A Generalized Regression Neural Network Model for Predicting the Curing Characteristics of Carbon Black-Filled Rubber Blends

Abstract

In this study, a new generalized regression neural network model for predicting the curing characteristics of rubber blends with different contents of carbon black filler cured at various temperatures is proposed for the first time The carbon black contents in the rubber blend and cure temperature were used as input parameters, while the minimum and maximum elastic torque, scorch time, and optimal cure time, obtained from the analysis of 11 rheological cure curves registered at 10 various temperatures, were considered as output parameters of the model. A special pre-processing procedure of the experimental input and target data and the training algorithm is described. Less than 55% of the experimental data were used to significantly reduce the total number of input and target data points needed for training the model. Satisfactory agreement between the predicted and experimental data, with a maximum error in the prediction not exceeding 5%, was found. It is concluded that the generalized regression neural network is a powerful tool for intelligently modelling the curing process of rubber blends even in the case of a small dataset, and it can find a wide range of practical applications in the rubber industry.

Influence of different vulcanizing agents on structures and properties of sepiolite-filled natural rubber composites

“…In the composite, the increase of cross-link density has resulted from good rubber-filler interactions [15,16]. t S1 and t 90 indicate the commencement of vulcanization and the time required to achieve 90% of the M H value, respectively [17]. Except for the phenolic resin system, t S1 and t 90 of NR composites cross-linked with sulfur and peroxide systems were longer than those of the neat cross-linked NR counterparts.…”

Carbon black distribution driven by its concentration and its effect on physico‐mechanical properties of styrene butadiene rubber and butadiene rubber miscible rubber blends

“…CB (N220-ISAF) with surface area, DBP number, the density of 100-120 m 2 /g, 113 cm 3 /100 g, 1.7 g/cm 3 , respectively was procured from CABOT Ltd. Processing grades of aromatic oil (aniline point of 40-50 C), stearic acid (specific gravity 0.85), zinc oxide (specific gravity 5.55), TMQ (2,2,4-trimethyl-1,2-dihydroquinoline) (specific gravity 1.10), microcrystalline wax (specific gravity 0.8), TBBS (N-tert-butyl-2-benzothiazole sulphenamide) (specific gravity 1.3), sulfur (specific gravity 2.05) was procured from Samira Chemicals, Kottayam, India.…”

“…RRB's can be classified into truly miscible RRB's and immiscible RRB's. [1][2][3][4] Miscible RRB's are morphologically homogenous while on the other hand, immiscible RRB's are heterogeneous. Occurrences of miscible RRB's are rare due to the high molecular weight of rubbers and most of the commercially utilized RRB's are immiscible in nature.…”

Physics-integrated Neural Differentiable (PiNDiff) Model for Composites Manufacturing

“…The composite laminate thickness (∼ 25 mm) considered in [58] was small, and thus the spatial variation of temperature and degree of cure was not significant, making it trivial for the data-driven modeling. To increase the complexity and make the problem more challenging, a thicker laminate (25…”

“…For example, Huang et al [22] developed a data-driven model to predict the mechanical properties of carbon nanotube (CNT)-reinforced cement composites, demonstrating better generalizability and predictability than traditional response surface-based methods. Nguyen et al [24] proposed a neural network-based constitutive model to capture the evolution of the matrix mechanical properties as a function of temperature and degree of cure in composites manufacturing process, and Kopal et al [25] utilized neural networks to predict the curing characteristics of carbon black-filled rubber blends. Tao et al [26] used experimental data to discover failure criteria of composites within sparse regression framework, which promotes sparsity to find the most parsimonious model form.…”