| dc.contributor.author | Brandštetter, Pavel | |
| dc.contributor.author | Skotnica, Martin | |
| dc.date.accessioned | 2007-08-14T08:08:30Z | |
| dc.date.available | 2007-08-14T08:08:30Z | |
| dc.date.issued | 1999 | |
| dc.identifier.citation | Computational Intelligence : Theory and Applications International Conference : 6th Fuzzy Days Dortmund, Germany, May 25-28, 1999. Proceedings. 1999, p. 220-224. | en |
| dc.identifier.isbn | 978-3-540-66050-7 | |
| dc.identifier.issn | 0302-9743 | |
| dc.identifier.issn | 1611-3349 | |
| dc.identifier.uri | http://hdl.handle.net/10084/61810 | |
| dc.language.iso | en | en |
| dc.publisher | Springer | en |
| dc.relation.ispartofseries | Computational Intelligence : Theory and Applications International Conference : 6th Fuzzy Days Dortmund, Germany, May 25-28, 1999. Proceedings | en |
| dc.relation.uri | http://dx.doi.org/10.1007/3-540-48774-3_25 | en |
| dc.title | Application of artificial neural network in control of vector pulse-width modulation inverter | en |
| dc.type | article | en |
| dc.identifier.location | Není ve fondu ÚK | en |
| dc.description.abstract-en | This paper describes neural controller that performs optimal switching strategy of VSI-PWM inverter. The strategy is based on rearranging the order of inverter output vectors in order to reduce the number of switched tranzistors during one time period, which leads to reduced power losses in switch tranzistors. Controller consists of one feedforward and one recursive neural network interconnected with computational blocks. The weights and thresholds of both neural nets are constructed rather than trained and require no adaption during the control process, which allows for very efficient implementation of the controller on digital signal processor. | en |
| dc.identifier.doi | 10.1007/3-540-48774-3_25 | |
| dc.identifier.wos | 000237352100025 | |