An environment-friendly leather waste-based liquid film mulching and its application for facilitating the growth of maize crops

Mao Yang,Li Yan,Yanchun Li,Pan Huang,Wenjia Han,Xugang Dang
DOI: https://doi.org/10.1016/j.psep.2022.01.073
IF: 7.8
2022-03-01
Process Safety and Environmental Protection
Abstract:Currently, environment-friendly high water-absorbing, water-retention, water-resistant, and water erosion resistance materials are the research hotspot in agricultural film. In this research, a new type of leather waste-based liquid film mulching (WG) was prepared by waterborne polyurethane (WPU) grafted with gelatin(G) extracted from leather waste. The interacting behavior of the prepared WG with soil particles was characterized including soil particles morphological changes, water-retention, water-resistant, water erosion resistance, and farming experiments. The SEM revealed that the WG could form a "coating layer" on the soil surface. The coating layer could improve the water-retention and water resistance of the soil. Meanwhile, the farming applications found that the soil sprayed WG could form a certain closed space which played a role in water-resistant, water erosion resistance, heat preservation, and moisture retention for maize. Especially, the water-absorbing, water-retention, and thermal-retention efficiency reached 80.5%, 20.2%, and 23.1 °C, respectively when the WG content was 50 g/m2 in the soil experimental sample. Moreover, after the WG was sprayed evenly on farmland, the germination period of the maize seeds was significantly shortened, the germination rate reached 92%, which was far higher than that of the bare soil. As an environment-friendly leather waste-based liquid film mulching, the WG could not only meet the growth of maize, increase yield, but also provide new ideas for the recycling of leather waste.Download : Download high-res image (231KB)Download : Download full-size imageContinuous monitoring of wastewater treatment processes is key to mitigate the risk of inadequately treated wastewater on the environment and public health. However, effective control of wastewater treatment processes is challenging because of the numerous relevant variables and their complex physio-chemical-biological interdependence. Most published related studies focused only on correlating the effluent concentration of chemical oxygen demand and/or suspended solids using only a limited number of wastewater influent variables. In addition, recent machine learning- (ML) based studies in wastewater treatment systems considered only individual classification algorithms rather than providing a comparison between different algorithm performances. In the current study, different algorithms were developed to categorize a range of wastewater treatment effluent characteristics based on multiple influent variables. To demonstrate their application, 23 ML classification algorithms were deployed on a membrane aerated biofilm reactor-generated dataset and their performances were evaluated considering two different group of metrics related to the removal efficiency and effluent quality. The analysis results showed that, among all considered algorithms, the ensemble bagged trees algorithm had the most superior performance in terms of its overall classification accuracy. An interpretability analysis was further performed on the treatment process variables to detect the correlation between the input and output variables and to assess variable importance. In practice, the developed algorithms can facilitate optimal operation and effective management of wastewater treatment plants. ML algorithms also present efficient tools for rapidly classifying the effluent characteristics in lieu of typical sampling and laboratory analysis processes.The use of unsupervised machine learning methods such as K-means, Hierarchical Agglomerative Clustering, and Self-organizing maps is constantly increasing in seismic interpretation. Regarding unsupervised methods, the K-means technique is one of the simplest ways to cluster seismic facies, although it presents neither a structure between the generated labels nor a measure of similarity when considering their order. To solve this drawback, we propose two automated label organization techniques that use principal component analysis (PCA) to organize those obtained from the algorithm, preserving some degree of similarity. To demonstrate the effectiveness of these methods, we interpreted two stratigraphic surfaces known as Maximum Transgressive Surface (MTS) and Maximum Regressive Surface (MRS), then extracted some attributes to run clustering experiments. Furthermore, we performed the principal component analysis and selected the first three components to be clustered. Subsequently, these components were used to organize the labels obtained with K-means through the two proposed techniques. Finally, we interpreted the outstanding results obtained from the met -Abstract Truncated-
engineering, environmental, chemical
What problem does this paper attempt to address?