Dr. Hongwei Wu

School of Physics, Engineering and Computer Science;

University of Hertfordshire

Country United Kingdom

Short biography and link to the website or Research Gate.

Dr Hongwei Wu is a Senior Lecturer in the School of Engineering and Computer Science at University of Hertfordshire. He is a Chartered Engineering (CEng), Member of IMechE and Fellow of the Higher Education Academy. Dr Wu obtained his PhD in 2004 at the Beihang University where he conducted research on thermofluids. He has an academic/research profile previously developed at Northumbria University, University of the West of Scotland, Birmingham University and Brunel University London in the UK and at the University of British Columbia & University of Alberta in Canada. His research has focused on energy system and energy recovery, two phase and multiphase flow, thermodynamic analysis, fluid mechanics and heat transfer, advanced cooling technology, computational fluid dynamics (CFD). He has published over 120 papers with more than 85 in peer-reviewedjournals. He serves as an Editor/Editorial Board Member of several International Journals. He also serves as general Chair and session Chairs/co-Chairs, TPC members at a number of International Conferences.

Presentation title:

A data driven deep neural network model for predicting boiling heat transferin helical coils under high gravity

Abstract：

A deep artificial neural network (ANN) model has been proposed to predict the boiling heat transfer in helical coils under high gravity conditions and compare with actual experimental data. A test rig is set up to provide the high gravity up to 11 g with the heat flux can be up to 15100 W/m² and the mass velocity range from 40 to 2000 kg m^-2 s^-1. In the current work, total 531 data samples have been used in the present ANN model. The proposed model was developed in Python Keras environment with Feed-forward Back-propagation (FFBP) Multi-layer Perceptron (MLP) using eight features (mass flow rate, thermal power, inlet temperature, inlet pressure, direction, acceleration, tube inner surface area, helical coil diameter) as the inputs and two features (wall temperature, heat transfer coefficient) as the outputs. The deep ANN model composed of three hidden layers with a total number of 1098 neurons and 300,266 trainable parameters has been found as optimal according to statistical error analysis. Performance evaluation is conducted based on six verification statistic metrics (, MSE, MAE, MAPE, RMSE and cosine proximity) between the experimental data and predicted values. The obtained results demonstrate that 8-512-512-64-2 neural network model has the best performance in predicting the helical coil characteristics with (R²=0.853, MSE=0.018, MAE=0.074, MAPE=1.110, RMSE=0.136 , cosine proximity=1.000) in testing stage. It is indicated that with the utilisation of deep learning, the proposed model is able to successfully predict the heat transfer performance in helical coils, especially achieved excellent performance in predicting the outputs having very large range of value differences.

Selected /recent 2 or 3 Journal publications.

1. A data driven deep neural network model for predicting boiling heat transfer in helical coils under high gravity, International Journal of Heat and Mass Transfer, Accepted on 21^st Nov. 2020.

2. A review of boiling heat transfer characteristics in binary mixtures, International Journal of Heat and Mass Transfer, 164:120570, 2021.

3. Experimental and analytical study of dual compensation chamber loop heat pipe under acceleration force assisted condition, International Journal of Heat and Mass Transfer, 153:119615, 2020.