The aim of this study is to demonstrate the comparison of an artificial neural network (ANN) and an adaptive neuro-fuzzy inference system (ANFIS) for the prediction performance of a vertical ground source heat pump (VGSHP) system. The VGSHP system using R-22 as refrigerant has a three single U-tube ground heat exchanger (GHE) made of polyethylene pipe with a 40 mm outside diameter. The GHEs were placed in a vertical boreholes (VBs) with 30 (VB1), 60 (VB2) and 90 (VB3) m depths and 150 mm diameters. The monthly mean values of COP for VB1, VB2 and VB3 are obtained to be 3.37/1.93, 3.85/2.37, and 4.33/3.03, respectively, in cooling/heating seasons. Experimental performances were performed to verify the results from the ANN and ANFIS approaches. ANN model, Multi-layered Perceptron/Back-propagation with three different learning algorithms (the Levenberg-Marquardt (LM), Scaled Conjugate Gradient (SCG) and Pola- Ribiere Conjugate Gradient (CGP) algorithms and the ANFIS model were developed using the same input variables. Finally, the statistical values are given in as tables. This paper shows the appropriateness of ANFIS for the quantitative modeling of GSHP systems. © 2010 Elsevier Ltd. All rights reserved.
Mendeley saves you time finding and organizing research
Choose a citation style from the tabs below