Estimation of Harmonics in Sparsely Monitored Distribution Networks

Accurate estimation of harmonics in partially monitored power networks with uncertain, power electronics interfaced low carbon technologies can facilitate efficient planning and operation of future net-zero distribution networks. A recent study demonstrated a methodology for estimating harmonic distortions in partially monitored, typically radial, residential distribution networks. This study extends the general applicability of the harmonic estimation method to higher voltage meshed networks. The Morris screening method based harmonic variation sensitivity analysis, and the sensitivity method based on electrical distance were proposed in this study and were combined together to determine the optimal/minimum number and location of PQ monitors. Appropriate scaling factors were recommended as well. This solves the problem of sub-optimal selection of monitoring locations when harmonic distortion is not highly correlated with the voltage drop at buses in the meshed network. Different types of harmonic injections from nonlinear loads are fitted using Kernel non-parametric distribution and are estimated separately according to different load types. The approach is validated, and its accuracy and reliability were demonstrated on a highly interconnected (meshed) section of the power system.