons 2007) is a very popular concept for the development and tuning of nonlinear controllers in the presence of input, output or state constraints. Many predictive control techniques based on MPC that use artificial neural network (ANN) as a predictor are established on multilayer feed-forward neural networks (Hagan et al. 2002), (Kanjilal 1995). In spite the multilayer feedforward neural networks (MFFNNs) have many advantages such as simple design and scalability, they have also many drawbacks such as long training times and choice of an appropriate learning stop time (the over-learning versus the early stopping problem). However, there is quite a number of types ANNs suitable for the modelling and prediction, for instance adaptive linear networks, radial basis function networks and recurrent networks (Liu 2001), (Meszaros et al. 1999), (Koker 2006). This paper is divided as follows: After short introduction to the recurrent neural networks, the used model predictive controller is explained. Then the model of the real time system is derived. After that the identification of the predictor (training of the artificial neural network) is described. When the identification is finished, the paper focuses on the model predictive control and evaluation of results. The contribution is finished by some concluding remarks. RECURRENT NEURAL NETWORKS Recurrent neural networks (sometimes are these networks called feedback neural networks) can be distinguished from feed-forward neural networks in that they have a loopback connection (Figure 1). In its most general form recurrent network consist of a set of processing units, while the output of each unit is fed as input to all other units including the same unit. With each link connecting any two units, a weight is associated which determines the amount of output a unit feeds as input to the other unit (Yegnanarayana 2005). Figure 1: Example of Recurrent Neural Network Recurrent neural networks have superior temporal and spatial behaviours, such as stable and unstable fixed points and limit cycles, and chaotic behaviours. These behaviours can be utilized to model certain cognitive functions, such as associative memory, unsupervised learning, self-organizing maps, and temporal reasoning (He 1999). Elman Neural Networks One of the most known recurrent neural networks is Elman neural network (Elman 1990). Typical Elman network has one hidden layer with delayed feedback. The Elman neural network is capable of providing the standard state-space representation for dynamic systems. This is the reason why this network architecture is utilized as a recurrent neural equalizer. Proceedings 23rd European Conference on Modelling and Simulation ©ECMS Javier Otamendi, Andrzej Bargiela, Jose Luis Montes, Luis Miguel Doncel Pedrera (Editors) ISBN: 978-0-9553018-8-9 / ISBN: 978-0-9553018-9-6 (CD) Generally, this network is considered as a special kind feed-forward network, including additional memory neurons and local feedback (Koker 2006). Typical structure of Elman neural network is depicted in fig. 2. Figure 2: Elman Neural Network MODEL PREDICTIVE CONTROL There are various approaches to predictive control by artificial neural networks. Generally we can say that these methods use ANN as the plant model in order to get its output predictions. The most used approach is model predictive control (Camacho and Bordons 1995). MPC is a broad control strategy applicable to both linear and nonlinear processes. The main idea of MPC algorithms is to use a dynamical model of process to predict the effect of future control actions on the output of the process. Hence, the controller calculates the control input that will optimize the performance criterion over a specified future time horizon:

Semi-batch reactor predictive control using artificial neural network

This paper deals with model predictive control (MPC) of chemical exothermic semi-batch reactor. A first order chemical reaction is considered to be running in the reactor. The reaction is strongly exothermic so the in-reactor temperature is rising very fast due to reaction component dosing. Thus, the temperature control is necessary. The simulation model of the plant was developed in the MATLAB/Simulink. The system is nonlinear because of chemical reaction kinetics, so its control is difficult by classical methods. The classical MPC objective function was modified in order to improve the control. The MPC controller uses an artificial neural network as a predictor.

Investigation of Surface Roughness while Ball Milling Process

Increasing urge to raise production rate and production quality in the industry brings new requests and challenges. One of them is demand for accuracy and precision of produced parts. Especially in the CNC machining, where the expectations are high, the companies face new issues. Therefore, it is very important to recognize, understand and cope with the technological factors influencing the production accuracy and surface quality of CNC machined parts.

Artificial Neural Networks In Prediction And Predictive Control

In this contribution the three various artificial neural networks are tested on CATS prediction benchmark. The results are compared and evaluated. Furthermore, these artificial neural networks are tested in model predictive control on the t-variant system. The aim of this paper is to present and compare artificial neural networks as interesting way how to model and predict nonlinear systems even with t-variant parameters. The key features of this paper are emphasis of the computational costs of the selected predictors and usage of adaptive linear network which offers short learning times and remarkable prediction error. INTRODUCTION The increasing demand on the quality, reliability, and economical profits leads to using of new modeling and control methods in the process industry. In past few decades the predictive control techniques have become very popular. One of the most used approaches is the Model Predictive Control (MPC) method (Camacho and Bordons 1995). The appropriate predictive model is a key question in nonlinear model predictive control. The predictive models can be divided into two main groups (Verdunen and Jong 2003): white box models and black box models. The white box modeling is established on a prior knowledge of mathematic description of basic physical rules of controlled process. White box models are excellent for process modeling and product development. The model constants have a physical meaning and are not dependent on process design. The main disadvantage of white box models is the time of development and higher complexity. Conversely, black box models such as artificial neural network (ANN) and fuzzy logic models are data-driven. They provide general method for process dynamics description from input-output data. First and foremost, the learning ability makes artificial neural networks versatile, user friendly and powerful tool for many practical applications (Hussain 1999). Many predictive control techniques based on MPC, which use artificial neural network as a predictor, are established on multilayer feed-forward neural networks (Hagan et al. 2002; Kanjilal 1995). In spite of the fact that the multilayer feed-forward neural networks (MFFNNs) have many advantages, such as simple design and scalability, they have also many drawbacks, such as long training times and choice of an appropriate learning stop time (the over-learning versus the early stopping). Nevertheless, there are quite a number of ANN types suitable for the modeling and prediction (Liu 2001; Meszaros et al. 1999; Chu et al. 2003). Moreover, features of these ANNs exceed abilities of the MFFNN in many cases. One of these ANNs is ADALINE (ADAptive LInear NEuron). What is more, ADALINE has one special feature – adaptivity. Owing to its simple structure it offers interesting way how to design adaptive neural predictor with reasonable computational demands. This paper is organized as follows: In the beginning multilayer feed-forward neural networks and adaptive linear networks are briefly introduced. Then the methodology of the simulations is explained, after that the results are presented and the paper is concluded by final remarks. MULTILAYER FEED-FORWARD NEURAL NETWORKS Multilayer feed-forward neural networks were derived by generalization from Rosenblatt’s perceptron, thus they are often called multilayer perceptrons (MLP). This type of artificial neural networks uses supervised training. One of the most known methods of supervised training is backpropagation algorithm; hence these ANNs are sometimes also called backpropagation networks. In the MFFNN the signals flow between the neurons only in the forward direction i.e. towards the output. Neurons in MFFNN are organized in layers and neurons of the certain layer can have inputs from any neurons of the earlier layer. The ability to predict of ANN is determined by capability of modeling of certain process. By applying the Kolmogorov theorem it was proved that for general function approximation is sufficient twolayer MFFNN (one hidden layer) if non-polynomial transfer functions are used and the hidden layer has enough neurons (Leshno et al. 1993). Proceedings 22nd European Conference on Modelling and Simulation ©ECMS Loucas S. Louca, Yiorgos Chrysanthou, Zuzana Oplatkova, Khalid Al-Begain (Editors) ISBN: 978-0-9553018-5-8 / ISBN: 978-0-9553018-6-5 (CD) Figure 1: Simplified Scheme of Two-layer MFFNN The two-layer MFFNN, which contains one output layer and one hidden layer, is depicted in the figure 1 (this structure is implemented in this paper). This MFFNN can be described by two equations:

Neural Networks in Modeling of CNC Milling of Moderate Slope Surfaces

Computer numerical control (CNC) allows achieving a high degree of automation of machine tools by pre-programmed numerical commands. CNC milling process is widely used in industry for machining of complex parts. The need of a description of the CNC milling process is necessary for production of precise parts. This paper introduces artificial neural network based modeling, while the CNC milling of moderate slope shapes is studied. The developed neural models consist of two inputs and two outputs. The created neural models were experimentally tested on the real data. Then, the evaluation and comparison of all models were performed.

Materials, construction and manufacture of the vehicle for inspection of piping systems

The article focuses on the design, construction and manufacture of an inspection vehicle intended to access difficult-to-reach places. The vehicle is able to monitor piping system failures at the view angle of 180° in deep depths and adverse environments. The individual components of the inspection vehicle, more detailed their programming and production on CNC machines are discussed. The vehicle is supposed to easily run into the piping systems and can be safely pulled out. A control system was created for motion and video signal transmission to the operator from above the ground. Aluminum alloy (EN AW 2024) is the predominant material of the manufactured components and at the same time has ideal processing and operating properties. Proposed inspection vehicle is a robust and functional solution with minimal maintenance.

Self-Organizing Migrating Algorithm Used for Model Predictive Control of Semi-batch Chemical Reactor

The current availability of powerful computing technologies enables using of complex computational methods. One of such complex method is also the self-organizing migrating algorithm (SOMA). This algorithm can be used for solving of various optimization problems. It may be used even for such complex task, as the non-linear process control is. In this paper, the capability of using SOMA algorithm for the model predictive control (MPC) of semi-batch chemical reactor is studied. The MPC controller including self-organizing migrating algorithm (SOMA) is used for the optimization of the control sequence. The reactor itself is used in chromium recycling process in leather industry.

Modeling and Model Predictive Control of Semi-Batch Chemical Reactor

The paper presents a control mechanism design for a semi-batch chemical reactor. The data obtained by chemical engineering analysis of real experiments are used to simulate the semi‑batch process. A mathematical model based on the real reactor geometry and size is used to simulate the whole process. The process simulations are implemented in MATLAB / Simulink environment and suitable PID and Model Predictive Control are also proposed. Because of that the chemical reactor is a complex and nonlinear system, the PID controller has to use an online identification to be able to deal with nonlinearities. Results obtained by simulations are compared and discussed.

An Approach for Residual Stress Determination by Mechanical Indentation Technique

Abstract Residual stresses lower the utility value of plastic parts. Determination of the induced stresses can help deal with them. Measurements are time-consuming and expensive. A new approach to measuring residual stresses, such as indentation measurement, can lead to the simple determination of residual stresses. The paper shows the relationship between the condition of injection moulding, the subsequent residual stress, and hardness through thickness. The computer model displays the field and magnitude of residual stress in the samples. The model results are then compared to measured parameters after indentation and the magnitude of residual stress determined by the standard hole drilling method.

Two adaptive approaches of nonlinear system control

Generally the artificial neural networks (ANN) are regarded as highly computational demanding method. The usage of ANN in model predictive control as an adaptive predictor is mostly impossible. The aim of this paper is to present and compare one possible way how to reduce computational costs of adaptive predictors based on artificial neural networks. This paper presents real-time system control by two adaptive control methods. The first method is based on the model predictive method with adaptive artificial neural network as a predictor. This artificial neural network offers interesting solution of the computation time problem while using artificial neural network as an adaptive (online) predictor. The second method is established on self-tuning approach. Both these methods are applied to a problem of control liquid level in interconnected tanks. Real-time experiments are performed using Amira DTS200 - three tank system. This system is characterized by non-linear behavior.