What will you learn in this Internship
- Overview of Artificial Intelligence and Machine Learning and their relevance in engineering disciplines, particularly electrical engineering.
- Introduction to Python/R and key libraries such as NumPy, Pandas, Scikit-learn, TensorFlow, and Keras.
- Brief history, evolution, and real-world applications in electrical systems.
- Understanding the AI/ML pipeline: data collection → preprocessing → model building → evaluation → deployment.
- Importance of data quality in ML model performance.
- Techniques for handling missing data, normalization, and encoding.
- Feature selection vs. feature extraction.
- Principal Component Analysis (PCA), Mutual Information, and domain-driven feature engineering examples from power and energy data.
Introduction to AI/ML: Concepts and Tools
Data Preprocessing and Feature Engineering
- Overview of supervised (regression, classification) and unsupervised (clustering, dimensionality reduction) methods.
- Application of models like Linear Regression, Decision Trees, SVM, K-means, and Hierarchical Clustering in electrical load prediction.
- Case studies and datasets (e.g., smart meter data or SCADA data).
- Introduction to Artificial Neural Networks (ANNs), CNNs, and RNNs for electrical signal processing.
- Time-series analysis using LSTM for fault signal detection or waveform prediction.
- Use of spectrograms and frequency-domain features in classification tasks.
Supervised/Unsupervised Learning for Load Forecasting
Deep Learning for Signal Processing
- Role of AI in energy demand forecasting, real-time grid monitoring, and decentralized power management.
- Applications in dynamic pricing, demand response, and energy theft detection.
- Case examples involving IoT-enabled grid environments.
- Challenges in solar and wind energy prediction due to weather variability.
- Use of ML/DL models to predict power generation using meteorological and historical data.
- Model deployment for energy scheduling and grid stability.
AI in Smart Grids
Renewable Energy Forecasting
- Use of AI to analyze machine data (vibration, temperature, current) for anticipating equipment failure.
- Introduction to predictive maintenance frameworks: condition-based and reliability-centered maintenance.
- ML classifiers for remaining useful life (RUL) estimation.
- Data-driven approaches for fault classification and location.
- Applications in transmission line fault analysis, transformer monitoring, and circuit breaker performance.
- Comparison of rule-based systems vs. intelligent models (ANN, SVM, Random Forest).
Predictive Maintenance of Machines
AI-based Fault Detection in Power Systems
- Use of AI to analyze machine data (vibration, temperature, current) for anticipating equipment failure.
- Introduction to predictive maintenance frameworks: condition-based and reliability-centered maintenance.
- ML classifiers for remaining useful life (RUL) estimation.
- Fundamentals of metaheuristic optimization algorithms: Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Artificial Neural Networks (ANN) in engineering design.
- Multi-objective optimization for electrical network design, load dispatch, and capacitor placement.
- MATLAB/Python implementation examples.
