This paper presents measurement results of the world wide first successful certification the electrical properties of a wind turbine, solely based upon measurements obtained at a system test bench with HiL-System and grid emulator. For all certification relevant tests the results are compared to field measurements. The impact of the real-time models in the HiL-System as well as the converter-based grid emulator are discussed in this paper. For full converter wind turbine, different requirements for the model depth could be determined depending on the tests. Nevertheless, higher-quality models that reflect the plant behaviour better are recommended to reduce uncertainties within the certification process. This paper also shows that especially for grid failure events grid emulators require real-time impedance control, in order to emulate grid failures properly. Based on these findings, recommendations for the requirements on test bench components are formulated in this paper, in order to contribute to new certification guidelines. Overall, we conclude that based on the experiences made at two different system test benches, the vast majority of certification measurements can be carried out without limitation at such system test benches.

Recently developed nacelle test benches for wind turbines, equipped with multi-physics Hardware-in-the-Loop (HiL) systems, enable advanced testing and even certification of next-generation wind turbines according to IEC61400-21. On the basis of three experiments carried out with a commercial 3.2 MW wind turbine, this paper shows to which extent test bench hardware and HiL systems influence certification results. For the crucial Fault-Ride-Through tests, all deviations were found to be below 1% compared to field and simulation results. For this test, the power HiL system and the accuracy of its impedance emulation are found to be of most relevance. The results for the test items Frequency Control and Synthetic Inertia were found to be more sensitive to shortcomings of the mechanical HiL with its control system. Based on these findings, the paper mentions general procedures to ensure the quality of test benches with HiL systems and, with that, ensure the quality of certification.

No abstract

We thank the referee Amir R. Nejad for the detailed review. We considered the remarks very helpful to further improve the manuscript's quality. General remarks: Referee general remarks: This article presents the application of HiL for drivetrain testing in onshore wind turbines.The article is well-structed and topic is of interest. Author's response to general remarks: We thank the referee for the positive feedback.

We thank the anonymous referee #1 for the extensive review. We considered the remarks very helpful to further improve the quality of the manuscript. General remarks: Anonymous referee general remarks: Very little information is provided on the drivetrain being tested or the wind turbine that is modeled. Also there are limited details provided on the implementation of each of the controllers. Further an evaluation of three standard control approaches on an unspecified system or plant is not novel and it is not justified for publication in its present state. Author's response to general remarks: The plant is a multi-MW experimental system C1 WESD Interactive comment Printer-friendly version Discussion paper test bench for wind turbines. The choice of rather fundamental control approaches is justified, since the HiL methods aim at experimental validation, impeded by the dimensions of the test bench and safety concerns. To keep the focus on the HiL control methods, we omitted most of the details on the device under test and the original wind turbine. The authors agree with the referee that further information is beneficial for interpretation and reproduction of the results and the corresponding details have been added to the text. Finally, a comparison of control approaches, in this case three, for a multi-MW system test bench with HiL functionality has not been published. We hope that with the following changes, the relevance of the findings for an international audience becomes clearer. Further remarks: Anonymous referee remark (1): The abstract needs to be rewritten to clearly articulate what novelty has been investigated and what was found. Author's response on remark (1): The quantitative comparison of the HiL control methods is novel as it offers a guideline for selecting the appropriate method for respective testing demands. The different performances of the HiL control methods were already included in the abstract of the discussion paper. Changes in the manuscript: An explicit statement of the novelty has been added to the abstract (l.15): "Our quantitative comparison of HiL control methods facilitates selecting the appropriate method for respective testing and certification demands." Anonymous referee remark (2): Page 2, line 35: "Only with a Hardware-in-the-Loop (HiL) control method that reproduces the WTs inertia.....". Please define what aspects of the control system do you define as HiL and whether it operates on a realtime platform? Author's response to remark (2): In general, HiL systems substitute missing components by simulation models and connect the simulation models to the actually present components. In the present application, the actually present components, or "hard-C2