Grid Voltage Frequency Estimation Using an Adaptive Complex Master-Slave Unscented Kalman Filter Public Deposited

http://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/r207ts24b

Descriptions

Attribute NameValues
Creator
Abstract or Summary
  • Due to the popularity of electronically controlled loads and the widespread use of alternative energy sources such as wind turbines and solar cells, the power quality at the distribution level must be carefully monitored. One of the many different means to monitor power quality is through frequency measurements. Nonlinearities cause harmonic generation effectively distorting the signals seen at the load, and so this seemingly simple task becomes a challenge. Recent advances in digital computing have allowed powerful linear estimators like the Kalman filter (KF) to be widely implemented for frequency estimation. In real physical applications, systems exhibit nonlinear behavior raising the need to adapt the Kalman Filter to fit nonlinear models. The Extended Kalman Filter (EKF) deals with nonlinearities in the system by linearizing the model around a known state. However, this makes the estimation process inaccurate because second order or higher nonlinearities are neglected. The recently developed Unscented Kalman filter (UKF) takes advantage of the Unscented Transformation (UT) to deal with nonlinear models without the need of linearization and with the same computational complexity as the EKF. However, any variations of the Kalman filter exhibit a very similar robustness problem when modeling uncertainty and are also sensitive to initial conditions. In order to overcome these limitations, an improved UKF algorithm based on the theory of strong tracking filters (STF) has been developed. If tuned properly, this new algorithm improves tracking for sudden changes and avoids divergence. However, if the process and/or measurement noise change with time, the filter will not estimate with the same accuracy it was tuned for. This thesis work proposes an adaptive algorithm based on an Unscented Kalman Filter (UKF), in order to improve the frequency estimation of power signals which undergo changes and are corrupted by white noise. The adaptive algorithm is based on a Master-Slave configuration, where the "master" estimates the state and the "slave", which operates in parallel, estimates the noise covariance matrix. Since the voltage signal is less distorted than the current signal, the former is employed to derive a complex state-space model to estimate the fundamental frequency. In order to evaluate the performance of the proposed algorithm, several simulations with synthetic data are implemented.
Resource Type
Date Available
Date Copyright
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Non-Academic Affiliation
Keyword
Subject
Rights Statement
Peer Reviewed
Language
Replaces
Additional Information
  • description.provenance : Submitted by Juan Pablo Munoz Constantine (munozjua@oregonstate.edu) on 2015-09-29T23:18:17Z No. of bitstreams: 2 license_rdf: 1370 bytes, checksum: cd1af5ab51bcc7a5280cf305303530e9 (MD5) MunozJuanP2015.pdf: 1066862 bytes, checksum: 65c76b2d7f51dfe5e2f2b584aa20b23b (MD5)
  • description.provenance : Made available in DSpace on 2015-10-02T23:24:18Z (GMT). No. of bitstreams: 2 MunozJuanP2015.pdf: 1061615 bytes, checksum: 3d34b0a19aaee8680b668a2f4e643a82 (MD5) license_rdf: 1370 bytes, checksum: cd1af5ab51bcc7a5280cf305303530e9 (MD5) Previous issue date: 2015-09-22
  • description.provenance : Approved for entry into archive by Julie Kurtz(julie.kurtz@oregonstate.edu) on 2015-09-30T18:17:23Z (GMT) No. of bitstreams: 2 MunozJuanP2015.pdf: 1061615 bytes, checksum: 3d34b0a19aaee8680b668a2f4e643a82 (MD5) license_rdf: 1370 bytes, checksum: cd1af5ab51bcc7a5280cf305303530e9 (MD5)
  • description.provenance : Submitted by Juan Pablo Munoz Constantine (munozjua@oregonstate.edu) on 2015-09-30T17:25:37Z No. of bitstreams: 2 MunozJuanP2015.pdf: 1061615 bytes, checksum: 3d34b0a19aaee8680b668a2f4e643a82 (MD5) license_rdf: 1370 bytes, checksum: cd1af5ab51bcc7a5280cf305303530e9 (MD5)
  • description.provenance : Approved for entry into archive by Laura Wilson(laura.wilson@oregonstate.edu) on 2015-10-02T23:24:18Z (GMT) No. of bitstreams: 2 MunozJuanP2015.pdf: 1061615 bytes, checksum: 3d34b0a19aaee8680b668a2f4e643a82 (MD5) license_rdf: 1370 bytes, checksum: cd1af5ab51bcc7a5280cf305303530e9 (MD5)
  • description.provenance : Rejected by Julie Kurtz(julie.kurtz@oregonstate.edu), reason: Rejecting to change the commencement date on the bottom of the title page from 2015 to read - Commencement June 2016. Also in the Table of Contents, chapter 5 shows it starts on page 37 but is actually 38. Chapter 6 shows page 63 but is 64 and the Bibliography should be changed to page 66. Everything else looks good. Once revised, log back into ScholarsArchive and go to the upload page. Replace the attached file with the revised file and resubmit. Thanks, Julie on 2015-09-30T16:31:29Z (GMT)

Relationships

Parents:

This work has no parents.

Last modified

Downloadable Content

Download PDF

Items