Artificial Neural Network Modeling for Thermophysical Behavior of Mxene Ionanofluids For Hybrid Solar Photovoltaic and Thermal Systems Applications

MXene materials proved to be an excellent contender for improving thermal systems' high energy and power density. MXene nanoparticles are synthesized and combined with ionic liquids to create an innovative fluid known as "MXene Ionanofluids," which has recently sparked an interest. Significantly few techniques have been developed for the properties analysis of these MXene nanoparticles. This paper aims to design a neural network model based on the Levenberg Marquardt algorithm, which can predict the thermophysical properties behavior of MXene Ionano fluids, including temperature and concentration, of nanofluids using experimental data. The experimental data were used to train the neural network as a learning dataset. A process of iteration was used to find the exemplary network architecture. The best architecture was chosen based on the possible combinations of hidden neurons. The neural network training, testing, and validation have all been completed successfully. The performance of the proposed LM-based back-propagation algorithm demonstrates that the error involved has been minimal and acceptable. The neural network could correctly predict the outcomes based on the numerical performance. ANN results have shown that the thermophysical properties behavior prediction error is 2%. A comparison study of ANN with RSM was performed to develop correction among input parameters and estimate the thermophysical properties behavior of MXene Ionanofluids. The comparison between experimental results and the proposed correlations revealed excellent practical compatibility