BioEnergy Research, 2014, Volume 7, Issue 2, pp 681-692.
S. B. Ghugare, S. Tiwary, V. Elangovan, S. S. Tambe.
Artificial Intelligence Systems Group, Chemical Engineering & Process Development Division, National Chemical Laboratory, Dr. Homi Bhabha Road, Pashan, Pune, 411 008, India and
Chemical Engg. Department, National Institute of Technology (NIT), Tiruchirapalli, India.
The higher heating value (HHV) is an important property defining the energy content of biomass fuels. A number of proximate and/or ultimate analysis based predominantly linear correlations have been proposed for predicting the HHV of biomass fuels. A scrutiny of the relationships between the constituents of the proximate and ultimate analyses and the corresponding HHVs suggests that all relationships are not linear and thus nonlinear models may be more appropriate. Accordingly, a novel artificial intelligence (AI) formalism, namely genetic programming (GP) has been employed for the first time for developing two biomass HHV prediction models, respectively using the constituents of the proximate and ultimate analyses as the model inputs. The prediction and generalization performance of these models was compared rigorously with the corresponding multilayer perceptron (MLP) neural network based as also currently available high-performing linear and nonlinear HHV models. This comparison reveals that the HHV prediction performance of the GP and MLP models is consistently better than that of their existing linear and/or nonlinear counterparts. Specifically, the GP- and MLP-based models exhibit an excellent overall prediction accuracy and generalization performance with high (>0.95) magnitudes of the coefficient of correlation and low (<4.5 %) magnitudes of mean absolute percentage error in respect of the experimental and model-predicted HHVs. It is also found that the proximate analysis-based GP model has outperformed all the existing high-performing linear biomass HHV prediction models. In the case of ultimate analysis-based HHV models, the MLP model has exhibited best prediction accuracy and generalization performance when compared with the existing linear and nonlinear models. The AI-based models introduced in this paper due to their excellent performance have the potential to replace the existing biomass HHV prediction models.
Solid biomass fuels (biofuels) form an important renewable energy source. The experimental estimation of the energy content of a biofuel in terms of higher heating value (HHV) is a tedious and time-consuming laboratory procedure. Thus, proximate and ultimate analysis based models have been developed for predicting HHV of solid biofuels (Figure 1a). The basic assumption underlying most of these models that there exist linear relationships between elements of the proximate/ultimate analyses and HHV, is not unambiguously supported by the corresponding experimental data (Figure 1b). Accordingly, this study has developed proximate/ultimate analysis based nonlinear models for HHV prediction using two artificial intelligence (AI) methods, namely genetic programming (GP) and multi-layer perceptron (MLP) neural networks (Figure 1c).
The novelty of this study is that GP-based models have been employed for the first time to predict HHV of biofuels. A special characteristic of GP is that it can search and optimize an appropriate linear or a nonlinear model that best fits a given set of data. For proximate as also ultimate analysis based models, GP has searched nonlinear forms for predicting HHV, thus indicating that these type of models are more suited than the earlier predominantly linear models. The GP and MLP-based nonlinear models developed in this study possess much better prediction and generalization performances when compared with all the earlier models. Also, a large and varied database consisting of proximate and ultimate analyses of 382 and 536 biofuels, respectively, has been used to develop the AI-based models thus giving them broader applicability. The HHV prediction models introduced in this paper, due to their excellent performance, possess a potential of replacing the existing models.
Figure 1: (a) status of existing models for the prediction of HHVs of solid biofuels, (b) nature of dependencies between some constituents of the proximate/ultimate analysis and HHV of solid biofuels, (c) nature of artificial intelligence based models developed for the prediction of biofuel HHVs.