Central Institute of Agricultural Engineering, Bhopal, India
Dept. of Agricultural and Food Engineering, Indian Institute of Technology, Kharagpur, India.
Litchi is one of the most environmentally sensitive tropical fruits crop. It is popular export cultivar due to its attractive red color and high fruits quality. Pericarp browning, desiccations, loss of quality and post harvest decay have been identified as major problems restricting expansion of the industries in litchi exporting countries. Modified atmosphere packaging (MAP) has been considered to be beneficial to maintain high humidity essential for prevention of water loss and browning of litchi pericarp. Accurate measurement of respiration rate and its modeling is an important aspect to the success of design and operational features of techniques like modified atmosphere storage. The respiration data generated at temperatures 0-30 oC in the step of 5 for litchi using the closed system method was used for modeling respiration rate using neural network technique. Here O2, CO2, temperature and time are considered as independent parameters and respiration rate as the dependent parameters. To establish a specific relationship between these parameter using neural network modeling three layers are taken i.e. input layer, output layer and a hidden layer. In this method first of all the experimental values are coded in between -1 to +1. Then by use of mathematical formulations in MATLAB programming the out put response (respiration rate) with respect to O2 and CO2 was found out at any specific temperature. This respiration rate provides the basis for modeling of modified atmospheric packaging system. In the MATLAB program using neural network the respiration rate can be obtained by giving the storage period and concentration of O2 and CO2 at specified temperature. The relative deviation at different storage temperature was found out and it is in good agreement with that of experimental values.
Key words: Litchi, fruits, Respiration, temperature.