Applied Data Science - EE25C08

 

EE25C08:

Applied Data Science

 

Course Objective: The Students will be equipped with data science competencies to analyze complex electrical systems through machine learning and statistical modelling, and foster application-based problem solving for optimizing smart grid operations, load forecasting, and renewable energy integration. This course enables to design of IoT-driven frameworks for real-time monitoring and anomaly detection in modern power networks, facilitates to develop industry-oriented skills in predictive maintenance and fault diagnosis for enhancing data-driven decision-making within digital energy ecosystems.

Foundations of Data Science

Data science lifecycle, Electrical engineering data : sensor, SCADA, IoT, Data preprocessing and visualization, Data Science Context- Prerequisites for a Data Scientist – Tools and Skills required.

*Activity – Microsoft Visio / MATLAB Simulink

1. Smart Meter Data Cleaning

2. Transformer Temperature Monitoring

 

Statistical Methods in Electrical Engineering

Probability and distributions- Statistical analysis of load uncertainty and renewable energy variability in power systems. Correlation and linear regression models for prediction of electrical parameters and energy consumption, Hypothesis testing

*Activity – Preparation of Cheat Sheet/Microsoft Visio / MATLAB Simulink

1. Fault Detection using Statistical Analysis

2. Load vs Temperature Correlation Analysis

 

Machine Learning Applications in Electrical Systems

Supervised and unsupervised learning techniques for power systems and smart grids, Regression methods for load and energy prediction, Classification algorithms for fault detection and system state identification, Clustering techniques for load pattern analysis and anomaly detection, and model evaluation using accuracy, precision, recall, and F1-score.

*Activity – Quiz, Reproduction of research paper.

 

Signal Processing and Data Analytics

Time series analysis of electrical signals for trend and forecasting, Fourier transform and FFT for frequency and harmonic analysis, feature extraction from signals for condition monitoring, and application of signal analytics for power quality assessment and predictive maintenance.

*Activity – Project Based Learning/Research writing

1. Power Quality Analysis to detect harmonics.

2. Predictive Maintenance to analyze motor vibration data.

 

Data-Driven Smart Grid and Sustainable Energy Systems

Smart grid architecture and communication, renewable energy forecasting and integration, IoT-based monitoring and real-time analytics, data-driven optimization for load management, and applications in smart grids and energy systems.

*Activity – Project Based Learning/Research writing

1. Solar Power Prediction

2. Smart Grid Optimization

 

Suggested Practices: (Analogy)

1. Simulation of Smart Meter Data Cleaning- Clean and preprocess power data to analyze energy

 

Assessment Weightage:

IAT 1: CO1 and CO2 IAT 2: CO3 and CO4

IAT 3: CO4 and CO5

 

References:

1. Le Xie, Yang Weng, Ram Rajagopal (2023). Data Science and Applications for Modern Power Systems. Springer, Power Electronics and Power Systems Series. ISBN: 978-3-031-29100-5.

2. S. Vijayalakshmi, Lekha J., Lija Jacob, Savita Dahiya, R. Gunavathi (Eds.) (2025). Smart Power Systems: Grid Modernization Using AI and IoT-Based Applications. Springer, Cham. ISBN: 978-3-031-89459-6.

3. B. Vinoth Kumar, P. Sivakumar, M.M. Rajan Singaravel, K. Vijayakumar (Eds.) (2021). Intelligent Paradigms for Smart Grid and Renewable Energy Systems. Springer. ISBN: 978-981-15-9968-2.

4. Bhargav Appasani, Nicu Bizon (Eds.) (2023). Smart Grid 3. 0: Computational and Communication Technologies. Springer. ISBN: 978-3-031-38506-3.