Frequency control and stability requirement on Hydro plants: System identification for performance and stability assessment
MetadataShow full item record
- Institutt for elkraftteknikk 
The Nordic transmission system operators (TSOs) have proposed new draft requirements for the providers of frequency containment reserves. These requirements include extensive tests to determine the dynamics of hydro power plants. The dynamics of the hydro power plant are used to verify whether or not the power plants qualify to provide these reserves. To be more precise, the tests require the power plant owners to measure the plants’ response when operating in open loop and with various sine sweep signals modulating the turbine governor setpoint. This is an intrusive approach and alternatives should therefore be investigated. In this work, three novel methods have been investigated as alternatives to the one proposed by the TSOs. The main novelties of these methods are that plants are allowed to continue closed loop operation during testing and that the added excitation is limited. The three methods are characterised by different requirements on input data, as follows: 1. Phasor measurement units (PMU) measurements close to the power plant, without added extra excitation. 2. Control system measurements from the power plant, without added extra excitation. 3. Control system measurements from the power plant, with added extra excitation. For all of the proposed methods it was analysed under which conditions the results are consistent (non biased). Although consistency cannot in general be guaranteed, it was argued that the bias due to lack of consistency is small. Moreover, the bias can be further reduced by adding external excitation to the identification procedure. To validate the methods, tests at two different power plants in the Norwegian power system were performed. The first test compared the PMU method to the one proposed in the coming draft requirements. From this test it can be seen that the PMU method yields similar results to the one proposed in the draft requirements. It was also shown that only one dataset is needed per operating state under investigation. That is true even if the method outlined in the draft requirements is used. This is an important observation as the draft proposes to use 10 tests per operating state under investigation. During the test at the other power plant it was demonstrated that the proposed method using control system measurements without added excitation could detect changes in the settings of the plant’s turbine governor (PID) parameters. Moreover, it was shown that this method is capable of estimating steady state gains of the governor controller that correspond very closely with the actual permanent droop setting of the plant . The methods were also demonstrated using the simulation softwares SIMULINK and PSS/E and a Monte Carlo Simulation (MCS) approach. This approach was used to investigate how large nonlinearities could be present before the results became too biased as well as some other aspects presented below. When using a PMU for the identification, the power system frequency is used as an estimate of the angular speed of the machine. Consequently, a natural question is, how large an error will this lead to? The MCS approach showed that frequency is indeed a good approximation of speed. This is perfectly true when studying the slowest turbine and governor dynamics, but for the faster dynamics there will be a bias in the estimate. When it comes to the performance of the three methods, the best results are obtained when the turbine governor uses angular speed of the rotor as feedback signal, and at the same time measurements from the power plant control system is utilized and extra excitation is added to the governor setpoint. It is possible to obtain a good estimate in the other cases too, but then some bias in the estimation should be expected, especially for the faster dynamics.
Has partsPaper 1: Jakobsen, Sigurd Hofsmo; Uhlen, Kjetil. Vector fitting for estimation of turbine governing system parameters. I: 12th IEEE Power and Energy Society PowerTech Conference PowerTech Manchester 2017. IEEE Press 2017 ISBN 978-1-5090-4238-8. https://doi.org/10.1109/PTC.2017.7980855 - © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
Paper 2: Jakobsen, Sigurd Hofsmo; Uhlen, Kjetil. Development of a test system for identification of turbine dynamics using the dc power flow. IFAC-PapersOnLine 2018 ;Volum 51.(2) s. 97-102 https://doi.org/10.1016/j.ifacol.2018.03.017
Paper 3: Jakobsen, Sigurd Hofsmo; Uhlen, Kjetil; Bombois, Xavier. Identification of Hydro turbine governors using PMU data. I: 2018 IEEE International Conference on Probabilistic Methods Applied to Power Systems - PMAPS. IEEE conference proceedings 2018 ISBN 978-1-5386-3596-4. s. https://doi.org/10.1109/PMAPS.2018.8440273 - © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
Paper 4: Jakobsen, Sigurd Hofsmo; Uhlen, Kjetil;Lie,P. System identification techniques for validating hydro power plant’s FCR performance. CIGRE
Paper 5: Jakobsen, Sigurd Hofsmo; Uhlen, Kjetil. Testing of a hydropower plant’s stability and performance using PMU and control system data in closed loop. - This paper is a postprint of a paper submitted to and accepted for publication in [journal] and is subject to Institution of Engineering and Technology Copyright. The copy of record is available at the IET Digital Library IET Generation, Transmission & Distribution 2019 ;Volum 13.(23) s. 5339-5348 http://dx.doi.org/10.1049/iet-gtd.2019.0804
Paper 6: Checking hydro power plants’ FCR performance using system identification in closed loop
Appendix A: Vanfretti, Luigi; Olsen, Svein H; Arava, VS Narasimham; Laera, Giuseppe; Bidadfar, Ali; Rabuzin, Tin; Jakobsen, Sigurd Hofsmo; Lavenius, Jan; Baudette, Maxime; Gomez-Lopez, Francisco J. An open data repository and a data processing software toolset of an equivalent Nordic grid model matched to historical electricity market data. Data in Brief 2017 ;Volum 11. s. 349-357 https://doi.org/10.1016/j.dib.2017.02.021 This is an open access article under the CC BY license
Appendix B: Jakobsen, Sigurd Hofsmo; Kalemba, Lester; Solvang, Espen Hafstad. The Nordic 44 test network.
Appendix C: An alternative derivation of the frequency divider formula using the dc power flow