Environmental impact assessment of multi-source solid waste based on a life cycle assessment, principal component analysis, and random forest algorithm.

As a national pilot city for solid waste disposal and resource reuse, Dongguan in Guangdong Province aims to vigorously promote the high-value utilization of solid waste and contribute to the sustainable development of the Greater Bay Area. In this study, life cycle assessment (LCA) coupled with principal component analysis (PCA) and the random forest (RF) algorithm was applied to assess the environmental impact of multi-source solid waste disposal technologies to guide the environmental protection direction. In order to improve the technical efficiency and reduce pollution emissions, some advanced technologies including carbothermal reduction‒oxygen-enriched side blowing, directional depolymerization‒flocculation demulsification, anaerobic digestion and incineration power generation, were applied for treating inorganic waste, organic waste, kitchen waste and household waste in the park. Based on the improved techniques, we proposed a cyclic model for multi-source solid waste disposal. Results of the combined LCA-PCA-RF calculation indicated that the key environmental load type was human toxicity potential (HTP), came from the technical units of carbothermal reduction and oxygen-enriched side blowing. Compared to the improved one, the cyclic model was proved to reduce material and energy inputs by 66%-85% and the pollution emissions by 15%-88%. To sum up, the environmental impact assessment and systematic comparison suggest a cyclic mode for multi-source solid waste treatments in the park, which could be promoted and contributed to the green and low-carbon development of the city.

Learning Interactions in Reaction Diffusion Equations by Neural Networks.

Partial differential equations are common models in biology for predicting and explaining complex behaviors. Nevertheless, deriving the equations and estimating the corresponding parameters remains challenging from data. In particular, the fine description of the interactions between species requires care for taking into account various regimes such as saturation effects. We apply a method based on neural networks to discover the underlying PDE systems, which involve fractional terms and may also contain integration terms based on observed data. Our proposed framework, called Frac-PDE-Net, adapts the PDE-Net 2.0 by adding layers that are designed to learn fractional and integration terms. The key technical challenge of this task is the identifiability issue. More precisely, one needs to identify the main terms and combine similar terms among a huge number of candidates in fractional form generated by the neural network scheme due to the division operation. In order to overcome this barrier, we set up certain assumptions according to realistic biological behavior. Additionally, we use an L2-norm based term selection criterion and the sparse regression to obtain a parsimonious model. It turns out that the method of Frac-PDE-Net is capable of recovering the main terms with accurate coefficients, allowing for effective long term prediction. We demonstrate the interest of the method on a biological PDE model proposed to study the pollen tube growth problem.