ANALISIS IKLIM MIKRO DI DALAM RUMAH TANAMAN UNTUK MEMPREDIKSI WAKTU PEMBUNGAAN DAN MATANG FISIOLOGIS TANAMAN TOMAT DENGAN MENGGUNAKAN METODE ARTIFICIAL NEURAL NETWORK

The aim of this research was to analyze the micro climate within a greenhouse in order to predict the flowering time and physiological maturity of tomato plant grown in the greenhouse. Heat unit was also used to determine the growing and development stage of tomato. Tomato was planted in the greenhouse for the period of August to December 2010 at Climatically Research Agency, Cimanggu, Bogor. The weather data collected using automatically weather station (aws) included air temperature, minimum and maximum temperature, relative humidity, and solar radiation intensity.The results of the research showed that heat unit of tomato during the growth period were 1692oC days. The average of air temperature and the relative humidity in the greenhouse was 27.1oC and 74.2%, respectively. The method of Artificial Neural Network (ANN) showed that coefficient correlations (R) of the flowering time and physiological maturity were 0.43655 and 0.48751, respectively, while the RMSE values were 5.2314 and 2.2539, respectively. Key words : Artificial neural network (ANN), greenhouse, and microclimate.