Mg Road Call Girls Service: 🍓 7737669865 🍓 High Profile Model Escorts | Banga...
14_july_DataScience_On_EMobility.pptx
1. DATA SCIENCE
FOR E-MOBILITY
Ashish Patel
Sr.AWS AI ML Solution Architect at
Book Author of Hands-on Time
Series
Analytics With Python
/ashishpatel2604
2. WHAT IS E-MOBILITY?
E-Mobility
refers to the use of electric vehicles (EVs) and
associated technologies that enable
sustainable transportation
DATA SCIENCE FOR E-MOBILITY
2
3. DATA SCIENCE FOR E-MOBILITY
E-MOBILITY : BENEFITS and GROWTHS
BENEFITS
Cleaner and more sustainable
More efficient
Lower operating costs
Quieter
Better for public health
GROWTHS
Market Rapid Growth
Global Demand Increase up to $800
billion by 2025
In the United States, the number of
EVs on the road is expected to reach
12 million by 2030
Factors : Government incentives,
technological advances, and
increasing consumer demand
3
4. DATA SCIENCE FOR E-MOBILITY
DATA SCIENCE IMPORTANCE FOR E-MOBILITY
• Data science can be used to improve the performance, efficiency, and
safety of e-mobility systems.
Optimize charging
infrastructure
Predict demand for
electric vehicles
Improve the design of
e-mobility vehicles
Data Science
Importance in
E-Mobility
Improved
efficiency and
sustainability
Increased
safety
Enhanced
user
experience
New business
opportunities
Examples
4
5. DATA SCIENCE FOR E-MOBILITY
DATA GENERATION E-MOBILITY
1 2 3 4
Speed
Acceleration
Battery level
Time
Duration of charging
Traffic Pattern
Urban Centres
Regions near
charging Station
Temperature
Humidity
5
6. DATA SCIENCE FOR E-MOBILITY
USECASES
Optimizing charging
infrastructure
Predicting demand for
electric vehicles
Improving the design
of e-mobility vehicles
Reducing traffic
congestion Improving air quality
DATA SCIENCE USE CASES FOR E-MOBILITY
6
7. DATA SCIENCE FOR E-MOBILITY
USE CASE: Charging Infrastructure Planning Problems
7
Global energy consumption and air pollution are driving the
shift to EVs
EVs are a cleaner and more sustainable form of transportation.
A well-functioning charging infrastructure is essential for the
large-scale adoption of EVs.
Planning the charging infrastructure is a complex task that
requires coordination between the power distribution network
and the road network.
8. DATA SCIENCE FOR E-MOBILITY
Challenges in Planning a Charging Infrastructure for
Electric Vehicles
8
High cost of
infrastructure
deployment
Interoperability and
standardization
of charging systems
Balancing charging
load with available
power capacity
Lack of accurate
data on EV usage
patterns and
charging behaviour
10. DATA SCIENCE FOR E-MOBILITY
10
Architecture Design for Machine Learning Approach
Input
Geospatial Data
Charging Demand Data
Weather Data
EV Fleet Data
Machine Learning Systems
Linear Regression, Decision Trees, Random Forests, or
Support Vector Regression
Regression
Feedback
Logistic Regression, Naive Bayes, Decision Trees,
Random Forests, or Support Vector Machines
Classification
K-means, DBSCAN, or Hierarchical Clustering
Clustering
PCA, t-SNE, Apriori or FP-Growth
Association Rules, Dimensionality
Reduction
Output
ESTIMATE CHARGING
DEMAND
CHARGING STATION
SUITABILITY PLACEMENT
TARGETED INFRASTRUCTURE
PLANNING
UNDERSTANDING
CHARGING BEHAVIOR
Algorithm Evaluation, Explainable AI, Data Drift
11. Decision : Optimal Placement of Charging Station
OPTIMAL
DECISION OF
CHARGING
STATION
PLACEMENT
Geographic
coverage
Proximity to
power supply
Accessibility
and
convenience
Cost-
effectiveness
11