Analiza stabilnosti elektroenergetskega sistema za majhne motnje s pomočjo Pronyjevega algoritma

Hodnik, Gašper (avtor), Bizjak, Grega (mentor)

Sodobni elektroenergetski sistemi (EES) delujejo v dereguliranih razmerah. Nepredvidljivo obnašanje uporabnikov, povečana poraba in zniževanje investicij v opremo silijo elektroenergetske sisteme v obratovanje v razmerah za katere niso bili načrtovani. Za zagotavljanje varnega in zanesljivega obratovanja EES se namenja veliko sredstev za izboljšanje nadzora in krmiljenja. Dejstvo, da sistemi obratujejo na robu meja stabilnosti, pomeni, da je stabilnost EES eno od ključnih področij preučevanja. Magistrsko delo se ukvarja s preučevanjem t. i. majhnih motenj v sistemih oz. s t. i. stabilnostjo za majhne motnje. Stabilnost za majhne motnje lahko preučujemo na dva načina, in sicer z analizo lastnih vrednosti, ki temelji na linearizaciji sicer nelinearnih enačb, ki opisujejo elektroenergetski sistem. Alternativa temu pristopu je analiza signalov veličin, ki jih zajamemo neposredno iz sistema. Za analizo lastnih vrednosti potrebujemo poleg enačb tudi podrobne podatke elementov sistema, tj. njihove impedance ter parametre raznih regulatorjev. Ker so izračuni, sploh za večje sisteme, dokaj kompleksni in obsežni, si navadno pomagamo s programskim paketom, ki omogoča tovrstne izračune. Za potrebe magistrske naloge je bil uporabljen program PSS® NEVA (v nadaljevanju »Neva«), ki je sestavni del programa PSS® NETOMAC (v nadaljevanju »Netomac«) proizvajalca Siemens AG. Analizo različnih signalov, ki se pojavljajo v elektroenergetskem sistemu, pa smo izvedli s Pronyjevim algoritmom, ki je bil napisan v obliki Matlabove funkcije. Ta funkcija je ključni del razvitega programa v Matlabu, ki sicer omogoča avtomatsko analizo signala v časovnem prostoru, ter prikaz frekvenčnega spektra z grafi in tabelami. Naloga je razdeljena na več poglavij. V uvodu je na kratko opisana problematika elektromehanskih nihanj. Sledi poglavje s podrobnejšim opisom Pronyjevega algoritma ter prikaz preprostega izračuna z nakazanimi potencialnimi težavami. V tretjem poglavju je opisan potek dela v posameznih programskih okoljih s poudarkom na razviti programski kodi v Matlabu. V četrtem poglavju so predstavljeni štirje primeri frekvenčne analize signalov s pomočjo več razvitih programskih kod v Matlabu, ki služijo različnim potrebam. Zadnje poglavje naloge je sklep.

• Jezik: Slovenski jezik
• Ključne besede: Pronyjev algoritem, stabilnost za majhne motnje, elektroenergetski sistem, Matlab
• Vrsta gradiva: Magistrsko delo/naloga
• Organizacija: FE - Fakulteta za elektrotehniko
• Leto izida: 2016
• PID: 20.500.12556/RUL-84922
• Datum objave v RUL: 08.09.2016

Sekundarni jezik

• Jezik: Angleški jezik
• Naslov: Power system small signal stability analysis with help of Prony algorithm

Izvleček

Modern power systems operate in a deregulated environment. Unpredictable behavior of participants, increased consumption and reduced investments are forcing power systems to operate under conditions they were not designed for. To ensure a safe and reliable operation of power systems a lot of resources are being invested in better power system supervision and control. The fact that power systems are operating close to stability limits, makes power system stability one of such topics.This thesis deals with power system stability for small disturbances in the network, also known as small system stability (SSS). Small system stability can be assessed using two main approaches. The first one is eigenvalue analysis which relies on the linearization of nonlinear equations describing power system. The alternative approach is to analyze the signals gathered directly from a real power system. Not only equations, but also a lot of data from power grid elements, e.g. their impedances and parameters of their regulators, are needed to perform eigenvalue analysis. SSS calculations, especially for bigger power systems, are very complicated and time consuming, therefore use of dedicated software packages is essential. For the needs of this masters thesis PSS® NEVA, a part of PSS® NETOMAC from SIEMENS AG, was used. The analysis of several different signals, which can be gathered from power systems, was performed by so called Prony algorithm programmed as a Matlab function. This function is the key part of the code developed in Matlab. The whole program enables an automatic analysis of measured signal in time domain and presenting its frequency spectrum with the help of figures and tables. The thesis consists of several chapters. In the introduction, the issues involving electro-mechanical oscillations are briefly discussed. The Prony algorithm is explained in detail in the second chapter as well as an example of the whole analysis. Possible issues connected with the whole procedure are also discussed. In the third chapter, all stages in development of the masters thesis are presented. Despite the work being done using several different programs, the focus is on the developed program in Matlab. In the fourth chapter, four cases of frequency analysis using different Matlab scripts were used. Each of those are suitable for a particular structure of raw data and diferent requirements of the user. The last chapter is the conclusion.

• Ključne besede: Prony algorithm, small signal stability, power system, Matlab