Search

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 thermogravimetric 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 artificial neural network model was studied using different 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 coefficient 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 artificial neural network model in predicting the devolatilisation behaviour of biomass.

This paper presents an experimental investigation of energy consumption and heat transfer performance characteristics of a safe mass–charge of liquefied petroleum gas refrigerant, enhanced with varying concentrations of TiO2 nano-lubricants (i.e. 0.2 gL−1, 0.4 gL−1 and 0.6 gL−1) in a domestic refrigerator. Performance parameters investigated at steady state included: instantaneous and mean power consumption, cooling capacity, coefficient of performance (COP), discharge thermal conductivity and discharge temperature. Analysis was based on temperature and pressure readings obtained from appropriate gauges attached to the test rig. Refrigerant properties were obtained from Ref-Prop NIST 9.0 software. Findings showed that reductions in mean power consumption were observed to be 14, 9 and 8% at 0.2 gL−1, 0.4 gL−1 and 0.6 gL−1 nano-lubricants respectively; the highest mean power consumption was obtained using pure compressor mineral oil while the lowest was with 0.2 gL−1 TiO2 nano-lubricant. The estimated mean cooling capacities for the various compressor lubricants were found to be higher with 0.4 gL−1 and 0.6 gL−1 nano-lubricants than pure compressor lubricant, and lower with 0.2 gL−1 nano-lubricant when compared with pure mineral oil lubricant. All the TiO2-based nano-lubricants were of higher instantaneous and mean COP values than the pure lubricant. All nano-lubricant mixtures were also found to give lower discharge temperatures than the pure lubricant. In conclusion, selected TiO2-based nano-lubricants improved the efficiency of the domestic refrigeration system considerably.

In a typical outdoor environment, the propagation of radio waves is usually random in nature, to the extent that the characterization of the wireless channel often becomes very difficult. Several models have been developed to predict the average Received Signal Strength (RSS) for specified distance ranges. However, the use of deterministic models requires high computational efficiency while the prediction results of empirical models may not be as accurate as required. On machine learning approach, the performances of multi-layered feed-forward network models are limited by slow convergence and local minimum, such that a global optimal solution is not guaranteed. In this paper, Extreme Learning Machine (ELM) algorithm is considered in the development of an optimal path loss prediction model for outdoor propagation scenario. Single Hidden Layer Feed-forward Neural Networks (SHLFNNs) are trained and tested with the path loss data that were computed based on the RSS data of a commercial 1800 MHz base station located along Lagos-Badagry highway in Nigeria. The training speed, learning effectiveness, and the generalization ability of Artificial Neural Network Back-Propagation (ANN-BP) and ELM algorithms are analysed. Experimental results show that ELM models are 140 times faster to train than the ANN-BP models. On prediction accuracy, the outputs of ELM, ANN-BP, Okumura–Hata, and COST-231 models have Root Mean Squared Error (RMSE) values of 2.896, 2.449, 7.456, and 6.116 dB respectively; and regression coefficient (R) values of 0.959, 0.973, 0.935, and 0.935 respectively, when compared to the target variable of the training dataset. When the models were tested with new input data that were excluded from the training process, RMSE values of 4.250, 6.622, 8.732, and 7.087 respectively; and R values of 0.893, 0.876, 0.904, and 0.904 respectively are obtained. In conclusion, the findings of this study confirm that ELM algorithm guarantees an optimal path loss model with fast training convergence, high prediction accuracy, and good generalization ability for radio network planning and optimization in outdoor environments.

Unfortunately, in the original publication of the article the third author name was misspelled as Faisal Abnisal.