Ensemble machine learning-based algorithm for electric vehicle user behavior prediction

Chung, Yu-Wei; Khaki, Behnam; Li, Tianyi; Chu, Chi-Cheng; Gadh, Rajit

doi:10.1016/j.apenergy.2019.113732

Cited by 141 publications

(62 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…MAPE is problematic when the actual value is close to 0 in the denominator and therefore creating a bias. On the contrary, SMAPE is the preferred metric for EV charging prediction because both the actual value and the predicted value is in the denominator [49].…”

Section: ) Evaluation Of Regressionmentioning

confidence: 99%

“…The authors in [49] utilized several ML models, including DT, K-NN and RF to predict session duration and energy consumption from two charging datasets. The first dataset contained charging sessions from University of California, Los Angeles (UCLA) campus, thus representing nonresidential charging behavior.…”

Section: Supervised Learning For Analysis and Prediction Of Charmentioning

confidence: 99%

“…Table 1 summarizes the research that used supervised learning for EV charging behavior predictions. [49] Predict session duration and energy consumption from both residential and non-residential SVR, RF and DKDE combined to form an ensembled model. SMAPE charging duration: 10.4%, SMAPE energy consumption: 7.54%.…”

Section: Supervised Learning For Analysis and Prediction Of Charmentioning

confidence: 99%

See 2 more Smart Citations

Machine Learning Approaches for EV Charging Behavior: A Review

et al. 2020

View full text Add to dashboard Cite

As the smart city applications are moving from conceptual models to development phase, smart transportation is one of smart cities applications and it is gaining ground nowadays. Electric Vehicles (EVs) are considered one of the major pillars of smart transportation applications. EVs are ever growing in popularity due to their potential contribution in reducing dependency on fossil fuels and greenhouse gas emissions. However, large-scale deployment of EV charging stations poses multiple challenges to the power grid and public infrastructure. To overcome the issue of prolonged charging time, the simple solution of deploying more charging stations to increase charging capacity does not work due to the strain on power grids and physical space limitations. Therefore, researchers have focused on developing smart scheduling algorithms to manage the demand for public charging using modeling and optimization. More recently, there has been a growing interest in data-driven approaches in modeling EV charging. Consequently, researchers are looking to identify consumer charging behavior pattern that can provide insights and predictive analytics capability. The purpose of this paper is to provide a comprehensive review for the use of supervised and unsupervised Machine Learning as well as Deep Neural Networks for charging behavior analysis and prediction. Recommendations and future research directions are also discussed.

show abstract

Section: ) Evaluation Of Regressionmentioning

confidence: 99%

Section: Supervised Learning For Analysis and Prediction Of Charmentioning

confidence: 99%

Section: Supervised Learning For Analysis and Prediction Of Charmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning Approaches for EV Charging Behavior: A Review

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Recently, there have been some efforts to predict charging behaviour with machinelearning methods. The impact of accurate predictions on charging scheduling has been demonstrated in [36], where the authors suggest a potential 27% decrease in peak load. However, when using historical data of parking durations and energy consumption to predict user behaviour, there is a lack of understanding of individual users' preferences for service attributes, such as the location of the charger and the charging rates.…”

Section: Representation Of Charging Demandmentioning

confidence: 99%

Optimal Pricing of Vehicle-to-Grid Services Using Disaggregate Demand Models

2021

View full text Add to dashboard Cite

The recent revolution in electric mobility is both crucial and promising in the coordinated effort to reduce global emissions and tackle climate change. However, mass electrification brings up new technical problems that need to be solved. The increasing penetration rates of electric vehicles will add an unprecedented energy load to existing power grids. The stability and the quality of power systems, especially on a local distribution level, will be compromised by multiple vehicles that are simultaneously connected to the grid. In this paper, the authors propose a choice-based pricing algorithm to indirectly control the charging and V2G activities of electric vehicles in non-residential facilities. Two metaheuristic approaches were applied to solve the optimization problem, and a comparative analysis was performed to evaluate their performance. The proposed algorithm would result in a significant revenue increase for the parking operator, and at the same time, it could alleviate the overloading of local distribution transformers and postpone heavy infrastructure investments.

show abstract

“…From Table 3 it can be shown that the MAPE for the surface roughness is higher than that in the maximum forming angle model. This is because using the MAPE with the small output values (i.e., Ra values) has a problem since it takes only the observed values in the denominator (see Equation (7)) which leads to a reduction in the overall accuracy [31]. Later, in the evaluation section, suitable criteria will be used for both models to show the accuracy of the predictive models.…”

Section: Optimum Gbrt Parameters For Maximum Forming Angle and Surfacmentioning

confidence: 99%

Experimental Investigation and Optimal Prediction of Maximum Forming Angle and Surface Roughness of an Al/SUS Bimetal Sheet in an Incremental Forming Process Using Machine Learning

et al. 2019

View full text Add to dashboard Cite

Bimetal sheets have superior properties as they combine different materials with different characteristics. Producing bimetal parts using a single-point incremental forming process (SPIF) has increased recently with the development of industrial requirements. Such types of sheets have multiple functions that are not applicable in the case of monolithic sheets. In this study, the correlation between the operating variables, the maximum forming angle, and the surface roughness is established based on the ensemble learning using gradient boosting regression tree (GBRT). In order to obtain the dataset for the machine learning, a series of experiments with continuous variable angle pyramid shape were carried out based on D-Optimal design. This design is created based on numerical variables (i.e., tool diameter, step size, and feed rate) and categorical variable (i.e., layer arrangement). The grid search cross-validation (CV) method was used to determine the optimum GBRT parameters prior to model training. After the parameter tuning and model selection, the model with a better generalization performance is obtained. The reliability of the predictive models is confirmed by the testing samples. Furthermore, the microstructure of the aluminum/stainless steel (Al/SUS) bimetal sheet is analyzed under different levels of operating parameters and layer arrangements. The microstructure results reveal that severe cracks are attained in the case of a small tool diameter while a clear refinement is observed when a high tool diameter value with small step down is used for both Al and SUS layers. different materials that have different characteristics, this strengthens the conductivity, corrosion resistance, and the yield strength of the combined sheet. The application of such processes to produce the desired composite parts combines the characteristics of both the forming process and the parent materials to enhance the quality and the formability of the low formable materials and reduce the cost and the weight of manufactured parts.The SPIF of composite sheets has generated considerable recent research interest. In the study of Al-Ghamdi and Hussain [1], the formability improvement of a Cu/steel bimetal sheet during SPIF was better than that in the stamping process due to the limitation of the delamination in the former process. In their further work [2], the relation between the maximum forming angle and the process parameters of the tri-layer Cu/steel/Cu was studied experimentally. The results of the study revealed that a complex relation was observed between the operating parameters of the tri-layer sheet. Gheysarian and Honarpisheh [3] investigated the effect of the layer arrangement, tool path, and tool diameter on formability (i.e., fracture depth) and surface quality of Al/Cu bimetal sheets. The results showed the improvement in the formability corresponding to a large tool diameter, spiral tool path, and copper layer as a contacted sheet. Later, Honarpisheh et al. [4] focused on the improvement of the maximum forming angl...

show abstract

Ensemble machine learning-based algorithm for electric vehicle user behavior prediction

Cited by 141 publications

References 22 publications

Machine Learning Approaches for EV Charging Behavior: A Review

Machine Learning Approaches for EV Charging Behavior: A Review

Optimal Pricing of Vehicle-to-Grid Services Using Disaggregate Demand Models

Experimental Investigation and Optimal Prediction of Maximum Forming Angle and Surface Roughness of an Al/SUS Bimetal Sheet in an Incremental Forming Process Using Machine Learning

Contact Info

Product

Resources

About