4.6 Article

Random Forest Ensemble-Based Predictions of On-Road Vehicular Emissions and Fuel Consumption in Developing Urban Areas

Related references

Note: Only part of the references are listed.
Article Energy & Fuels

The Development of CO2 Instantaneous Emission Model of Full Hybrid Vehicle with the Use of Machine Learning Techniques

Maksymilian Madziel et al.

Summary: This paper proposes a relatively computationally efficient method for creating a CO2 emission model that provides viable results for full hybrid vehicles. The study finds that the currently used emission models may not be accurate, particularly for full hybrid vehicles. The authors use a method called GPR to create the CO2 emission model, which is validated using PEMS data.

ENERGIES (2022)

Article Green & Sustainable Science & Technology

A Mixed Ensemble Learning and Time-Series Methodology for Category-Specific Vehicular Energy and Emissions Modeling

Ehsan Moradi et al.

Summary: The study finds that an ensemble time-series machine-learning methodology, specifically using Long Short-Term Memory (LSTM) architecture, is effective in capturing the dynamic effects of speed, acceleration, and grade on fuel and emission rates, outperforming traditional models. These developed ensemble models serve as representatives for vehicle categories and can be utilized in eco-driving services and environmental assessment modules.

SUSTAINABILITY (2022)

Article Transportation

Investigating the attitudes of Egyptian drivers toward traffic safety

Chantal Timmermans et al.

Summary: One of the main pillars for improving road safety is understanding the traffic safety culture and driving behavior of local drivers. This study aimed to determine the differences in traffic safety attitudes and acceptance of risky driving behavior among Egyptian drivers. The study identified three clusters of drivers based on their driving behavior and found a similarity between social norms and personal attitudes. The results can be used to enhance road safety programs in Egypt.

IATSS RESEARCH (2022)

Article Energy & Fuels

Predicting Gasoline Vehicle Fuel Consumption in Energy and Environmental Impact Based on Machine Learning and Multidimensional Big Data

Yushan Yang et al.

Summary: The underestimation of fuel consumption has various impacts and it is important to recognize the real-world fuel consumption of vehicles. In this study, a comparative analysis is conducted to forecast the real-world fuel consumption rate of light-duty gasoline vehicles in China using five regression models. The random forest regression model performs the best, and the most important factors affecting fuel consumption are brake and accelerator habits, engine power, and the fuel economy consciousness of vehicle owners.

ENERGIES (2022)

Article Physics, Multidisciplinary

Solar irradiance short-term prediction under meteorological uncertainties: survey hybrid artificial intelligent basis music-inspired optimization models

Behrooz Keshtegar et al.

Summary: This study discusses several hybrid artificial intelligence models for short-term prediction of solar radiation. The results show that the GGHS-MLNN and DGHS-MLNN models provide superior predictions compared to other models in terms of accuracy and tendency.

EUROPEAN PHYSICAL JOURNAL PLUS (2022)

Review Green & Sustainable Science & Technology

Vehicle Telematics for Safer, Cleaner and More Sustainable Urban Transport: A Review

Omid Ghaffarpasand et al.

Summary: This paper reviews the contribution of vehicle telematics to safer, cleaner and more sustainable urban transport. It discusses collection methods and technical challenges such as data processing, storage and privacy concerns. The paper highlights the roles of telematics in intelligent transportation systems, optimized routing services, safer road networks, and fairer insurance premiums estimation. It also outlines the potential of telematics in facilitating new-to-market urban mobility technologies, signalised intersections, vehicle-to-vehicle communication networks, and other IoT and IoV technologies.

SUSTAINABILITY (2022)

Article Green & Sustainable Science & Technology

Vehicle Emission Changes in China under Different Control Measures over Past Two Decades

Ning Yang et al.

Summary: Vehicle emissions have become a significant source of air pollution in urban cities, especially in the Beijing-Tianjin-Hebei region in China. The increasing vehicle ownership in the past two decades has led to changes in vehicle emissions under various management measures. The emissions of pollutants have shown a trend of first rising and then falling, except for CO2. Eliminating high emission vehicles has proven to be an effective control measure in reducing vehicle emissions.

SUSTAINABILITY (2022)

Article Energy & Fuels

Hourly predictions of direct normal irradiation using an innovative hybrid LSTM model for concentrating solar power projects in hyper-arid regions

Abdallah Djaafari et al.

Summary: This study develops hybrid LSTM models for assessing hourly direct normal irradiation (DNI) using meteorological datasets. The proposed models show low relative errors and reduced input dimensions, making them reliable for estimating DNI in arid desert areas and locations with similar climatic features.

ENERGY REPORTS (2022)

Article Energy & Fuels

Performance of Two Variable Machine Learning Models to Forecast Monthly Mean Diffuse Solar Radiation across India under Various Climate Zones

Jawed Mustafa et al.

Summary: Machine learning models were used in this study to forecast monthly-average diffuse solar radiation for different climatic zones in India. The models were compared and ranked based on their predictive power, and it was found that the K-nearest neighbours models generally outperformed random forest models. The study provides effective tools for accurately forecasting diffuse solar radiation in various climates.

ENERGIES (2022)

Article Green & Sustainable Science & Technology

A Neural Network and Principal Component Analysis Approach to Develop a Real-Time Driving Cycle in an Urban Environment: The Case of Addis Ababa, Ethiopia

Amanuel Gebisa et al.

Summary: This study developed a driving cycle model for Addis Ababa using a neural network and principal component analysis approach based on real-time data from passenger vehicles. Unlike standard driving cycles, this model better reflects the current scenario and resolves the issue of deciding the duration of the driving cycle.

SUSTAINABILITY (2022)

Article Transportation

Trip energy consumption estimation for electric buses

Jinhua Ji et al.

Summary: This study developed a trip energy consumption estimation model for electric bus fleet planning, operation, and life-cycle assessment. Real-world data from 31 electric buses operating in Jilin Province, China were collected to consider the effects of temperature variations. The model accurately estimated the energy consumption of the traction and battery thermal management system, as well as the air conditioner system, demonstrating its usefulness for fleet scheduling.

COMMUNICATIONS IN TRANSPORTATION RESEARCH (2022)

Article Engineering, Multidisciplinary

Combined Thermal Performance Enhancement of Parabolic Trough Collectors Using Alumina Nanoparticles and Internal Fins

Mohamed H. Yehia et al.

Summary: Parabolic trough collectors, combined with nanofluid and internal fins, have been found to enhance thermal efficiency by up to 7.25%. Although these techniques reduce thermal losses and improve energy utilization, they also result in increased pressure losses. The combination of both techniques shows the highest improvement in thermal performance.

INTERNATIONAL JOURNAL OF ENGINEERING RESEARCH IN AFRICA (2022)

Article Energy & Fuels

A Comparative Study on Fuel Consumption Prediction Methods of Heavy-Duty Diesel Trucks Considering 21 Influencing Factors

Jian Gong et al.

Summary: This study analyzed the influence of various factors on fuel consumption during driving heavy-duty diesel trucks (HDDTs) using naturalistic driving data. Different machine learning algorithms were used to build fuel consumption predictors, with random forest showing the best performance in prediction accuracy. The conclusions can help transportation companies in formulating driving training strategies and reducing energy consumption and emissions.

ENERGIES (2021)

Article Green & Sustainable Science & Technology

Selection methodology of representative meteorological days for assessment of renewable energy systems

Muhammed A. Hassan et al.

Summary: The study proposes a new approach to reduce costs and improve accuracy of renewable energy data reconstruction by using representative days. Including system-irrelevant meteorological parameters increases data disagreement, while the autoencoder feature extraction method outperforms conventional statistical methods.

RENEWABLE ENERGY (2021)

Article Economics

Modelling the energy consumption of electric vehicles under uncertain and small data conditions

Yang Liu et al.

Summary: The study combines machine learning and controlled experiments to propose a new model for estimating the energy consumption of electric vehicles. By parameter search and data augmentation methods, the accuracy of the base learner is improved, and experimental results show that the proposed model outperforms existing machine learning models in estimation accuracy.

TRANSPORTATION RESEARCH PART A-POLICY AND PRACTICE (2021)

Article Environmental Sciences

A representative urban driving cycle for passenger vehicles to estimate fuel consumption and emission rates under real-world driving conditions

Omar A. Huzayyin et al.

Summary: The driving patterns in developing countries differ significantly from those in developed countries, especially in metropolitan cities. Current driving cycles are not accurate in estimating emission rates, and a study in Greater Cairo proposes a specific driving cycle and develops two global driving cycles based on onboard measurements. The cycle based on the K-medoids algorithm is found to be more representative and superior in estimating fuel consumption and emission rates compared to commonly used cycles like the American transient FTP-75 and the European modal ECE cycle.

URBAN CLIMATE (2021)

Article Environmental Sciences

Features Importance Analysis of Diesel Vehicles' NOx and CO2 Emission Predictions in Real Road Driving Based on Gradient Boosting Regression Model

Hung-Ta Wen et al.

Summary: The study investigated the impact of input features on emissions models of light diesel vehicles and utilized gradient boosting regression models for training and prediction. The results showed varying effects of different numbers of features on model training, with the best models achieving high coefficient of determination values.

INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH (2021)

Article Environmental Sciences

Prediction of instantaneous real-world emissions from diesel light-duty vehicles based on an integrated artificial neural network and vehicle dynamics model

Jigu Seo et al.

Summary: This paper presents a road vehicle emission model that integrates an artificial neural network (ANN) model with a vehicle dynamics model to predict the instantaneous carbon dioxide (CO2), nitrogen oxides (NOx) and total hydrocarbon (THC) emissions of diesel light-duty vehicles. Real-world measurement data were used to train a multi-layer feed-forward ANN model, which showed satisfactory prediction accuracy for road vehicle emissions. The optimal combination of experimental variables was selected as the ANN input, achieving both practicality and accuracy in predicting vehicle emissions.

SCIENCE OF THE TOTAL ENVIRONMENT (2021)

Article Management

Comparison of artificial intelligence and empirical models for energy production estimation of 20 MWp solar photovoltaic plant at the Saharan Medium of Algeria

Kada Bouchouicha et al.

Summary: This study estimated the electric power production of a 20 MWp solar photovoltaic plant in Algeria using minimal knowledge about weather conditions, utilizing linear and nonlinear simulation models. Comparison between artificial neural network (ANN) and multiple linear regression (MLR) models showed that ANN-based models were superior in prediction accuracy, with cascade-forward neural network (CNN) models providing the most accurate results. The proposed forecasting model has practical implications for solar PV forecasting and social implications for improving grid electricity system operations.

INTERNATIONAL JOURNAL OF ENERGY SECTOR MANAGEMENT (2021)

Article Engineering, Civil

Vehicle Fuel Consumption Prediction Method Based on Driving Behavior Data Collected from Smartphones

Ying Yao et al.

JOURNAL OF ADVANCED TRANSPORTATION (2020)

Article Environmental Sciences

Modelling of instantaneous emissions from diesel vehicles with a special focus on NOx: Insights from machine learning techniques

Clemence M. A. Le Cornec et al.

SCIENCE OF THE TOTAL ENVIRONMENT (2020)

Article Engineering, Civil

Estimating Light-Duty Vehicle Emission Factors using Random Forest Regression Model with Pavement Roughness

Fengxiang Qiao et al.

TRANSPORTATION RESEARCH RECORD (2020)

Article Environmental Sciences

Estimation of bus emissionmodels for different fuel types of buses under real conditions

Chao Wang et al.

SCIENCE OF THE TOTAL ENVIRONMENT (2018)

Article Environmental Studies

NOx emissions from diesel light duty vehicle tested under NEDC and real-word driving conditions

A. Ramos et al.

TRANSPORTATION RESEARCH PART D-TRANSPORT AND ENVIRONMENT (2018)

Article Construction & Building Technology

Incorporating driving cycle based fuel consumption estimation in green vehicle routing problems

Surendra Reddy Kancharla et al.

SUSTAINABLE CITIES AND SOCIETY (2018)

Article Construction & Building Technology

Development of Mashhad driving cycle for passenger car to model vehicle exhaust emissions calibrated using on-board measurements

Mohammad Amin Pouresmaeili et al.

SUSTAINABLE CITIES AND SOCIETY (2018)

Proceedings Paper Computer Science, Theory & Methods

Vehicle emission prediction using remote sensing data and machine learning techniques

Jiazhen Chen et al.

33RD ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING (2018)

Article Green & Sustainable Science & Technology

Vehicular CO Emission Prediction Using Support Vector Regression Model and GIS

Omer Saud Azeez et al.

SUSTAINABILITY (2018)

Article Environmental Sciences

Modeling of real time exhaust emissions of passenger cars under heterogeneous traffic conditions

Rohit Jaikumar et al.

ATMOSPHERIC POLLUTION RESEARCH (2017)

Article Energy & Fuels

Exploring the potential of tree-based ensemble methods in solar radiation modeling

Muhammed A. Hassan et al.

APPLIED ENERGY (2017)

Article Green & Sustainable Science & Technology

Potential of four different machine-learning algorithms in modeling daily global solar radiation

Muhammed A. Hassan et al.

RENEWABLE ENERGY (2017)

Article Engineering, Civil

Evaluation of Light-Duty Gasoline Vehicle Rated Fuel Economy Based on In-Use Measurements

Tanzila Khan et al.

TRANSPORTATION RESEARCH RECORD (2016)

Article Statistics & Probability

Modern Multivariate Statistical Techniques: Regression, Classification and Manifold Learning

John Maindonald

Journal of Statistical Software (2015)

Article Engineering, Civil

Effects of Errors on Vehicle Emission Rates from Portable Emissions Measurement Systems

Gurdas S. Sandhu et al.

TRANSPORTATION RESEARCH RECORD (2013)

Proceedings Paper Transportation

Statistical Vehicle Specific Power Profiling for Urban Freeways

Zhuo Yao et al.

INTELLIGENT AND INTEGRATED SUSTAINABLE MULTIMODAL TRANSPORTATION SYSTEMS PROCEEDINGS FROM THE 13TH COTA INTERNATIONAL CONFERENCE OF TRANSPORTATION PROFESSIONALS (CICTP2013) (2013)

Review Environmental Sciences

Validation of road vehicle and traffic emission models - A review and meta-analysis

Robin Smit et al.

ATMOSPHERIC ENVIRONMENT (2010)

Article Computer Science, Artificial Intelligence

Random forests

L Breiman

MACHINE LEARNING (2001)