Evolving connectionist approaches to compute thermal conductivity of TiO2/water nanofluid

Conventional working fluids which are used in the heat transfer mediums have restricted the ability of heat removal. In this investigation, thermal performance of TiO2 nanoparticles immersed in DI (de-ionized) water was evaluated. Introducing a combination of experimental and modeling approaches to forecast the amount of thermal conductivity using four different neural networks can be mentioned as the predominant aim of this investigation. Between MLP-ANN, ANFIS, LSSVM, and RBF-ANN Methods, the LSSVM produced better results with the lowest deviation factor and reflected the most accurate responses between the proposed models. The regression diagram of experimental and estimated values shows an R2 value of 0.9806 for training sets and 0.9579 for testing sections of the ANFIS method in part a, and in the b, c and d parts of the diagram, coefficients of determination were 0.9893 & 0.9967 and 0.9974 & 0.9992 and 0.9996 & 0.9989 for train and test stages of MLP-ANN, RBF-ANN and LSSVM models, respectively. Also, the effects of different parameters were investigated using a sensitivity analysis method which demonstrates that the temperature is the most affecting parameter on the thermal conductivity with a relevancy factor of 0.66866.