Harmonic Current Emission Estimation of Modular Multilevel Converters in MV Charging Stations

Abstract

This paper proposes a novel framework for modular multilevel converter (MMC) current harmonic emission estimation from medium voltage (MV) charging stations. The framework is enlightened by wave decomposition, exploiting an analytical derivation of the emission, without a time-domain model. Accordingly, voltage harmonic distribution in grids can be estimated with a frequency-domain model, using the estimated voltage-dependent current emission in the iterations. Compared to conventional timedomain simulation, such method not only enables analytical harmonic estimation integrated with a large load-flow model, but also benefits from the potential of distributed computation. A 10.5 kV grid-tied single-phase half-bridge MMC simulation case on PLECS has been implemented and investigated for comparison.Corresponding results revealed that the estimated harmonic series kept the absolute average difference to the PLECS model lower than 0.02% and a correlation ratio higher than 99.990% with the simulated ones. The effectiveness of the proposed estimation framework is successfully verified.

Constraint Concentration in Distribution Networks

“…The DNC-e is non-convex and difficult to process for Gaussian elimination. Since the accuracy of LinDistFlow model has been numerically verified acceptable as tested in [4], we use this convex variant as an approximated network model, thus contributing to DNC-c1 as written below.…”

“…For further illustration, E j denotes the set of links in the shortest path from POC j to the slack bus. Intermediate parameters l max ij , ∆l ij , ∆P ij , ∆Q ij can be calculated as (4).…”

“…To verify the numerical efficacy of the proposed scheme, a Dutch network model depicted in Fig. 2 is employed for testing, whose parameters are provided in [4]. Bus 9 is the slack bus.…”

“…A brute-force concentration approach involves computing the multidimensional feasible region of critical variables, alias "hosting capacity region" (HCR), which advances the concept of "hosting capacity" [4], [5]. Leveraging computational geometry techniques, this region can be assessed by an expansive polytope.…”