doktorska disertacije
Rok Pajer (Author), Miran Rodič (Mentor), Amor Chowdhury (Co-mentor)

Abstract

Doktorska disertacija obravnava problem emulacij baterijskih paketov za uporabo v razvoju komponent za električna vozila (EV). Predstavljeni so vsi potrebni segmenti, od izbire in identifikacije modelov baterij, ki so primerni za izvajanje v realnem času, razvoja ustreznega pretvornika za emulacijo do razvoja emulacijskega algoritma. Predlagan emulacijski algoritem temelji na kaskadni strukturi, notranjo regulacijo tvorijo ločene regulacijske zanke vejnih tokov na osnovi drsnega režima. Zanke se izvajajo s konstantno stikalno frekvenco, mogoča je implementacija na digitalnih signalnih krmilnikih (DSP). Zaradi uporabe ocene motenj, dobljenih z diskretnimi opazovalniki motenj, je zagotovljeno sledenje drsni ravnini brez efekta drhtenja (angl. chattering). Tokovni regulatorji si delijo skupno referenco toka, zagotovljena je enakost tokov in dinamik, kar omogoča redukcijo sistema več vej pretvornika na sistem 2. reda. Zunanjo regulacijo zagotavlja napetostni regulator, ki ima vključen model emulirane baterije, kar omogoča replikacijo obnašanja baterije v frekvenčnem področju modela. Predkrmilje izhodnega toka v kombinaciji z opazovalnikom motnje, podobnim kot pri tokovni regulaciji, zmanjša vpliv bremena na obnašanje. Posebno je v nalogi obravnavan problem različnih bremen, predlagana so pravila nastavitve parametrov, ki zagotavljajo stabilnost pretvornika tudi ob priklopu bremen konstantne moči (CPL), ki so glavni porabniki v EV. Delovanje predlaganih algoritmov je potrjeno s simulacijami in z eksperimentalnim vezjem.

Keywords

emulacije baterij;vodenje v drsnem režimu;opazovalnik motnje;predkrmiljenje izhodnega toka;bremena konstante moči;electric vehicles;battery emulations;sliding mode control;disturbance observer;output current feedforward;constant power loads;CPL;

Data

Language: Slovenian
Year of publishing:
Typology: 2.08 - Doctoral Dissertation
Organization: UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher: [R. Pajer]
UDC: 004.057.7.021:681.51:621.314(043.3)
COBISS: 30498563 Link will open in a new window
Views: 946
Downloads: 141
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Battery dynamics emulation by control of a multileg switching type converter
Secondary abstract: The thesis covers the problem of the battery emulation for the purpose of development of the components for electric vehicles (EV). All the required components are presented, from selection and identification of battery models, appropriate for real time execution, the development of power converter to the development and implementation of emulation algorithm. The algorithm is cascade structured, with the inner current control loops in the sliding mode. Designed in discrete time, they are appropriate for constant frequency execution on the digital signal processor (DSP). Using estimated model disturbance from discrete disturbance observers, the currents are in sliding mode without the undesirable chattering effect. The loops share the same current reference and the stationary currents are made equal as well as current dynamics, which makes it possible to reduce system description to second order. The outer voltage control loop incorporates the emulated battery model and replicates a real battery characteristic in the model frequency range. Output current feedforward, in combination with the disturbance observer, similar as in the current loops, reduces the impact of load on dynamics. The problem of different loads is also investigated, especially of the constant power loads (CPL), which are the main ones in EV. The operation of proposed algorithms is verified with simulations and experimental circuit.
Secondary keywords: Električna vozila;Disertacije;Baterije;
Type (COBISS): Doctoral dissertation
Thesis comment: Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko
Pages: XXI, 141 str.
ID: 11810181