Vol.:(0123456789) 1 3 Journal of Thermal Analysis and Calorimetry (2020) 140:1811–1823 https://doi.org/10.1007/s10973-019-08915-0 Thermal decomposition of rice husk: a comprehensive artifcial intelligence predictive model Peter Adeniyi Alaba 1  · Segun I. Popoola 2  · Faisal Abnisal 3  · Ching Shya Lee 1,4,5  · Olayinka S. Ohunakin 6,7  · Emmanuel Adetiba 2,8  · Matthew Boladele Akanle 2  · Muhamad Fazly Abdul Patah 1  · Aderemi A. A. Atayero 2  · Wan Mohd Ashri Wan Daud 1 Received: 7 March 2019 / Accepted: 11 October 2019 / Published online: 2 November 2019 © Akadémiai Kiadó, Budapest, Hungary 2019 Abstract This study explored the predictive modelling of the pyrolysis of rice husk to determine the thermal degradation mechanism of rice husk. The study can ensure proper modelling and design of the system, towards optimising the industrial processes. The pyrolysis of rice husk was studied at 10, 15 and 20 °C min −1 heating rates in the presence of nitrogen using thermo- gravimetric analysis technique between room temperature and 800 °C. The thermal decomposition shows the presence of hemicellulose and some part of cellulose at 225–337 °C, the remaining cellulose and some part of lignin were degraded at 332–380 °C, and lignin was degraded completely at 480 °C. The predictive capability of artifcial neural network model was studied using diferent architecture by varying the number of hidden neurone node, learning algorithm, hidden and output layer transfer functions. The residual mass, initial degradation temperature and thermal degradation rate at the end of the experiment increased with an increase in the heating rate. Levenberg–Marquardt algorithm performed better than scaled conjugate gradient learning algorithm. This result shows that rice husk degradation is best described using nonlinear model rather than linear model. For hidden and output layer transfer functions, ‘log-sigmoid and tan-sigmoid’, and ‘tan-sigmoid and tan-sigmoid’ transfer functions showed remarkable results based on the coefcient of determination and root mean square error values. The accuracy of the results increases with an increasing number of hidden neurone. This result validates the suitability of an artifcial neural network model in predicting the devolatilisation behaviour of biomass. Keywords Rice husk · Thermal decomposition · Artifcial intelligence · Neural network · Pyrolysis · Heating rate Introduction The depletion of fossil fuels and accompanied environmental impact such as water and air pollution, global warming and acid rains have propelled energy and fuel diversifcation, inciting immense research eforts towards the development of renewable and sustainable alternative sources of energy. Biomass waste has been identifed as a more promising source of renewable energy among all other alternatives towards meeting the global demand [1, 2]. * Peter Adeniyi Alaba adeniyipee@live.com * Wan Mohd Ashri Wan Daud ashri@um.edu.my 1 Department of Chemical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia 2 Department of Electrical and Information Engineering, Covenant University, Ota, Ogun State, Nigeria 3 Department of Chemical Engineering, Faculty of Engineering, King Abdulaziz University, Rabigh 21911, Saudi Arabia 4 University of Malaya, 50603 Kuala Lumpur, Malaysia 5 UMR5503 Laboratoire de Génie Chimique (LGC), Toulouse, France 6 The Energy and Environment Research Group (TEERG), Mechanical Engineering Department, Covenant University, Ota, Ogun-State, Nigeria 7 Faculty of Engineering and the Built Environment, University of Johannesburg, Johannesburg, South Africa 8 HRA, Institute for Systems Science, Durban University of Technology, P.O. Box 1334, Durban, South Africa