In recent years, more and more people are buying Electrical Vehicles (EVs) for environmental, aesthetic, and financial reasons. The number of car companies inventing EVs for their brand are increasing. Companies such as Tesla, Ford, and Rivian are taking advantage of this move toward EVs. The goal of our project is to examine just how much the range of EVs changes due to other factors such as battery pack. In this paper we analyze the relation between the range and other variables such as acceleration, top speed, battery pack, efficiency, fast charge and price in order to give us a deeper understanding of what element influences the range of EVs the most. Additionally, we can also see what car manufacturers should do to improve their EVs, and provide useful data for car companies.
Divyanshu Gupta, Kaggle (2021), Cars Dataset with Battery Capacity [Data File]. Retrieved from https://www.kaggle.com/datasets/divyanshugupta95/cars-dataset-with-battery-pack-capacity
The data was collected from different companies such as Tesla, Porsche, BMW. The data set also gives us the specific make and model of the cars. The dataset contains 14 explanatory variables with 1 response variable and a total of 102 data points.
The main covariate that we believe will have the largest impact on the range of an electric vehicle is battery packs in kilowatts per hour. The other variables that we are looking at would be acceleration, top speed, efficiency, how fast the car charges, and price. Qualitative variables would be the plug style, number of seats, power train (all wheel drive vs. four wheel drive), and type of car. This mix of qualitative and quantitative variables will allow us to give a clearer understanding of how different factors affect the range of EVs.
|0||Tesla||Model 3 Long Range Dual Motor||4.6||233||460||70.0||161||940||Yes||AWD||Type 2 CCS||Sedan||D||5||55480|
|1||Volkswagen||ID.3 Pure||10.0||160||270||45.0||167||250||Yes||RWD||Type 2 CCS||Hatchback||C||5||30000|
|2||Polestar||2||4.7||210||400||75.0||181||620||Yes||AWD||Type 2 CCS||Liftback||D||5||56440|
|3||BMW||iX3||6.8||180||360||74.0||206||560||Yes||RWD||Type 2 CCS||SUV||D||5||68040|
|4||Honda||e||9.5||145||170||28.5||168||190||Yes||RWD||Type 2 CCS||Hatchback||B||4||32997|
|97||Nissan||Ariya 63kWh||7.5||160||330||63.0||191||440||Yes||FWD||Type 2 CCS||Hatchback||C||5||45000|
|98||Audi||e-tron S Sportback 55 quattro||4.5||210||335||86.5||258||540||Yes||AWD||Type 2 CCS||SUV||E||5||96050|
|99||Nissan||Ariya e-4ORCE 63kWh||5.9||200||325||63.0||194||440||Yes||AWD||Type 2 CCS||Hatchback||C||5||50000|
|100||Nissan||Ariya e-4ORCE 87kWh Performance||5.1||200||375||87.0||232||450||Yes||AWD||Type 2 CCS||Hatchback||C||5||65000|
|101||Byton||M-Byte 95 kWh 2WD||7.5||190||400||95.0||238||480||Yes||AWD||Type 2 CCS||SUV||E||5||62000|
102 rows × 15 columns
We tested different outlier removal methods such as taking away data that was two and three standard deviations away from mean. Figure 1 compares the linear relationship of covariates against Range_Km with and without outliers respectively: