Actively research and apply artificial intelligence in the management and operation of the power grid.

Over the years, Quang Tri Power Company (PC Quang Tri) has actively researched, developed, and applied science and technology to the fields of automation in operation, inspection, and evaluation of power grid quality, investment management, construction, and customer service, contributing to improving the company's production and business activities.

Some images of power grid safety hazards were detected by drones/UAVs - Photo: TN

Quang Tri Power Company is a pioneering unit within the Vietnam Electricity Group (EVN) in researching and developing a software integrated with field information and power grid technical management software. Notably, the company has conducted research into artificial intelligence (AI) in automation to detect abnormal phenomena in power lines and substations on the operating power grid using imagery.

These include programs such as: automatically checking objects of interest in images captured and stored on the EVN-IMIS investment management system. This program has automated the inspection and analysis of images taken annually at investment projects; or the artificial intelligence application program that automatically detects abnormal temperatures from images captured from live equipment on the power grid. The program automatically analyzes and issues warnings, helping technical staff to take appropriate measures to address these anomalies and prevent potential electrical incidents.

In 2022, Quang Tri Power Company researched and implemented an AI application to detect potential power grid safety hazards from images/ videos collected by drones. Although the power sector has implemented many programs to support power grid management and operation, such as the Project Management Information System (PMIS) and medium-voltage field inspection (KTHT) software, with the aim of digitizing the inspection of power lines and substations, the detection of problems based on images from PMIS and KTHT programs is still largely done visually.

This method is very time-consuming in detecting hazards from images and videos. Therefore, images and videos collected from drones/flycams are synchronized into the PMIS-AI program and automatically analyzed to detect potential power grid safety hazards, replacing the work of workers performing visual inspections with binoculars. Thus, the application of AI models in detecting power grid safety hazards from images/videos collected by drones has yielded positive results in power grid operation and management.

To ensure the system operates with high accuracy, in addition to building models, normalizing data, labeling objects, and training the object recognition program, the company has applied the YOLV5 modeling solution to the PMIS-AI program.

With this model, the processing time for a 4MB image takes only 1/10 of a second. Therefore, Quang Tri Power Company is a pioneer in participating in research in this field, especially with many proposed solutions for widespread implementation. A prime example is the application of artificial intelligence in image recognition during construction investment management, and the automatic recognition of thermal cameras for units under the Central Power Corporation, which has been highly praised and effectively applied in practice.

In 2024, the project "Research and application of artificial intelligence to detect power grid safety risks from images/videos collected by drones/UAVs during flight missions" by the team of authors: Master Phan Van Vinh, Nguyen Van Tai, Le Cong Hieu, Le Van Minh, and Nguyen Xuan Thuy from Quang Tri Power Company won second prize at the 17th National Technical Innovation Competition (2022-2023) organized by the Vietnam Union of Science and Technology Associations and the Vietnam Fund for Technical Innovation Support (VIFOTEC) in the field of information technology, electronics, and telecommunications.

With its AI application solution for detecting power grid safety risks from images/videos collected by drones/UAVs programmed with automatic flight paths according to flight missions, the Quang Tri Provincial People's Committee's AI recognition software combines data analysis to provide warnings and detect power grid safety risks from images/videos collected by flying drones.

The application utilizes the Yolov8 artificial intelligence model and other supporting tools (LabelMe for labeling, Google Colab for training) to detect abnormalities/anomalies in 110kV and 22kV power lines through images and videos collected from drones/flying cameras. Specifically, it focuses on detecting frayed bare conductors, loose insulator ties, dirty, broken, or cracked insulation, and other unusual objects on the power grid.

Automated flight path programming for drones flying over power grids is an advanced technology in the field of power grid safety and efficiency monitoring. The system is designed to monitor the power grid automatically and continuously, providing comprehensive information to detect potential power grid safety hazards. This solution enhances the effectiveness of power grid safety monitoring; saves costs; reduces monitoring expenses; increases accuracy; improves operational efficiency; and minimizes time and manpower.

With the aim of maximizing the power of digital technology to improve efficiency in technical management and ensure safe operation of the power grid, the research and application of AI in technical management is an inevitable trend. This will greatly contribute to increasing labor productivity and the efficiency of power quality management. From there, it will provide a stable and safe power supply to serve the socio-economic development of the locality.

Tan Nguyen

Source: https://baoquangtri.vn/tich-cuc-nghien-cuu-ung-dung-tri-tue-nhan-tao-trong-quan-ly-van-hanh-luoi-dien-189890.htm