Assessment of Fouling in Plate Heat Exchangers with Machine Learning Algorithms

Fouling is the accumulation of undesired particles on heat transfer surfaces which affects the heat transfer performance of a heat exchanger negatively. The accumulation of these particles prevents heat from being transformed through the heat exchangers by generating a fouling layer-like insulation. The main aim of the thesis is to investigate the machine learning algorithms to classify and predict the fouling status of PHE used in combi-boilers, to generate the background of the predictive maintenance, besides investigating the fouling effect on PHEs in terms of heat transfer and energy consumption by using a 1-D model.