I. Farkas

A neural network topology for modelling grain drying

This paper is concerned with modelling moisture distribution in agricultural fixed-bed dryers using a neural network (NN). Ten different NN topologies were studied for modelling and the most appropriate one was selected to use. Inlet and outlet air temperatures, absolute humidities and air flow were considered as the input variables to the layers of the drying bed. Some topologies include also grain temperature for better performance. Randomly varying time series data simulating inlet conditions were used for training the neural network. The data were taken from a physically-based simulation model instead of real measurements. The simulation of three scenarios corresponding to constant, slow and fast input dynamics were compared. Average and maximum deviations were used as performance measures to evaluate and compare the models. On the basis of the comparisons, the topology of the best model was identified. The results show that moisture distribution in the drying bed could be well modelled using a neural network.

Analytical and experimental study of a modular solar dryer

A plenty of recently established small Hungarian farms need new small scale agricultural equipment to expand the productivity and to fulfil the increasing quality control demand. As a response for this demand a modular solar dryer was developed and built which can help the quality saving with renewable energy. In this paper the features and the operation of the dryer is presented first. The effect of the solar air collector module for the physical properties of drying air was studied along with the calculation of efficiency, too. The mass flow of the drying air through the system is one of the most important factor concerning to the whole process, for that reason the velocity distribution is also presented. Finally some measuring results are presented for drying of apple.

Mathematical and physical foundations of drying theories

ABSTRACT In this paper a survey is given concerning to the stochastic modelling approaches in transport processes with a special emphasis on application possibilities for simultaneous heat and mass transfer in drying. First, the mostly used classical modelling methods for drying are discussed which lead to a linear parabolic type of PDE systems supposing constant (state-independent) conductivity coefficients. Powerful discretisation methods are shown for their solution. Basic principles of variational calculus are discussed then with an attention on direct methods. As a simple application a first-order approximation example is formed, and the solution of the system equation is presented. It is also shown, that the thermodynamical state-dependence of the conductivity coefficients has a crucial influence on the flow pattern of the coupled heat and mass transfer, which is particularly obvious in the cases, when the so-called percolative phase transitions take place. It effects a discrete change of the conductivity coefficients and their probabilities as well. An illustration is shown for percolative phase transition. Describing statistical properties of percolative

Modelling of Thin-Layer Drying Using Neural Network

Abstract This paper deals with a neural network application concerning to an agricultural fixed bed driers. Aim of the study is to set up a relationship between material moisture distribution and physical parameters of drying air, such as temperature and humidity. Input data was randomly changed, meanwhile output was generated by OCallaghans model based on the input. Number of layers in the fixed bed drier for the neural network was also determined using OCallaghans model.

Climate Control in a Solar Operated Laboratory Greenhouse

Abstract This work shows a study to build up a temperature control for a laboratory scale greenhouse which can be, anyhow, used also for a general case in connection with climate environmental problems. The parameters of the system were identified on the base of measurements inputting with standard signals. The identification shows that heat capacity dependence on ventilation can not be ignored. The constants of PID controller were determined taking into consideration the calculated and the measured parameters. A trial was made to use an empirical approach to the determination of PID controller, as well.

Developing of a Process Control Laboratory for Education in Agricultural Engineering

Abstract This paper describes the development of a process control laboratory at the Department of Physics and Process Control, Godollő University of Agricultural Sciences, Hungary to be used for educational purposes in agricultural engineering. The main intention of the development is to provide practical exercises for the MSc and PhD students studying on the course Process control and for the related other subjects. The laboratory set-up covers technological processes as hot water making, greenhouses, drying, etc. ADAM modules are used for data acquisition and control purposes. Recently, a simplified greenhouse and a solar collector model are introduced for making practice of modelling, simulation, parameter identification and control studies.

Data logging and monitoring tools used for simulation and modelling of a solar system

Abstract The topic detailed in this paper is a data acquisition and monitoring system. The current applications of this system are mainly related to solar energy. The acquisition system was based on ADAM 4000 series modules. These are one-chip microcontroller based modules having analogue and digital inputs and outputs. Choosing these devices the modularity and extendibility of the system are ensured. Furthermore additional electronic boards were developed to fit these modules and the applied sensors or in control cases the switching or controlling circuits. The sensors applied in the system are described along with their calibration methods. The relevant software tools and its options are also explained.

Modeling and Identification of Agricultural Driers

Abstract A block-oriented simulation system (BOSS) has been developed for the modeling and identification of different types of agricultural driers. The basic system consists of a meteorological data generator, a mass flow rate setting, a moisture generator, a heater, a mixer, a drier and some supplementary blocks. Experimental and analytical modeling techniques of drying processes were reviev/ed especially for grain and hay. A fixed-bed, a simple crossflow and a recycled crossflow drying arrangements were studied creating their block-oriented scheme.

Estimating moisture content in a fixed-bed grain dryer

Abstract This paper deals with a neural network application concerning to the determination of moisture distribution in an agricultural fixed-bed dryers. The aim of this study is to determine the influence on the different type of training and validation data used for neural network (NN) model. The validation of three kinds of data as constant, slow and fast were applied. The input data were generated an identified from a physically based model. It has been concluded that for satisfactory validation of an NN model different number of training data series should be linked together. Average deviation and maximum difference were used to estimate the influence of different training and validation input data.

Modelling of a Fixed Bed Grain Dryer Using Neural Network

Abstract This paper deals with a neural network application concerning to fixed bed grain dryer. Aim of the study is to set up a relationship between material moisture distribution and physical parameters of drying air, such as temperature and humidity. Five different neural network structures were studied on two different series input data containing inlet and outlet air temperatures and humidities and air flow. Randomly changed input data was used for training the neural network. The data were taken from a physically based model instead of real measurements. The result show that moisture content of the drying bed can be calculated from air parameters using neural network.