An Investigation of Harmonic Fingerprints by using Artificial Neural Network in Distribution Grid
Konferenz: NEIS 2019 - Conference on Sustainable Energy Supply and Energy Storage Systems
19.09.2019 - 20.09.2019 in Hamburg, Deutschland
Tagungsband: NEIS 2019
Seiten: 5Sprache: EnglischTyp: PDF
Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt
Autoren:
Song, Xinya; Cai, Hui; Zeng, Y. L.; Jiang, Teng; Schlegel, Steffen; Westermann, Dirk (Power Systems Group, Technische Universität Ilmenau, Germany)
Inhalt:
This paper provides a new algorithm to enhance the observability of distribution grid by using artificial neural network (ANN) in a distribution grid. Each electrical device has their unique harmonic nature, which could be used as fingerprint for identification. Training with this characteristic and the corresponding power profile ena-bles the ANN to identify the electrical devices and estimate their power profiles. The training is based on massive harmonic measurement, which carry out with self-developed low voltage (LV) measurement unit with WESENS-adapter form A-EBERLE and medium voltage (MV) measurement unit with DEWESOFT. This algorithm is validated with a field-test with 34times LV Unit and 3times MV unit in center Germany.