DU Kang, ZHONG Wenqi, CHEN Xi, ZHOU Guanwen
Waste plastics, with their high hydrogen-to-carbon (H/C) atomic ratios, can act as hydrogen donors during coal pyrolysis, thereby enhancing tar yield and quality. Thus far, a study has been conducted on the co-pyrolysis characteristics of coal and waste plastic, along with a rapid prediction method for tar yield. An experimental system for the co-pyrolysis of coal and waste plastic is established to examine the distribution patterns of pyrolysis products, such as gas, tar, and char, at varying temperatures and coal-to-waste plastic ratios. The results indicate a significant synergistic effect during the co-pyrolysis of coal and plastic waste. As the blending ratio of waste plastic increases, the tar yield also increases, with the value of the synergistic effect parameter initially rising and then falling. As the blending ratio continues to increase, the formation of a liquid phase becomes more prevalent on the surface of coal particles during the pyrolysis process, which inhibits tar release and leads to a gradual decrease in the positive synergistic effect of the waste plastic on tar yield. Based on these findings, a rapid prediction model for tar yield has been developed using neural networks and optimized with a Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), achieving a 10.52% reduction in the average prediction error under training conditions. The proposed model is utilized to predict the tar yield for new conditions in the database, with the relative error generally maintained within (–20%, 30%), demonstrating good accuracy and utility.