Development of a predictive emissions model using a gradient boosting machine learning method

Si, Minxing; Du, Ke

doi:10.1016/j.eti.2020.101028

Cited by 38 publications

(26 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gradient boosting models like XGBoost perform supervised regression tasks through an iterative approach to predict a target variable (i.e., emissions), optimizing predictive performance by combining multiple "weak" trees to t new models that are more accurate predictors of a response variable. 55,56 The XGBoost gradient-boosting model has XGBoost has been widely used in air quality monitoring [57][58][59] and greenhouse gas (GHG) emissions estimation 60 for its high e ciency, exibility, and portability. Si and Du (2020) 56 further note additional advantages of XGBoost, which requires less data preprocessing and has fewer hyperparameters, parameters an ML model uses to control the learning process for tuning.…”

Section: Model Selection -Xgboostmentioning

confidence: 99%

Predicting European Cities’ Climate Mitigation Performance using Machine Learning

Hsu

Wang

Tan

et al. 2022

Preprint

View full text Add to dashboard Cite

Although cities have risen to prominence as climate actors, emissions data scarcity has been the primary challenge to evaluating their performance. Here we develop a scalable, replicable machine learning methodology for evaluating the mitigation performance for nearly 50,000 local and municipal actors in the European Union from 2001–2018. We find that participation in one of the largest voluntary transnational climate initiatives is associated with a 1.6 percent reduction in annual emissions. Overall, these cities representing 301 million inhabitants have reduced nearly 186 million tons of carbon dioxide emissions. Compared to only 35 percent of external cities that have reduced emissions, 84 percent of cities participating in transnational climate governance have reduced emissions over the same time period. Participating cities reporting emissions data on average have higher annualized per capita reduction compared to cities without reported emissions. These findings provide quantitative evidence urban climate governance initiatives’ effect on global climate mitigation.

show abstract

Section: Model Selection -Xgboostmentioning

confidence: 99%

Predicting European Cities’ Climate Mitigation Performance using Machine Learning

Hsu

Wang

Tan

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Random Forest is an ensemble machine learning technique capable of performing regression using multiple decision trees with a statistical technique called bagging. GBM is also an ensemble category algorithm developed by Friedman [65], which uses multiple tree structures upon the loss function gradient to boost its performance and optimize prediction capability [66].…”

Section: Predictions Via Machine Learning Algorithmsmentioning

confidence: 99%

Rheological behavior of stabilized diamond-graphene nanoplatelets hybrid nanosuspensions in mineral oil

Ilyas

Ridha

Sardar

et al. 2021

Journal of Molecular Liquids

View full text Add to dashboard Cite

Recent advancements in thermo-fluid technology assisted with highly thermal conductive nanomaterials have shown assuring outcomes. It is also proven that thermal conductivity alone cannot define the overall heat transfer characteristics, and the viscous properties are equally significant towards thermal management. Therefore, this research involves investigating the rheological behavior of hybrid nanosuspensions containing high thermally conductive diamond and graphene nanoplatelets (1:1). These nanomaterials are dispersed in mineral oil using a two-step technique. Hybrid nanofluids' stability is achieved using a non-ionic stabilizer Span85, exhibiting no sedimentation for a minimum of five months. Nanomaterial characterizations are performed to study morphology, purity, and chemical analysis. The flow behavior of hybrid nanosuspensions is investigated at varying nanomaterial mass concentrations (0-2 %), temperatures (298.15-338.15 K), and shear rates (1-2000 s -1 ). Hybrid nanofluids exhibit shear-thinning behavior, which is also correlated with the Ostwald-de-Waele model. The temperature-viscosity relationship is well predicted using the Vogel-Fulcher-Tammann model. Hybrid nanofluids show a maximum enhancement of 35% viscosity at 2% concentration. A generalized twovariable correlation is used to express viscosity as a function of temperature and nanofluid concentration with an excellent agreement. Three different machine learning methods, i.e., Artificial Neural Network (ANN), Gradient Boosting Machine (GBM), and Random Forest (RF) algorithms are also introduced to predict the viscosity of hybrid nanofluids based on the three input parameters (temperature, concentration, and shear rate). The parity plots conclude that all algorithms can predict big-data viscous behavior with high precision.

show abstract

“…With the continuous improvement of computing speed in recent years, machine learning methods have been more widely used. More and more scholars apply them for disaster mapping under earthquake considering that machine learning methods could provide the ability to learn from historical data for producing insight into extreme events (Yang et al, 2015;Choubin et al, 2019;Pourghasemi et al, 2019;Jena et al, 2020;Hou et al, 2020;Si and Du, 2020;Luo et al, 2020). Aghamohammadi et al (2013) used artificial neural network (ANN) for estimating the human loss of building damage under earthquake based on the data of the 2003 Bam earthquake.…”

Section: Introductionmentioning

confidence: 99%

Earthquake-hit Population Spatial Evaluation Based on Correlation Characteristics and BP Neural Network

Zhou

Che

Xiang-hua

2021

Preprint

View full text Add to dashboard Cite

Rapid spatial evaluation of earthquake-hit population after earthquake occurrence is required in the disaster emergency rescue management, due to its significant support for decreasing casualties and property losses. The correlation between earthquake-hit population and influencing factors are analyzed using the data from the 2013 Ms7.0 Lushan earthquake. Ten influencing factors including elevation, slope angle, population density, per capita GDP, distance to fault, distance to river, NDVI, PGA, PGV and distance to epicenter, are classified into environmental factors and seismic factors. The correlation analysis reveals characteristics that there is a nonlinear relationship between the earthquake-hit population and various factors, and per capita GDP and PGA factor have a stronger correlation with earthquake-hit population. Moreover, the spatial variability of influencing factors would affect the distribution of earthquake-hit population. The earthquake-hit population is evaluated using BP neural network with optimizing training samples based on the spatial characteristics of per capita GDP and PGA factors. Different number of sample points are generated in areas with different value intervals of influencing factors, instead of the random distribution of sample points. The minimum value of RMSE (Root Mean Square Error) from testing set is 18 people/km2, showing good accuracy in the spatial evaluation of earthquake-hit population. Meanwhile, the optimizing samples considering spatial characteristics could improve the convergence speed and generalization capability comparing to random samples. The trained network was generalized to the 2017 Ms7.0 Jiuzhaigou earthquake to verify the prediction accuracy. The mean absolute error of earthquake-hit population evaluation results in different counties under the Jiuzhaigou earthquake were 18357 people and 26121 people for optimizing samples and random samples, respectively. The evaluation results indicate that BP neural network considering the correlation characteristics of factors has the capability to evaluate the earthquake-hit population in space, providing more detailed information for emergency service and rescue operation.

show abstract

Development of a predictive emissions model using a gradient boosting machine learning method

Cited by 38 publications

References 15 publications

Predicting European Cities’ Climate Mitigation Performance using Machine Learning

Predicting European Cities’ Climate Mitigation Performance using Machine Learning

Rheological behavior of stabilized diamond-graphene nanoplatelets hybrid nanosuspensions in mineral oil

Earthquake-hit Population Spatial Evaluation Based on Correlation Characteristics and BP Neural Network

Contact Info

Product

Resources

About