Study on HHV prediction of municipal solid wastes: A machine learning approach

Abstract

In this work, artificial neural network (ANN) models and particle swarm optimization (PSO) models based on machine learning were built to predict the HHVs of MSW quickly. Four kinds of BP ANN models and two PSO models were built using proximate analysis and ultimate analysis of 33 MSW samples as input variables. As a comparison, three classical linear empirical models employed from publications were also used. The modeling results show that the input variables had significant influence on the prediction accuracy. The ANN model based on proximate analysis had lower precision for HHV estimation. The ultimate analysis had better prediction performance while the combination of ultimate analysis and proximate analysis (ANN-4 model) had the best accuracy. With regard to ANN-4 model, the largest relative deviation for all samples was lower than 10%. Due to the complex composition of MSW, the linear empirical model was not suitable for accurate prediction of the calorific value of MSW. Nonlinear empirical formulas obtained by PSO models improved the prediction performance for most samples. In general, the ANN modeling method could predict the thermochemical properties of MSW and provide rapid and effective guidance for the operation of MSW incineration process.

Prediction of municipal solid waste generation and analysis of dominant variables in rapidly developing cities based on machine learning – a case study of China

“…BP neural network adds several hidden layers (one or several layers) of neurons between the input layer and the output layer. These neurons are called hidden units, which will affect the relationship between input variables and output values (Zhu and Yang, 2022).…”

Machine Learning Prediction of Phosphate Adsorption on Six Different Metal-Containing Adsorbents

“…In order to optimize the phosphate adsorption process and direct the choice and preparation of adsorption materials, we can simultaneously apply the ML approach to the adsorption of phosphate. Previous documents have reported the applications of ML in solving multiple environmental problems such as heavy metal removal, , micropollutant oxidation, , seawater desalination, , carbon dioxide adsorption, , and municipal solid-waste treatment. , A number of ML models such as bagging, linear regression (LR), neural networks (NNs), and support vector machines (SVMs), and tree-based ML models have been developed in previous studies. Particularly, decision tree-based algorithms, including gradient boosting decision tree (GBDT), decision tree (DT), extreme gradient boosting (XGBoost), categorical boosting (CatBoost), light gradient boosting machine (LightGBM), and random forest (RF), are a subcategory of supervised ML models. ,− DT is a tree structure that is like a binary tree or multitree.…”

Estimation of fast pyrolysis bio‐oil properties from feedstock characteristics using rough‐set ‐based machine learning

“…As a result, there has been a notable trend in the use of ML to complement mechanistic models for such applications. 7,8 For example, Cheng et al 9 used a random forest ML model coupled with life cycle assessment and economic analysis on slow pyrolysis for biochar production in negative emissions systems. Elmaz et al 10 tested ML approaches based on regression, support vector machine (SVM), decision tree, and perceptron models to predict gasification performance of biomass gasification.…”