Registrations have closed.
Power Load Forecasting Using Machine Learning (STTP)
62 62 people viewed this event.
Session Highlights:
- Purpose:
- Equip faculty with practical knowledge on building machine learning models for load demand forecasting.
- Emphasize the significance of accurate forecasting in the power industry for improved efficiency and reliability.
- Methodology and Workflow:
- Data Collection and Preprocessing
- Exploratory Data Analysis (EDA) for understanding patterns and correlations
- Feature Engineering for enhanced model accuracy
- Model Selection and Evaluation, focusing on regression-based approaches
- Hyperparameter Tuning for optimizing model performance
- Model Deployment with Flask/Django for integration with utility systems
- Visualization and Reporting for effective presentation of forecasted results
Key Takeaways:
- Practical experience in developing, evaluating, and deploying machine learning models for power load forecasting.
- Understanding of time series forecasting techniques and their applications in the power sector.
- Insight into data visualization and reporting for effective decision-making.
Additional Details
Speakers - Ashish Shetty, Esha Prakash, Cezzane Khan