The share of electricity generation in the power system being based on power electronics is increasing, which will impact the system in different ways, such as an increased risk for undesired interactions. An example is doubly fed induction generator (DFIG) based windfarms which have been shown to present negative resistance in (parts of) the sub-synchronous range (i.e., below the system frequency of 50 or 60 Hz). If such a wind farm is radially connected (deliberately or not) to a series compensated line, undamped or poorly damped sub-synchronous oscillations could occur due to sub-synchronous resonance. One possible cause of such interactions is related to the wind farm control system, and in such cases, the interaction between the wind farm and system leading to sub-synchronous oscillations is referred to as sub-synchronous control interaction (SSCI). This thesis aims to describe different types of so-called sub-synchronous oscillations, with a focus on SSCI. An investigation is performed to find out under what circumstances there is a risk of SSCI, and how one can evaluate this risk. Different methods of obtaining the impedance of non-linear systems (e.g., a wind farm) are discussed, with the method used in this thesis being a dynamic impedance scan. The dynamic impedance scan is implemented in PSCAD and uses a voltage (or current) perturbation of one frequency at a time and measures the current (or voltage) response at that frequency, subsequently giving the impedance as the voltage/current ratio. Combined with the impedance of the grid, screening studies were performed to identify the risk of SSCI under different conditions. A 200 MW photovoltaic (PV) farm is designed and implemented in PSCAD. The PV farm is connected to the same bus as a 200 MW DFIG wind farm, resulting in a hybrid wind and PV farm. The hybrid wind and PV farm is in turn connected to a series compensated transmission system made to resemble a typical Swedish system. Functionality is added to the PV farm that enables it to act as a static synchronous compensator (STATCOM) to damp sub-synchronous oscillations if needed. Simulations are performed in PSCAD showing that the PV farm is able to damp the sub-synchronous oscillations occurring when the wind farm is radially connected with the series compensated line due to a fault, thereby avoiding disconnection or damage to equipment. One of the main conclusions is that assessing the risk of SSCI (screening) is not an exact science, but a highly complex matter. This conclusion is drawn due to contradictory implications given when analysing the measured grid and wind farm impedances. For example, a series resonance point in the combined reactance (grid + wind farm) would suggest that there would be oscillations at this frequency during a fault, but this may not always be the case. The opposite also occurred, i.e., oscillations of a certain frequency occurred even though no series-resonance point was seen in the combined impedance. Nonetheless, the screening method did manage to identify risk cases based on a set of criteria listed in the thesis, although electromagnetic transient analysis (EMT) time-domain simulations should always be performed for verification. The other main conclusion is that a PV farm installed at the point of common coupling (PCC) of a wind farm, i.e., a hybrid wind and PV farm, is able to damp sub-synchronous oscillations by acting as a PV-STATCOM. The use of combined assets, such as utilizing a PV farm to counteract SSCI in a wind farm, means that additional investments, for example in the form of a STATCOM, for this purpose could be avoided.
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:du-42262 |
Date | January 2022 |
Creators | Löfgren, Isabelle |
Publisher | Högskolan Dalarna, Institutionen för information och teknik |
Source Sets | DiVA Archive at Upsalla University |
Language | English |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0024 seconds