Adaptive neuro-fuzzy inference system modeling to predict the performance of graphene nanoplatelets nanofluid-based direct absorption solar collector based on experimental study

Abstract

In this study, an artificial neural network (ANN) adaptive neuro-fuzzy inference system (ANFIS) model was developed to predict the performance of a low-temperature direct absorption solar collector (DASC) enhanced with nanofluids. The accuracy and the effect of different types of clustering methods such as Subtractive Clustering and Fuzzy c-means (FCM) clustering investigated. Various inlet temperatures, weight percentage of graphene nanoplatelets nanofluids based deionized water (0.0005, 0.001, 0.005 wt%) and flow rates (0.0075, 0.015, 0.0225 kgs−1) were tested. Five inputs used for the ANFIS model: the inlet temperature, rate of solar radiation, ambient temperature, mass flow rate and weight percentage of nanofluid. The ANN provides a single output: the efficiency of the collector. In total, 1536 data points utilized for modeling, including 1075 for the ANN training process and 461 for testing. Overall, the ANFIS model was very accurate at predicting the collector performance. This was indicated by the root mean square error (RSME), the mean absolute percentage error (MAPE), coefficient of determination (R2), and mean bias error (MBE), which calculated to be 0.008, 0.031, 0.999 and 7e-4 respectively. The ANFIS model agrees well with the experimental results.