Weighting factor design based on Artificial Neural Network for Finite Set MPC operated 3L-NPC converter

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

1 Citation (Scopus)
17 Downloads (Pure)

Abstract

Optimum design of the weighting factors for a multi-objective cost function is one of the major challenges of Finite-Set Model Predictive Control (FS-MPC) operated power electronic converters. Especially for multi-level topologies, where multi-objectives must be included in the cost function to ensure a safe operation of the converter, the complexity of the optimization problem is rapidly growing with each new objective included in the cost function. In this paper a new approach for design of the weighting factors for a three level neutral point clamped (NPC) converter using artificial neural network (ANN) is proposed. The ANN is used as a surrogate model of the detailed converter model. In the first step a detailed converter model is simulated for different weighting factor combinations. From the simulations obtained performance metrics (e.g. total harmonic distortion (THD), average switching frequency, DC-link voltage balance) are used to train the ANN. Once the network is trained, it can be used to estimate the performance metrics for any combination of weighting factors. By defining a fitness function using the metrics, weighting factor combinations that optimize the function are found to be very fast. The design is also validated on an experimental set-up, where the measured performance metrics are compared to the ones predicted by the ANN. It is concluded that the results match very well with a difference being below 10%.
Original languageEnglish
Title of host publication34th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2019
Number of pages6
PublisherIEEE Press
Publication date24 May 2019
Pages77-82
Article number8722062
ISBN (Print)978-1-5386-8331-6
ISBN (Electronic)978-1-5386-8330-9
DOIs
Publication statusPublished - 24 May 2019
EventApplied Power Electronics Conference and Exposition APEC 2019 - Anaheim, United States
Duration: 17 Mar 201921 Mar 2019
http://www.apec-conf.org/

Conference

ConferenceApplied Power Electronics Conference and Exposition APEC 2019
CountryUnited States
CityAnaheim
Period17/03/201921/03/2019
Internet address
SeriesI E E E Applied Power Electronics Conference and Exposition. Conference Proceedings
ISSN1048-2334

Fingerprint

Cost functions
Neural networks
Model predictive control
Harmonic distortion
Switching frequency
Power electronics
Topology
Electric potential
Optimum design

Keywords

  • weighting factor design
  • Artificial Neural Network (ANN)
  • finite set model predictive control (FS-MPC)
  • 3L-NPC

Cite this

Novak, M., Dragicevic, T., & Blaabjerg, F. (2019). Weighting factor design based on Artificial Neural Network for Finite Set MPC operated 3L-NPC converter. In 34th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2019 (pp. 77-82). [8722062] IEEE Press. I E E E Applied Power Electronics Conference and Exposition. Conference Proceedings https://doi.org/10.1109/APEC.2019.8722062
Novak, Mateja ; Dragicevic, Tomislav ; Blaabjerg, Frede. / Weighting factor design based on Artificial Neural Network for Finite Set MPC operated 3L-NPC converter. 34th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2019. IEEE Press, 2019. pp. 77-82 (I E E E Applied Power Electronics Conference and Exposition. Conference Proceedings).
@inproceedings{ef82ab2d0ad44ef99ec20dde1b09b094,
title = "Weighting factor design based on Artificial Neural Network for Finite Set MPC operated 3L-NPC converter",
abstract = "Optimum design of the weighting factors for a multi-objective cost function is one of the major challenges of Finite-Set Model Predictive Control (FS-MPC) operated power electronic converters. Especially for multi-level topologies, where multi-objectives must be included in the cost function to ensure a safe operation of the converter, the complexity of the optimization problem is rapidly growing with each new objective included in the cost function. In this paper a new approach for design of the weighting factors for a three level neutral point clamped (NPC) converter using artificial neural network (ANN) is proposed. The ANN is used as a surrogate model of the detailed converter model. In the first step a detailed converter model is simulated for different weighting factor combinations. From the simulations obtained performance metrics (e.g. total harmonic distortion (THD), average switching frequency, DC-link voltage balance) are used to train the ANN. Once the network is trained, it can be used to estimate the performance metrics for any combination of weighting factors. By defining a fitness function using the metrics, weighting factor combinations that optimize the function are found to be very fast. The design is also validated on an experimental set-up, where the measured performance metrics are compared to the ones predicted by the ANN. It is concluded that the results match very well with a difference being below 10{\%}.",
keywords = "weighting factor design, Artificial Neural Network (ANN), finite set model predictive control (FS-MPC), 3L-NPC",
author = "Mateja Novak and Tomislav Dragicevic and Frede Blaabjerg",
year = "2019",
month = "5",
day = "24",
doi = "10.1109/APEC.2019.8722062",
language = "English",
isbn = "978-1-5386-8331-6",
series = "I E E E Applied Power Electronics Conference and Exposition. Conference Proceedings",
pages = "77--82",
booktitle = "34th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2019",
publisher = "IEEE Press",

}

Novak, M, Dragicevic, T & Blaabjerg, F 2019, Weighting factor design based on Artificial Neural Network for Finite Set MPC operated 3L-NPC converter. in 34th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2019., 8722062, IEEE Press, I E E E Applied Power Electronics Conference and Exposition. Conference Proceedings, pp. 77-82, Applied Power Electronics Conference and Exposition APEC 2019, Anaheim, United States, 17/03/2019. https://doi.org/10.1109/APEC.2019.8722062

Weighting factor design based on Artificial Neural Network for Finite Set MPC operated 3L-NPC converter. / Novak, Mateja; Dragicevic, Tomislav; Blaabjerg, Frede.

34th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2019. IEEE Press, 2019. p. 77-82 8722062 (I E E E Applied Power Electronics Conference and Exposition. Conference Proceedings).

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Weighting factor design based on Artificial Neural Network for Finite Set MPC operated 3L-NPC converter

AU - Novak, Mateja

AU - Dragicevic, Tomislav

AU - Blaabjerg, Frede

PY - 2019/5/24

Y1 - 2019/5/24

N2 - Optimum design of the weighting factors for a multi-objective cost function is one of the major challenges of Finite-Set Model Predictive Control (FS-MPC) operated power electronic converters. Especially for multi-level topologies, where multi-objectives must be included in the cost function to ensure a safe operation of the converter, the complexity of the optimization problem is rapidly growing with each new objective included in the cost function. In this paper a new approach for design of the weighting factors for a three level neutral point clamped (NPC) converter using artificial neural network (ANN) is proposed. The ANN is used as a surrogate model of the detailed converter model. In the first step a detailed converter model is simulated for different weighting factor combinations. From the simulations obtained performance metrics (e.g. total harmonic distortion (THD), average switching frequency, DC-link voltage balance) are used to train the ANN. Once the network is trained, it can be used to estimate the performance metrics for any combination of weighting factors. By defining a fitness function using the metrics, weighting factor combinations that optimize the function are found to be very fast. The design is also validated on an experimental set-up, where the measured performance metrics are compared to the ones predicted by the ANN. It is concluded that the results match very well with a difference being below 10%.

AB - Optimum design of the weighting factors for a multi-objective cost function is one of the major challenges of Finite-Set Model Predictive Control (FS-MPC) operated power electronic converters. Especially for multi-level topologies, where multi-objectives must be included in the cost function to ensure a safe operation of the converter, the complexity of the optimization problem is rapidly growing with each new objective included in the cost function. In this paper a new approach for design of the weighting factors for a three level neutral point clamped (NPC) converter using artificial neural network (ANN) is proposed. The ANN is used as a surrogate model of the detailed converter model. In the first step a detailed converter model is simulated for different weighting factor combinations. From the simulations obtained performance metrics (e.g. total harmonic distortion (THD), average switching frequency, DC-link voltage balance) are used to train the ANN. Once the network is trained, it can be used to estimate the performance metrics for any combination of weighting factors. By defining a fitness function using the metrics, weighting factor combinations that optimize the function are found to be very fast. The design is also validated on an experimental set-up, where the measured performance metrics are compared to the ones predicted by the ANN. It is concluded that the results match very well with a difference being below 10%.

KW - weighting factor design

KW - Artificial Neural Network (ANN)

KW - finite set model predictive control (FS-MPC)

KW - 3L-NPC

UR - http://www.scopus.com/inward/record.url?scp=85067130338&partnerID=8YFLogxK

U2 - 10.1109/APEC.2019.8722062

DO - 10.1109/APEC.2019.8722062

M3 - Article in proceeding

SN - 978-1-5386-8331-6

T3 - I E E E Applied Power Electronics Conference and Exposition. Conference Proceedings

SP - 77

EP - 82

BT - 34th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2019

PB - IEEE Press

ER -

Novak M, Dragicevic T, Blaabjerg F. Weighting factor design based on Artificial Neural Network for Finite Set MPC operated 3L-NPC converter. In 34th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2019. IEEE Press. 2019. p. 77-82. 8722062. (I E E E Applied Power Electronics Conference and Exposition. Conference Proceedings). https://doi.org/10.1109/APEC.2019.8722062