2nd Trabzon International Energy and Environment Symposium, Trabzon, Türkiye, 26 - 29 Temmuz 1998, ss.517-520
Dimension reduction for the data set by using input trained feedforward neural nets is presented in this study for dehumidifying coils. A similar method to nonlinear extension of principal component analysis, input trained neural networks are used to extract useful information in data set obtained experimentally for dehumidifying coils. The data set is represented by a few latent variables after training of IT nets.