| Infrared drying for fruits and vegetables is a complex unsteady state process with simultaneous heat and mass transferment, which not only affected by the drying conditions, but also having great differences with the types of material, internal structure, physical and chemical properties and the external shapes. As the infrared drying process affected by many factors, in this paper, the mechanism of infrared drying for fruits and vegetables was made a thorough study, in order to accurately predicte and controll the infrared drying process of fruits and vegetables.Firstly, the characteristics of infrared drying of apple slices and its drying quality were further studyed. The factors influencing infrared radiation drying rates, such as radiation intensity, radiation distance, material temperature, material thickness, the types of material and status were analyzed. The optimum drying parameters were obtained by orthogonal experiments. The optimum heating program and drying cycles were obtained for the design of new process and the improvement of the traditional equipment, which can both save energy and ensure its drying quality.Secondly, the network model structure between moisture content and all the controlling factors was built based on feed-forward neural network, the selected structure of the applied neural network, with its five inputs, single output and 11 hidden neurons were used. All data series obtained from different drying runs were used for training and test, mathematical model responding to inner relationship of the experimental data was obtained by finite iteration calculation, and it was trained and simulated systemically by using Matlab neural network toolbox. It was conclusion that the model could be built by the BP neural network, cost-effectively, accurately and rapidly during far infrared drying of apple slices within the trial stretch, it was found that the predictions of the artificial neural network model fitting the experimental data preferably, and the applications of the artificial neural networks can be used for the online state estimation moisture content more suitably and accurately.Thirdly, the traditional drying models were fitted to experimental data and the drying coefficients were determined by means of nonlinear least square, based on the Matlab software using the gauss-newton algorithm. The experimental data from infrared drying of apple slices were used as measured samples, and the tests were performed with the materials temperature of 60℃, the radiation power of 750W, the radiation distances of 100mm and materials thickness of 5mm. Fifteen different mathematical drying models were compared by some evaluation targets such as coefficient of determination (R), square sum of error (SSE) and root mean square error (RMSE) to estimate the drying models, and compared with the mathematical model of infrared drying of fruits and vegetables was built based on BP neural network. It may be concluded that the Modified Page equation-Ⅱmodel could sufficiently predict and control the infrared drying process for fruits and vegetables.fourthly, a test device with bench was a on-line measurment system was designed and for the studying of infrared drying for fruits and vegetables. The hardware parts of this system was composed of the electric-resistance strain gauge mass transducer and AD590 temperature transducer, physical signals of mass and temperature collected were converted to faint electric signals, which were amplified with analogue circuits, the analogue signals were converted into digital signals by PCI8310 data acquisition board. and digital signals were sent and stored to the PC. The software parts were designed using VB, mass and temperature corresponding to voltage values were obtained by function transformation based on VB procedures, generating drying curves, radiation temperature and material temperature change curves, so the characteristics of infrared drying for fruits and vegetables were diaplayed accurately and timely.
|
PDF Full Text Request