Optimizing Urban Solid Waste-to-Energy Production Systems with Backpropagation Neural Network Model

Environmental concerns and global MSW generation have spurred waste-to-energy (WtE) research. These projects support the global bioeconomy and biorefineries to generate renewable energy and reduce fossil fuel use. Preprocessing, feature extraction, and model training comprise the approach. Outliers are removed and input values are normalized using zero-mean normalization. For feature extraction, GW was chosen because to its localization and frequency domain capabilities. Model training uses the Backpropagation Neural Network (BPNN). The proposed model outperformed SVMs and ANNs with 91.45% accuracy. Thus, the model's strength is waste-to-energy conversion. The results indicate that the suggested method optimizes MSW processing for renewable energy. This method is promising for WtE technology, which will help the globe switch to renewable energy.