Power system integration of VSC-HVDC connected offshore wind power plants

(28/07/2019) Power system integration of VSC-HVDC connected offshore wind power plants This report presents an overview of challenges and solutions for the integration into the power system of offshore wind power plants (WPPs) connected to onshore grids through a voltage-source converter based high voltage direct current (VSC-HVDC) transmission system. Aspects that are touched upon are (i) principles for the control of offshore alternating current (AC) networks behind offshore VSC-HVDC converters, (ii) power system services that could be featured by VSCHVDC connected WPPs and (iii) clustering of multiple WPPs to co-ordinately provide desired control actions. After a brief introduction to justify the study, describe the state-of-art and formulate the project’s objectives, the report is essentially divided into three parts, as follows. Control principles of offshore AC networks The control of offshore AC networks relies purely on power electronics, especially if Type 4 wind turbine generators (WTGs) are used. Assuming the WTGs are controlled in a “standard” way (based on established literature), two state-ofart control strategies for the offshore HVDC converter are compared in different operational scenarios: (Option 1) nested voltage-current control scheme based on vector control and (Option 2) direct AC voltage control with addition of active damping. The design of controllers at no-load is discussed, after which Option 2 appears superior. Recommendations for enhanced performance with Option 1 are given. Further analysis is performed when a WPP is connected to the network, highlighting the fact that Option 2’s performance is less dependent on the control parameters than Option 1’s. This could be an advantage, since it may allow for independent design of other elements in the network. On the other hand, it may be disadvantageous, since small room for performance improvement is left. The latter point is somehow confirmed by scenarios where multiple HVDC converters are sharing the control of the network. In this situation, Option 1 with proper active and reactive power droop loops appears superior at a first glance. However, Option 2 seems more easily adaptable to different scenarios. Anyhow, a more complete assessment is necessary for a final conclusion on the absolutely best control scheme. Power system services First among the power system services under focus is the control of the AC voltage at the onshore HVDC station. In particular, interesting results are derived in terms of AC voltage control for connection to weak AC networks as well as of long-term voltage stability. New illustrative diagrams in the reactive power – AC voltage plane are drawn for connection to weak grids, shedding light on some peculiarities such as the non-linearity of the continuous short circuit power contribution from the HVDC station and the importance of using AC voltage control when connecting to weak grids. In terms of longterm voltage stability, the focus is on the HVDC converter behaviour when reaching its current limitation while the network approaches its voltage stability limit. The benefits of not prioritising active power during current-limited operation are demonstrated on a simple system and the possible implications on the control of a WPP potentially connected behind the HVDC converter are discussed. Active power balance control is the second service being analysed in this report. In ACDC grids, this is linked to the control of AC frequency and DC voltage. The two control services are treated one by one. Since the former has already widely been discussed in literature, focus is on formulating recommendations for real-life implementation of the service, by comparing a communication-based scheme with a communication-less one on a pointto-point VSC-HVDC connection of WPP. For most of the cases, use of communication may be considered the best solution. Other inherent limitations that are observed in WPPs are discussed. DC voltage control is briefly analysed from the standpoint of WPPs, once again taking into account their realistic limitations and their implications for the other players in the control of the DC network. Power oscillation damping (POD) is the last service within the scope of this report. Aspects related to its implementation are analysed. In particular, a deeper assessment of the robustness of POD with active and reactive power is made, emphasising the better performance of active power based POD as opposed to reactive power, mainly due to the presence of voltage regulators. Furthermore, a better understanding of which information should be exchanged between manufacturers, utilities and transmission system operators is gained. Crucial information is for example (i) the active power and voltage sensitivity of synchronous generators to injection of active and reactive power from the HVDC converter, (ii) the voltage regulation characteristics of the network in the vicinity of the HVDC station and (iii) limiting characteristics of WPPs such as inherent control and communication delays, presence of mechanical resonances at the same frequency as POD and active power ramp-rate limitations. Clustering of wind power plants The proof of concept of clustering significantly different WPPs is conducted in the last part of the report, demonstrating the possibility of implementing coordinated and synchronised active power control. An experimental validation is used to corroborate part of the results, which also provides support for the validity of some of the simulation results provided earlier. Furthermore, the characteristics of the test devices allow supporting some of the recommendations that are proposed earlier in the report, predominantly with regard to the implementation of POD.