PREDICTION OF PSYCHROMETRIC PARAMETERS USING NEURAL NETWORKS

Abstract

A neural network approach was used for the prediction of the psychrometric parameters in a non-iterative manner. Neural network models were developed for the each of the three main variables - dry bulb temperature, wet-bulb temperature and relative humidity - as a function of the other two variables. Models were also developed for the prediction of the dew point temperature using the dry bulb and wet bulb temperatures, and for the saturation vapor pressure as a function of the dry bulb temperature. The sensitivity of the neural network performance to the form of input and output variables employed was also investigated. The prediction accuracy of the neural network models were found to be very good, with errors less than 4%.

Keywords:

• Psychrometry
• Drying
• Relative humidity
• Wetbylb temperature
• dry bulb temperature
• Artificial Intelligence
• humidification
• cooling towers

Additional information

Notes on contributors

S. Sreekanth

Post doctoral fellow, Dept. of Biosystems Engineering

H.S. Ramaswamy

Professor, Dept. of Food Science

S. Sablani

Research associate, Agricultural and Agri-food Canada