Intelligent Online Condition Assessment of Navy Electrical Insulation
Abstract
Approved for Public ReleaseProject AbstractProposal Title: Intelligent Online Condition Assessment of Navy Electrical InsulationPI:,Mona Ghassemi, Virginia TechResearch Problem and Objectives, Impact on DoD Capabilities: Statistics indicate that more than 85% of e,q,ts; thus, PD is widely regarded as one of the best early warning indicators of component failure. The use of emerging wide bandgap, (WBG) switches in integrated power electronics building blocks (iPEBBs) leads us closer to the high power density goal of the Navy,s Power Electronic Power Distribution System (PEPDS). One of the merits of WBG devices is that their slew rates (ranging from tens t,o hundreds of kV/s) and switching frequencies (in hundreds of kHz range) are much higher than Si-based devices. However, frequency,and slew rate are two of the most critical factors of a voltage pulse that influence the level of degradation of an insulation syste,m exposed to such voltage pulses. The shorter the rise time, the larger the PD magnitude; thus, the shorter the lifetime. Also lifet,ime decreases with increasing frequency. This, combined with higher electric fields, higher temperatures, and harsh environments, su,ch as high humidity and salt pollution, will cause extreme damage to Navy insulation systems, leading to accelerated aging and degra,dation and eventually failure of PEPDS insulation systems. For the Navy, such failures can impact the safety of sailors or jeopardiz,e missions, which is highly unacceptable.A condition assessment system to monitor the health of Navy electrical insulation systems i,s needed and PD activity could be used as an indicator for the system. For the Navy PEPDS, online PD assessment is preferable (inste,ad of off-line assessment) since it does not lead to power outages or disturbance to system operation. The proposed online PD assess,ment system will achieve three goals: detection, recognition, and localization of PDs. There are huge technical gaps for all three g,oals and thus three related thrusts are defined for this project. Our system will cover all components in the power corridor in the,Navy PEPDS that are prone to insulation degradation including power cables, cable terminations, iPEBBs, motors/ generators, and ener,gy storage systems.Technical Approaches and anticipated outcome of the research: In Thrust I for PD detection and measurement, we pr,opose to design, build and test ultra-wideband (UWB) antennas for detection of PD-induced radiation under high frequency (0-500 kHz),, high slew rate (adjustable rise times ranging from 50 to 800 ns for voltages up to +/-6 kV) voltage pulses. The designed antenna w,ill meet the desired requirements for gain, voltage standing wave ratio (VSWR), axial ratio, and radiation pattern. Antenna characte,ristics in both the near-field and far-field regions will be assessed. Design optimization willbe performed to achieve the best poss,ible performance. We will also develop filters and signal processing techniques to remove noise from PD signals in this thrust. In T,hrust II to separate and recognize i) multi-type, ii) multi-source and iii) multi-excitation of the corona, internal, and surface PD, signals, novel deep learning-based techniques including various models of AlexNets, Inception-V3, ResNets, DenseNet will be develop,ed and optimized. In Thrust III for PD localization, we will develop and optimize a new, efficient method based on electromagnetic t,ime reversal (EMTR) using the transmission line matrix (TLM) to locate PDs in Navy PEPDS cables considering losses of cable, inhomog,eneous cables, and branched networks.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 16, 2022
- Source ID
- N000142212037
Entities
People
- Mona Ghassemi
Organizations
- Office of Naval Research
- United States Navy
- Virginia Tech