Performance prediction of a hybrid microgeneration system using Adaptive Neuro-Fuzzy Inference System (ANFIS) technique

This study investigates the application of Adaptive Neuro-Fuzzy Inference System (ANFIS) technique to predict the performance of a hybrid microgeneration system. The hybrid system consists of an internal combustion engine (1kWe and 3.2kWth) integrated with a high efficiency condensing furnace (16.4kWth). Real life system performance data has been collected during a heating/shoulder season in a controlled field-trial at Canadian Centre for Housing Technologies for total of 26days. Four ANFIS models, were developed, trained and validated with the collected filed-trial performance data sets and applied to predicting the system operating temperatures. The MATLAB® ANFIS models were then interfaced with TRNSYS building and controller modules to establish a whole-system model to predict the hybrid microgeneration unit’s seasonal performance.