AI based predictive maintenance for electrical energy systems

We present solutions for predictive maintenance by combining Virtual Sensors with powerful AI tools. Developed new models of the underlying devices that can run in real time and thus serve as Virtual Sensor fed by real operation data from the actual devices. The VS will monitor thermal, mechanical, and electrical stresses. The data from the VS will be used in failure models to predict the remaining lifetime of the devices allowing for fault-tolerant and overload usage of the said devices, as well as condition-based maintenance. This is possible if the models are used in combination with AI or machine learning engines running in the clouds.

Project Aim and Main Tasks

The main goal of the proposal is to develop a specialized unsupervised diagnosis and prognosis platform for electrical energy systems.

Main Tasks

  1. Development of physical models of different energy system components
  2. Implementation of Virtual Sensors based on the Digital Twin concept
  3. Development of an Artificial Intelligence-based system for predictive maintenance
  4. Development of a small-scale demonstrator of the above concepts
  5. Development of a power electronic converter with sensor-fault-tolerant control algorithm
  6. Development of self-monitoring of power electronic converter with self-detection of parasitic components

Results

  • 6 journal papers (CA WoS)
  • 3 article-based doctoral theses
  • 6 other journal papers
  • 10 IEEE conference papers

Team (VILNIUS TECH)

  • prof. dr. Algirdas Baškys
  • prof. dr. Artūras Serackis
  • assoc. prof. dr. Raimondas Pomarnacki

Team (University of Adger)

prof. dr. Van Khang Huynh

Team (RTU)

dr. Jānis Zaķis

Team (TalTech)

prof. dr. Anton Rassõlkin

assoc. prof. dr. Toomas Vainmann

Project principal investigators

Project is coordinated by Vilnius Gediminas Technical university (VILNIUS TECH). The consortium includes University of Adger, Tallin University of technology (TalTech), and Riga Technical University (RTU).