Unbalanced three-phase distribution system frequency estimation using least mean squares method and positive voltage sequence

• Published in: IET science, measurement & technology Vol. 8; no. 1; pp. 30 - 38
• Main Authors: Marcetic, Darko P, Tomic, Josif J, Kusljevic, Miodrag D
• Format: Journal Article
• Language: English
• Published: Stevenage The Institution of Engineering and Technology 01.01.2014; The Institution of Engineering & Technology
• Subjects: Algorithms; circular vector trajectory; CLMS algorithm; complex least mean squares algorithm; computer simulations; constant delay block; Electric potential; frequency estimation; grid‐connected power converters; harmonic distortion levels; high signal harmonics; LabVIEW software environment; Least mean squares; Least mean squares algorithm; least mean squares methods; low‐pass filters; Mathematical analysis; noise measurement; positive voltage phase‐sequence vector; power convertors; power distribution; power engineering computing; power grids; power harmonic filters; signal frequency feedback loop; three‐phase grid voltage; two‐phase αβ‐plane; Unbalance; unbalanced three‐phase electrical distribution system frequency estimation; Vectors (mathematics); virtual instrumentation; Voltage; voltage amplitude unbalance; least mean squares methods; noise measurement; three-phase grid voltage; power harmonic filters; harmonic distortion levels; unbalanced three-phase electrical distribution system frequency estimation; positive voltage phase-sequence vector; LabVIEW software environment; low-pass filters; voltage amplitude unbalance; power grids; complex least mean squares algorithm; constant delay block; power convertors; high signal harmonics; computer simulations; power engineering computing; frequency estimation; circular vector trajectory; virtual instrumentation; grid-connected power converters; power distribution; signal frequency feedback loop; CLMS algorithm; two-phase αβ-plane
• ISSN: 1751-8822; 1751-8830
• DOI: 10.1049/iet-smt.2012.0003

The subject of this study is a frequency estimation algorithm suitable for grid-connected power converters placed at a weak coupling point of a three-phase electrical distribution system. An upgraded version of the widely used complex least mean squares (CLMS) algorithm for frequency estimation is introduced to cope with different voltage amplitude unbalance and harmonic distortion levels, both frequently present in power system at distribution level. First, it is suggested that the CLMS algorithm uses only a positive phase-sequence component of voltage vector, the component that is inherently symmetrical and by cancelling the phase unbalance preserves the circular vector trajectory in a two-phase αβ-plane. This study shows that it is even possible to use the positive voltage phase-sequence vector extracted using a constant delay block, thus avoiding potential instability issues in the case of signal frequency feedback loop. Second, possible high signal harmonics and signal measurement noise are both removed using low-pass filters prior to CLMS algorithm deployment. Computer simulations and experiments are performed under a variety of conditions to validate the effectiveness of the proposed technique. Experimental results are achieved using the dataset sampled from the actual three-phase grid voltage at distributed level and with data processing done in the LabVIEW software environment.