Machine Learning and Hyperspectral Imaging-Aided Forecast for the Share of Biogenic and Fossil Carbon in Solid Waste

Abstract

Traditional methods for analyzing the biogenic and fossil carbon shares in solid waste are time-consuming and labor-intensive. A novel approach was developed to directly classify the carbon group and predict carbon content using the hyperspectral imaging (HSI) spectra of solid waste in conjunction with state-of-the-art tree-based machine learning models, including random forest (RF), extreme gradient boost, and light gradient boost machine (LGBM). All of the classifiers and regressors were able to achieve an accuracy above 0.95 and an R 2 of 0.96 in the test set, respectively. In addition, two model interpretation approaches, the Shapley additive explanation and model explainer, were applied. The results showed that the predictions of the developed models were based on a reasonable understanding of the overtone and shake of the functional groups (C–H, N–H, and O–H). Furthermore, the developed models were validated by an external test set, which did not overlap with the data used for model construction. The RF and LGBM showed robust performance with a 0.790 accuracy for carbon group classification and a 0.806 R 2 for carbon content prediction. Overall, the optimal models provided a rapid method for characterizing the biogenic carbon share in solid waste based on raw HSI spectra without preprocessing.

A Novel High-Throughput Detection Method for Plastic Debris in Organic-Rich Matrices Based on Image Fusion

“…To avoid overexposure, the exposure time was set at 3.1 ms and the frame rate was set at 50 Hz. Other setting parameters remained consistent with Lan et al, 27 sample. Scanning occurred at 600 dots per inch (dpi) in reflective mode to avoid the loss of details in plastic debris and matrices.…”

“…Plastic products were collected from solid waste. 27 All plastic resins and products were identified using attenuated total reflectance (ATR)-FTIR (Nicolet iS20, Thermo Fisher Scientific), and the corresponding spectra are shown in Figures S1 and S2. LDPE film was cut using scissors, while other hard plastics were frozen with liquid nitrogen and milled into fragments.…”

“…Hyperspectral imaging (HSI) system with pushroom mode is a promising technique for characterizing the solid waste. 27 HSI can simultaneously capture spatial and spectral information on plastic debris in the organic-rich matrix simultaneously. 28,29 However, owing to the technical and physical limitations of HSI sensors, trade-offs exist between spectral and spatial resolution.…”

“…Researchers have optimized the detected area of Raman imaging to 12.4 mm × 12.4 mm using optical and collection techniques, yet detecting areas at the centimeter level remains time-consuming. , This limitation results in a narrow size range of detected MPs, not to mention macroplastics. Hyperspectral imaging (HSI) system with pushroom mode is a promising technique for characterizing the solid waste . HSI can simultaneously capture spatial and spectral information on plastic debris in the organic-rich matrix simultaneously. , However, owing to the technical and physical limitations of HSI sensors, trade-offs exist between spectral and spatial resolution.…”

Machine Learning-Based Prediction of the CO2 Concentration in the Flue Gas and Carbon Emissions from a Waste Incineration Plant

“…Compared to traditional modeling approaches, such as direct testing and the IPCC factor calculation, machine learning (ML) algorithms are adept at modeling complex nonlinear processes. In recent years, ML has gained significant attention in the field of MSW management. − In incineration operations, ML models have been used to forecast the performance of a combustion boiler when processing waste plastics and to predict the gas composition in medical waste incineration plants. , Zhu et al developed an extreme gradient boosting (XGBoost) model for predicting CO 2 emissions in thermal power plants using a few key observable factors. However, compared to thermal power plants, the fuel composition in waste incineration plants is more diverse and the factors influencing CO 2 emissions can vary based on the incineration process and flue gas purification conditions.…”