Optimizing plastic waste management and pollution control using a deep neural network-based predictive model
摘要
This study analyzes the mathematical model of pollution arising from industrial plastic waste management (PPWM) using the integration of a Physics-Informed Neural Networks methodology. The model consists of five categories: Industrial plastic waste (I), Landfill waste incineration (B), Plastic waste disposal (D), Plastic waste recycling (R), and Environmental pollution (P). The Particle Swam Optimization- Physics-Informed Neural Networks (PSO-PINNs) algorithm employs an ensemble approach to PINN, facilitating precise predictions while minimizing undesirable characteristics. The model’s efficacy is validated by the alignment of actual values with projected PPWM solutions. Critical environmental aspects, such as combustion, disposal, and recycling, are analyzed to assess their impact on pollution levels. The results indicate that mitigating these factors significantly reduces pollution, highlighting the necessity for improved waste management systems. The model’s reliability is validated by its consistent attainment of minimal residuals and preservation of symmetry around zero error. The AE ranges from