The potential of electric reflectometry

At CEA-List we have done extensive research on electric reflectometry, a non-destructive testing method, for cable diagnostics. It works by injecting a signal into the system or medium being tested and recording the echo reflected back to the injection port at each impedance discontinuity. The echoes can then be analyzed to glean information about the impedance profile.

We have developed non-intrusive reflectometry methods for embedded in-line diagnostics. These methods, which include MCTDR (multi-carrier time domain reflectometry) and OMTDR (orthogonal multi-tone time domain reflectometry), do not interfere with the native signals of the system being tested. We have integrated these methods into complete systems on low-cost circuit boards with the necessary injection, acquisition, and processing capabilities. A system can be probed in under a millisecond, fast enough to detect an electric arc in an aircraft during flight.

Signal processing and AI have been integrated to further improve the performance (fault location, accuracy, sensitivity) of the system. The goal is to identify end-of-life indicators and be able to determine how faults will evolve in order to assess remaining cable lifetime the PTC Viper project for predictive maintenance purposes, the subject of a PhD dissertation underway. The reliability of the diagnostics was also assessed in the presence of vibrations for a space application in research with France’s space research center, CNES (Figure 1), and under high temperatures for a nuclear application.

CEA-List leverages extensive expertise in cable diagnostics to bring electric reflectometry to a wide range of other use cases:

• Reflectometry was determined feasible for the structural health monitoring of hydrogen tanks in research with Tier 1 automotive supplier Faurecia.
• Electric reflectometry was applied to an oil & gas use case (the measurement of water, gas, and oil present) in 2023. The technique’s capacity to detect the interfaces of media with different dielectric permittivities was tested on a wellbore model with a complex architecture and severe operating conditions (debris).
• The technique was also applied to energy storage system monitoring, with lithium-ion battery characterization and modeling at high frequencies from 1 MHz to 300 MHz (Figure 2) completed in 2023. A PhD dissertation underway in 2024 will deepen this research and bring in AI to estimate remaining battery pack life.

 

• The topology of a multi-branch energy network was reconstructed, a capability that can be used either to detect physical intrusions (energy theft) or to generate a wiring diagram of a building, for example. A new iterative algorithm was patented and deployed on test houses at CEA-Liten’s INCA lab in Le Bourget-du-Lac (PTC MOCA, Figure 3), where correct reconstructions were obtained for 90% of measurements in under an hour (compared to state-of-the[1]art convergence times of 24 hours).

• In research conducted for the Secure-HOP project (financed by Bpifrance), a low-cost prototype in the form of a new circuit board integrating the compressed OMTDR reflectometry chain (Figure 4) and a heat-sensitive cable, was produced. A patent for the emission architecture developed was filed. The system is estimated to cost 25% less than thermal and gas detection for aeronautics applications.

• The technique was also used to detect quench (sudden unexpected state changes in superconducting coils). Specifically, a method was developed to rapidly detect quench based on real-time reflectrometry-based observation of superconductivity. The method was patented (in 2022 as a result of the R4QUENCH project with the CEA’s fundamental research division).

• The technology in use: WIN MS, a CEA-List startup created in 2012, is commercializing aeronautics maintenance and railway cable monitoring solutions leveraging CEA-List MCTDR technology.
• Patents: CEA patents for network topology reconstruction1 and multi-level measurement are being leveraged in partnerships in the energy and oil & gas industries. Finally, the Bpifrance-financed Secure-HOP project resulted in a prototype of a low-cost, embedded OMTDR reflectometry system for fire detection.