Waste Plastic Thermal Pyrolysis Analysis by a Neural Fuzzy Model Coupled with a Genetic Algorithm

Waste plastic (WP) thermal pyrolysis is a promising method which could utilize WP to produce fuels. This study investigated this method to provide a direction for prospective industrial and commercial productions. The process temperature, the WP conversion and the pyrolysis rate are the decisive factors for industrial applications. Therefore, thermogravimetric (TG) experiments were conducted at different heating rates to obtain the experimental WP mass fraction, the WP conversion and the pyrolysis rate, which varied with the temperature and heating rate. Furthermore, a neural fuzzy model and a genetic algorithm (GA) were adopted to determine the optimal operating conditions over different temperature ranges. The neural fuzzy model-predicted WP conversion and pyrolysis rate were highly consistent with the experimental results, indicating the high accuracy of the neural fuzzy model method for this application. Moreover, the WP conversion and the pyrolysis rate optimized by the GA were 97.68% at 5.00 °C/min and 497.89 °C, and 60.66 wt%/min at 20.00 °C/min and 492.09 °C, respectively.