Novel Approaches to Modelling Flammability Characteristics of Polymethyl Methacrylate (PMMA) via Multivariate Adaptive Regression Splines and Random Forest Methods

Abstract

Soft-computing techniques for fire safety parameter predictions in flammability studies are essential for describing a material fire behaviour. This study proposed, two novel Artificial Intelligence developed models, Multivariate Adaptive Regression Splines (MARS) and Random Forest (RF) methods, to model and predict peak heat release rate (pHRR) of Polymethyl methacrylate (PMMA) from Microscale Combustion Calorimetry (MCC) experiment. From the statistical analysis, MARS presented the highest coefficient of determination (R 2 ) values of (0.9998) and (0.9996) for training and testing respectively, with low MAD, MAPE and RMSE values. Comparatively, MARS Alhassan et al.; AJRCOS, 4(4): 1-14, 2019; Article no.AJRCOS.53584 2 outperformed RF in the predictions of pHRR, through its model algorithms that generated optimized equations for pHRR predictions, covering all non-linearity points of the experimental data. Amongst the input variables (sample mass, THR, HRC, pTemp and pTime), heating rate (β), highly influenced pHRR outcome predictions from MARS and RF models. However, to validate the performance and applicability of the proposed models. Results of MARS and RF were benchmarked with that from Artificial Neural Network (ANN) methods. The MARS and RF models observed the least error deviation when compared with pHRR results for PMMA from the ANN models. This study therefore, recommends the adoption of MARS and RF in the predictions of flammability characteristics of polymeric materials. Original Research Article

Wood Dust Flammability Analysis by Microscale Combustion Calorimetry

“…The LOI values of wood are generally between 23–26. Researchers proposed empirical formulas between LOI and HRC, LOI = 12% + 4000/HRC, LOI = 125/HRC [ 7 , 9 ]. However, the LOI value and HRC value do not fit such relations.…”

“…It is simple and wieldy with excellent repeatability and reproducibility as well as a long-life span. The range of sample sizes required is 0.5–50 mg with a sample heating rate of 0–10 K s −1 [ 7 , 8 ]. The MCC experiment uniquely produces separate burning stages, i.e., controlled pyrolysis of the solid material and combustion of the volatile gases.…”

“…According to the design and standards of MCC [ 7 , 8 , 9 ], there are two standard measurement methods, Method A [ 17 ] and Method B [ 18 ]. The Method A procedure, also referred to as controlled thermal decomposition, involves the pyrolysis of the sample in inert gas (nitrogen, helium, etc.)…”