More information
Description
Renel IKE provides solutions for energy and electromechanical projects and services.
Keeping the power transmission infrastructure well-functioning and without faults is crucial for the reliable electricity transmission day-to-day. This is achieved through inspections followed by preventative measures and/or repairs if/when needed, depending on the findings. Until now, the inspections are conducted manually, relying on human resources. Mapping and monitoring of the powerline infrastructure and, more specifically, High and Medium Voltage network, is carried out by inspection workers who need to be on-site (walk patrols) and fill out specific survey-based questionnaires that are designed to assess the condition of the network.
The ALTITUDE project aims to develop an innovative solution to perform inspections of the aerial network automatically, using Unmanned Aerial Vehicles (UAVs - drones) and Machine Learning (ML) and AI algorithms. Using drones will make the inspections less time consuming, more reliable and safer for personnel involved. For this project, Renel has partnered with two Greek entities: Innora, a mico-enterprise that will be responsible to develop the communication protocol used by the UAVs, and the Hellenic Electricity Distribution Network Operator (HEDNO), who will be the pilot user of the solution within the project.
The project will lead to increased competitiveness of enterprises registered in Greece. Through the project, 2 new jobs will be created and I new Intellectual Property Rights application is expected to be submitted.
Summary of project results
The project ALTITUDE aimed to develop a web-based tool for the inspection and maintenance of medium voltage (MV) power networks using drone imagery and artificial intelligence (AI). The main objectives were to improve the efficiency, accuracy, and safety of the inspection process, as well as to reduce the costs and environmental impact of the maintenance operations.
- Web platform and AI functionalities: The project developed a web-based Geographic Information System (web GIS) that allows users to upload drone imagery of medium voltage (MV) structures and apply a deep learning algorithm for the detection of defects. The system consists of several parts, such as the web site, the processing application, the database, and the map server. The system can perform AI inference for various faults, such as tilt, missing disks, and thermal anomalies (overheating).
- Digitization and data visualization: The project also enabled the digitization of MV network structures and power lines, by creating a digital twin of the network elements. The digitized data can be visualized on different basemaps and orthomosaics, and can be filtered and downloaded by a user. The uploaded images can also be displayed on the map and annotated for AI training
The project involved four main tasks: drone data acquisition, web platform creation, power network mapping, and testing and evaluation. The project outputs included a drone communication protocol, a web GIS system, an AI algorithm for defect detection, and an inspection reporting functionality.
The project will assist network operators in the mapping and inspection of the medium voltage power networks as well as the identification and prevention of faults. It will contribute in increased speed as well as lower cost of such inspections, as well as increased safety for the relevant personnel involved in such activities. The project also led to the creation of 2 new permanent positions for the Project Promoter. 1 Intellectual Property Rights application for the software and algorithms developed is also planned to be submitted by the Project Promoter shortly after project completion.
Lat but not least, the commercialisation of the developed solution will contribute to increased profitability (turnover and net operating profit) for the Project Promoter.