Shiva Gorjian - Agricultural Machinery Engineering Department, Faculty of agriculture, TarbiatModares University, Tehran, Iran
teymour tavakkoli hashjin - Agricultural Machinery Engineering Department, Faculty of agriculture, TarbiatModares University, Tehran, Iran
barat ghobadian - Agricultural Machinery Engineering Department, Faculty of agriculture, TarbiatModares University, Tehran, Iran
ahmad banakar - Agricultural Machinery Engineering Department, Faculty of agriculture, TarbiatModares University, Tehran, Iran

A study has been performed to predict distillate production of a point-focus parabolic solar still (PPSS) was operated for seven sunny, relative cloudy and dusty days in October. The aim of this study is to determine the effectiveness of modeling solar still distillate production using artificial neural networks (ANNs) and local weather data. A mathematical model is also presented to predict the thermal losses, and hourly productivity of the PPSS based on energy balance and heat transfer equations. The study used the environmental and operational variables affecting solar still performance, which are the hourly beam solar insolation, hourly air temperature, hourly wind velocity and wind incidence angle. The objectives of the study are to assess the sensitivity of the ANN predictions to different combinations of input parameters as well as to determine the minimum amount of inputs necessary to accurately model the solar still performance. The results showed that the ANN-model gave the best estimation with the accuracy of more than 99%. By using the correlation coefficient (R), it was found that 93-97% of the variance was accounted for by the ANN model. Satisfactory results for the PPSS suggest that, with sufficient input data, the ANN method could be extended to predict the performance of other solar still designs in different climate regimes