PENGEMBANGAN MODEL JARINGAN SYARAF TIRUAN UNTUK PENDUGAAN SUHU UDARA DI DALAM RUMAH KACA

Herry Suhardiyanto (Departemen Teknik Pertanian, Fakultas Teknologi Pertanian, Intitut Pertanian Bogor, Kampus IPB Darmaga Bogor 16680)
Suroso Suroso (Departemen Teknik Pertanian, Fakultas Teknologi Pertanian, Intitut Pertanian Bogor, Kampus IPB Darmaga Bogor 16680)
Titin Nuryawati (Alumnus Departemen Teknik Pertanian, Fakultas Teknologi Pertanian, Intitut Pertanian Bogor, Kampus IPB Darmaga Bogor 16680)

Jurnal Keteknikan Pertanian
Vol. 21 No. 1 (2007): Jurnal Keteknikan Pertanian

ABSTRACT During the daytime in tropical region, air temperature inside the greenhouse higher than the outside air temperature. The prediction of air ternpereture inside the greenhouse can be done by using Artificial Neural Network (ANN) model. The neural network model consist of three layers, there are input layer, hidden layer and output layer. The input layer consist of eight nodes, there are wind velocity, air humidity, air pressure, outside air temperature, daily rainfall, solar radiation, roof temperature and floor temperature. The output layer is inside air temperature of the greenhouse. The ANN models were developed with different proportion of training and validation data. Validation of the model had been done by using standard error prediction, bias and Coefficient of Variation. It had been shown that the ANN model could explain the complicated relationship among greenhouse parameter, effectively. Diterima: 27 Pebruari 2007; Disetujui: 9 Maret 2007.