Projects per year
Abstract
Systematic measurements of harmonic voltage and current are becoming more prevalent due to the increasing issue of excessive harmonic values in distribution and transmission grids. 10-minute aggregated values are used for post-analysis and harmonic assessment, but also to investigate if limit values for specific harmonic orders have been exceeded, identify locations for mitigating measures, and assess customer connections in the grid. They also assess the accuracy of complex harmonic simulation models, primarily used by Transmission System Operators.
The utilisation of historical harmonic distortion measurements is examined in this paper, focusing on measured 5th, 7th, 11th, and 13th harmonic voltages. To identify probable relationships between individual variables, the study uses multivariable data analysis. For exploratory analyses, principal component analysis is used. As a case study, measurements from the Western Danish 400 kV transmission grid are used. A set of guidelines is provided for conducting exploratory analyses, as well as a thorough demonstration of the interpretation of the results from the tool used.
It has been demonstrated how multivariate data analysis can be applied on harmonic voltage measurements in interaction with the operational properties of harmonic sources, thus explaining the correlation between source contributions and specific substations in the grid. The impact of other variables, such as operation state of transmission lines, can also be revealed when multivariate methods are used. These relationships could not be easily found with the help of univariate methods.
The utilisation of historical harmonic distortion measurements is examined in this paper, focusing on measured 5th, 7th, 11th, and 13th harmonic voltages. To identify probable relationships between individual variables, the study uses multivariable data analysis. For exploratory analyses, principal component analysis is used. As a case study, measurements from the Western Danish 400 kV transmission grid are used. A set of guidelines is provided for conducting exploratory analyses, as well as a thorough demonstration of the interpretation of the results from the tool used.
It has been demonstrated how multivariate data analysis can be applied on harmonic voltage measurements in interaction with the operational properties of harmonic sources, thus explaining the correlation between source contributions and specific substations in the grid. The impact of other variables, such as operation state of transmission lines, can also be revealed when multivariate methods are used. These relationships could not be easily found with the help of univariate methods.
Original language | English |
---|---|
Title of host publication | CIGRE Paris Session 2024 |
Publisher | CIGRE (International Council on Large Electric Systems) |
Publication date | 2024 |
Article number | C4-11649-2024 |
Publication status | Published - 2024 |
Event | Cigre Session 2024 - Paris, France Duration: 25 Aug 2024 → 30 Aug 2024 |
Conference
Conference | Cigre Session 2024 |
---|---|
Country/Territory | France |
City | Paris |
Period | 25/08/2024 → 30/08/2024 |
Series | CIGRE Science & Engineering |
---|
Keywords
- Correlation analysis
- Electrometrics
- Harmonic distortion
- Harmonic measurements
- Harmonic voltage
- meshed grid
- Principal Component Analysis
- PCA
- Statistical Analysis
Fingerprint
Dive into the research topics of 'Exploratory Analyses of Power System Harmonic Measurements Using Principal Component Analysis'. Together they form a unique fingerprint.Projects
- 2 Finished
-
DANPAC 2020
Bukh, B. S. (PI), Bak, C. L. (PI), Faria da Silva, F. M. (Project Participant) & Hansen, C. (Project Manager)
01/09/2020 → 31/08/2023
Project: Research
-
Methods for harmonic analysis in meshed transmission systems
Bukh, B. S. (PI)
01/09/2020 → 31/12/2023
Project: PhD Project
Research output
- 1 PhD thesis
-
On the Propagation of Harmonics in Meshed Transmission Power Systems with a Large Number of Underground Cables
Bukh, B. S., 2023, Aalborg Universitetsforlag. 346 p.Research output: PhD thesis
Open AccessFile356 Downloads (Pure)