The comparison of neural network and hybrid neuro-fuzzy based inferential sensor models for space heating systems

Abstract

Inferential sensors are used to infer the critical control variables that are otherwise difficult, if not impossible, to measure in broad range of engineering fields. All inferential sensors are based on an inferential modelling module that represents the dynamics between the inputs and the outputs. Two commonly used artificial intelligence based approaches for the development of the inferential modelling modules are: (1) Neural Networks and (2) Adaptive Neuro-Fuzzy Inference Systems. This paper is presenting the estimation of average air temperature in the built environment by using Integer Neural Network and Adaptive Neuro-Fuzzy Inference System based inferential sensor models. By comparing the results of these models with one another, advantages and disadvantages of each are discussed.