PowerNext-AI | Hackathon Phase
Problem Statements
The shortlisted teams are required to register for one of the three available problem statements. A brief abstract for each problem statement has been shared to help teams understand the challenge and make an informed selection.
Challenge Abstracts
AI-Driven Optimisation of High-Voltage Impulse Generator Settings for 400 kV Insulator Testing
High-voltage impulse testing is a critical part of validating the insulation performance of 400 kV power-system equipment against lightning and switching surges. However, determining suitable impulse-generator settings is often a time-consuming process that relies on engineering calculations followed by multiple trial shots to achieve waveform parameters within prescribed tolerances. Even minor changes in insulator configuration, measuring systems, or laboratory setup can significantly affect the generated impulse waveform. This challenge invites participants to develop an intelligent decision-support system that combines first-principles impulse-circuit modelling with machine learning to predict waveform characteristics and recommend optimal generator settings. Using generator, load, and test-bay parameters, the solution should estimate charging voltage, stage utilization, front resistance, and tail resistance while ensuring compliance with standard lightning (1.2/50 µs) and switching (250/2500 µs) impulse requirements. The outcome will enable faster test preparation, improved waveform accuracy, reduced trial-and-error effort, and a scalable framework for next-generation data-driven high-voltage testing laboratories.
Digital Twin for Intelligent Health Monitoring and Predictive Maintenance of Distribution Transformers
Distribution transformers form the backbone of electrical distribution networks and play a vital role in ensuring reliable power delivery to consumers. However, these assets operate under continuously changing conditions such as fluctuating loads, ambient temperature variations, voltage disturbances, harmonics, and ageing-related degradation, making them vulnerable to unexpected failures. Traditional maintenance approaches are largely schedule-based and often lack real-time visibility into equipment health. This challenge aims to develop a Digital Twin, a dynamic virtual representation of a physical distribution transformer, capable of continuously monitoring operating conditions and assessing asset health using electrical and thermal parameters. Participants are expected to combine data analytics, artificial intelligence, and predictive modelling techniques to estimate transformer health, detect anomalies, generate early warning alerts, and recommend maintenance actions. The solution should include an intuitive dashboard that visualizes transformer performance, health indices, fault indicators, and maintenance insights. The outcome has the potential to improve reliability, reduce downtime, optimize maintenance costs, and support the transition toward smarter and more resilient power distribution systems.
Intelligent Automated Test Report Generation for High-Volume Laboratory Operations
Testing laboratories generate thousands of test reports every year, making report preparation a resource-intensive process that often involves manual compilation of test results, verification of customer information, data formatting, and document generation. As testing activities are conducted across multiple locations, consolidating data from different sources into a standardized report format can lead to delays, increased workload, and the possibility of human errors. This challenge focuses on developing an intelligent web-based test report generation system that automatically collects test data, retrieves customer information from laboratory databases and portals, validates inputs, and generates reports in approved formats with minimal manual intervention. Participants are expected to design a scalable solution that combines database management, workflow automation, document generation, and analytics capabilities to streamline laboratory operations. The proposed system should improve report turnaround time, enhance data accuracy, reduce administrative effort, and provide a foundation for digital transformation of laboratory documentation processes, ultimately improving operational efficiency and customer satisfaction.
Registration Process & Rules
Once a team registers for a specific problem statement, detailed information, datasets, resources, and further instructions will be shared directly with the Team Lead via email.
Please Note the Following:
- 1Each team can register for only one problem statement.
- 2A maximum of 10 teams will be allowed per problem statement.
- 3Registrations will be accepted on a first-come, first-served basis.
- 4Once a problem statement reaches its capacity of 10 teams, no further registrations will be accepted for that challenge.
- 5Teams are encouraged to review the abstracts carefully and complete their registration at the earliest to secure their preferred problem statement.