Vietnamese engineer wins first prize at artificial intelligence competition in Hong Kong

A group of three Vietnamese engineers has just won first prize at the Global AI Challenge for Building E&M Facilities held in Hong Kong (China). The award ceremony took place at the Hong Kong Science Park on August 8.

This year's competition attracted more than 200 teams from 26 countries and regions, taking place from June to August, both online and in person.

The competition has two categories, “Developing a General AI Model to Predict Cooling Demand for Multiple Buildings” and “Proposing Innovative Solutions in the Construction and Engineering Industry”. This will be a platform for researchers, students, AI experts, startups and industry professionals to exchange ideas and showcase how advanced AI technology can drive the development of the construction and engineering industry.

Mr. Nguyen Hoang Vu, AI engineer and CEO of start-up company AIZ, said that according to information from the competition's organizers, buildings are the largest energy consumers globally, accounting for more than 40% of total energy use.

In Hong Kong, heating, ventilation and air conditioning (HVAC) systems alone in buildings cost about HKD 12.3 billion (over US$1.5 billion) in electricity each year.

The competition aims to build AI models that can predict building cooling loads, helping to save operating costs and easily deploy in many different types of buildings.

Faced with this requirement, the Vietnamese engineering team focused on an approach that can be flexibly deployed on many types of devices, including those with limited computing resources, while still meeting quality requirements.

Engineer Luong Duc Long ( Viettel Group) shared that in order to solve the problem of building a model to support predicting the cooling capacity of buildings, the group tested many different models, from which they selected the best model that could solve this problem.

By participating in the competition, the team had the opportunity to learn more knowledge and problem solving methods from their peers in many countries. They also hope to be able to apply this problem to Vietnam, to help reduce the huge energy consumption in buildings.

Engineer Dang Quang Minh (VCCorp Company) said that besides choosing the right model, the difficulty of the competition is that the provided data sets are limited by lack of data, or incorrect data due to problems with building sensors or limitations due to new buildings.

Therefore, the team also spent most of the competition time processing and analyzing the raw data of the buildings provided by the organizers. This contributed greatly to the team's final results.