**5. Conclusions**

The D/I methodology with open-air capacitive sensor is capable of recording switching transients in an EMC harsh environment. However, the decoupling inherent to open-air sensors is a tedious process and may limit its application to specific situations:


The need of pre-assumptions is an obvious drawback when dealing with significant cross-coupling, which may especially occur at large substations. Future research is directed to reduce the number of assumptions to be made in establishing the coupling matrix. Complete information can be extracted by employing the responses from different sensors upon the front of a switching event (as in [4]) in combination with information provided by the power frequency responses as presented here. Having the topologies of Figure 3 in mind, upon an initial travelling wave on one of the lines, only the response ratios of neighbouring sensors are needed. There are four combinations which, together with information from power frequency fitting, would completely determine the coupling matrix.

**Author Contributions:** F.B. and P.W. prepared and performed the experiments, designed the analysis methods and wrote the paper. F.B. performed the power system modelling in PSCAD. F.S. provided technical feedback and reviewed the paper.

**Funding:** This research was funded by TenneT TSO B.V. within the framework of the Randstad380 cable research project.

**Acknowledgments:** TenneT is acknowledged for performing the switching actions in their 380 kV grid. Movares Nederland B.V. is acknowledged for providing their set of integrators with EMC shielding. Marcel Hoogerman, Hennie van der Zanden, Frank Beckers and Armand van Deursen from Eindhoven University of Technology are acknowledged for assisting with the measurement campaigns and their preparations.

**Conflicts of Interest:** The authors declare no conflict of interest.
