Implementation of Multi Objective - Modulated Model Predictive Control Into Virtual Synchronous Machine: Implementation of Multi Objective - Modulated Model Predictive Control Into Virtual Synchronous Machine
Translated title
Implementation of Multi Objective - Modulated Model Predictive Control Into Virtual Synchronous Machine: Implementering af Multi Objective - Modulated Model Predictive Control til Virtual Synchronous Machine
Author
Matijevic, Edi
Term
4. term
Education
Publication year
2019
Submitted on
2019-05-31
Pages
81
Abstract
Efterhånden som flere lande integrerer rene og vedvarende energikilder i elnettet, bliver decentrale løsninger som mikronet mere udbredte. En udfordring er, at mange vedvarende kilder kobles på via effektelektronik uden den mekaniske inerti, som traditionelle roterende generatorer har. Mindre inerti gør netfrekvensen mere følsom over for ændringer. En løsning er at styre konvertere, så de efterligner en synkronmaskine – en såkaldt virtuel synkronmaskine (VSM). Normalt styres VSM’er med kaskadekontrolløkker: en langsommere ydre spændingsløkke og en hurtigere indre strømløkke. Dette projekt udvikler modelprædiktiv styring (MPC), som samler flere styringsmål i én omkostningsfunktion. Derved kan antallet af særskilte løkker reduceres med henblik på at forbedre overgangsresponsen – altså hvor hurtigt og stabilt systemet reagerer på forstyrrelser. Den nye MPC sammenlignes med den traditionelle kaskadekontrol. Der gennemføres laboratorieforsøg med parallelle VSM’er i ø-drift (flere enheder, der kører sammen uden forbindelse til hovednettet), og resultaterne bruges til at vurdere de to metoder.
As more countries integrate clean, renewable sources into power grids, decentralized setups like microgrids are gaining traction. A key challenge is that many renewables are connected through power electronics and lack the mechanical inertia of traditional spinning generators. Lower inertia makes grid frequency more sensitive to disturbances. One way to address this is to control converters so they mimic a synchronous machine—a Virtual Synchronous Machine (VSM). Conventional VSMs use cascaded control loops: a slower outer voltage loop and a faster inner current loop. This project develops a Model Predictive Control (MPC) approach that combines multiple control objectives into a single cost function. This reduces separate control loops with the aim of improving transient response—that is, how quickly and smoothly the system reacts to changes. The new MPC is compared with the cascaded approach. Laboratory experiments are conducted with parallel VSMs operating in islanded mode (multiple units running together without the main grid), and the results are used to assess both methods.
[This abstract was generated with the help of AI]
Keywords
Documents