Publishable Summary for 14IND14 MNm Torque Torque measurement in the MN m range

Publishable Summary for 14IND14 MNm Torque Torque measurement in the MN m range Overview The overall aim of this project is to provide traceability for torque measurements in the MN m range for nacelle test benches. Such a development will support the wind energy industry by significantly improving testing conditions. Within the framework of this project, existing nacelle test benches will be reviewed, multi-component effects of superimposed forces and bending moments will be investigated and novel traceable calibration methods will be developed. Need In the last few decades, the combination of climate change and the increase in electricity consumption led to a general demand for more renewable energy. The EU Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources is a direct consequence of this general demand. It requests the countries of the European Union to set overall national targets for the use of renewable energy sources. One of the main pillars in the new energy mix in most countries is onshore and offshore wind energy. To improve the technical development of wind turbines as well as their cost-effectiveness, several large test benches have been constructed to supply full scale testing facilities. These new test benches can be used to test full nacelles (the upper part of a wind power station) under conditions that are similar to the field. Current nacelles have a power rating of around 3 MW (onshore) and 6 MW (offshore) and even 20 MW machines are thought to be possible (source: EU-Project Upwind). Therefore, the test benches have to apply torque loads in the multi-mn m range. The measurement of this torque load is needed for the steady control of the torque loading system and also for the determination of the efficiency of entire nacelles or single components that are tested on the test bench. At the moment, the operators of the above-mentioned test benches cannot measure torque loads very precisely. The main reason for this is that they are not traceable to a torque standard. The largest torque standard worldwide, at PTB in Germany, covers a maximum torque of 1.1 MN m while the largest nacelle test bench can exert up to 18 MN m. The traceability of torque measurements is important in order to be able to reliably verify the quality of the measurements. Another problem is the occurrence of other mechanical loads, such as longitudinal and lateral forces and bending moments, on the test benches which simulate the real wind conditions. The effect of these loads on the torque measurement is mostly unknown. Last but not least, all test benches differ from each other depending on their focus of testing (functionality, efficiency, ) but also their power ratings range from 1 20 MW. This variation makes universal approaches very difficult. Objectives The specific scientific and technological objectives of the project are to: 1. Review existing nacelle test benches and their boundary conditions. The review will include the range of loads that can be applied and the dimensions of the test bench, as well as existing methods of torque measurement and calibration and the levels of uncertainty achieved. 2. Develop novel traceable calibration methods for torque values on nacelle test benches in the form of transfer standards for the range above 1 MN m. In order to enable the multi-use of transfer standards, a unified approach for several nacelle test benches will be applied. Two different approaches will be used in the project: a commercial torque transducer will be used with an extrapolation procedure for the MN m range and a force lever system will be designed to directly reach the MN m range. 3. Investigate the effect of multi-component loading on the measurement of torque. In particular, cross-talk effects, in the case of 6-component loading (3 directional forces, 2 directional bending, torque), will be studied to describe effects on the torque measurements which occur on large nacelle test benches. Report Status: PU Public Publishable Summary Issued: January 2018 This publication reflects only the author s view and the Commission is not responsible for any use that may be made of the information it contains. 1/5

4. Develop a calibration procedure for large nacelle test benches. The calibration procedure will enable the traceability of torque loads up to 20 MN m and will include an uncertainty model that considers cross-talk effects. 5. Engage with industries that utilise the MN m range for torque measurements to facilitate the take up of the technology and measurement infrastructure developed by the project, to support the development of new, innovative products, thereby enhancing the competitiveness of European industry. Progress beyond the state of the art At present, several approaches have been used to solve the problem of the missing traceability for torque measurement in the range above 1 MN m: Torque calculation based on strain measurements (obtained using strain gauges); torque calculation based on low torque measurements on the high speed shaft in transmission test benches; torque calculation based on electrical power measurements and torque calculation of combined force and distance measurements. All of these four options have drawbacks: the use of strain gauges lacks accurate knowledge of material properties and dimensions which could be substituted with a calibration of the transducer; measurements on different parts of the transmission line (either electric power or high speed shaft) rely on estimations of friction losses in e.g. gears or engines; the use of the product between force and distance still suffers from the lack of investigations on the entailed measurement uncertainty. To progress beyond the state of the art, this project will analyse calibration approaches for nacelle test benches to evaluate their uncertainty. Traceability in the MN m range will be provided by calibration in partial load and subsequent extrapolation on a 5 MN m torque transfer standard, as well as by a force-lever system for up to 20 MN m which will be designed. The smallest expected uncertainty for the main torque component Mz is < 1 % for up to 5 MN m (state of the art: 0.1 % for up to 1.1 MN m). Multi-component effects on the torque measurement will be determined using small-scale investigations, Finite Element Modelling (FEM) simulations and tests on a nacelle test bench. Moreover, design principles for multi-component transducers will be developed which enable a quantification of the cross-talk effects on the torque measurements in the MN m range. Finally, a traceable calibration method will be developed for nacelle test benches. During the first 27 months of the project, a proposal for a torque calibration of a nacelle test bench was developed and tested with a 5 MN m torque transfer standard. Furthermore, several designs for a force lever system have been developed for which a patent application has been filed and an upscaling of the proposals is planned for the last phase of the project. Results In this section, achieved and envisioned results are presented according to the main objectives (as listed above): 1. An overview of existing nacelle test benches and their boundary conditions has been produced and includes eight test benches. This overview serves as a basis for completing all other objectives as it provides information on the mechanical facilities (torque, bending moments, forces) as well as on geometrical dimension to adapt transducers. Furthermore, a first overview of influencing factors for the torque measurement on these test benches have been determined. 2. Novel traceable calibration methods for torque measurements on nacelle test benches will be developed. This objective is approached using two methods: a. Based on the calibration of a 5 MN m torque transducer up to 1.1 MN m detailed FEM simulations of this transducer were established. The simulations have been improved and compared. It was found that it is not possible to use the FEM results for the development of an extrapolation method and a first method was, therefore, based on experimental measurements in the kn m range and partial measurements with the 5 MN m torque transducer. This extrapolation method is used for the analysis of the calibration data of the nacelle test bench of RWTH. b. A force lever system is being developed within the project which divides the torque load into a force measurement at a known distance. Up to now, first designs for a 5 MN m force lever system were developed and discussed and force transducers were chosen for the systems. A set of influences on the measurement uncertainty of such a system has been determined 2/5

and the effects will be studied using FEM. An internal guide on how to perform these investigations and on how to evaluate the designs and chose the best one is under preparation. Furthermore, first ideas on the measurement uncertainty budget have been discussed. 3. Knowledge about the effects of multi-component loads on the torque measurement. These will mainly be cross-talk effects of additional forces and moment components on the torque measurement. A proposal for a multi-component calibration in the kn m range was made and implemented using three different torque transducers with multi-component measurement capabilities and a developed lever-hanger-mass system. Furthermore, design principles for multi-component measurements in the MN m torque range will be studied. To this end, the transducers used for the multi-component measurements are used for comparison with FEM-simulations. Three FE-models of two of the used transducers have been developed and compared to the measurement results and analytical calculations. The results show that the FE-models can be used for the analysis of different transducer designs. 4. A calibration procedure for large nacelle test benches was developed in a first draft version within the project. As the measurements are a pilot test, a quite extensive set of tests were agreed upon. These consist of: (i) zero-point determination with and without rotation; (ii) tests for the determination of the temperature development at the test bench and the torque transducers; (iii) stationary tests with different torque loads at a low rotational speed for the comparison with the calibration data from PTB; (iv) stationary tests with different torque loads and different rotational speeds to determine the effect of the rotation; (v) test with different additional mechanical loads (lateral and longitudinal forces and bending moments) and (vi) different dynamic tests such as emergency brakes and acceleration and slow deceleration. The tests have been successfully performed on the 4 MW test bench of RWTH University at the Chair for Wind Power Drives in Aachen (Germany) with a 5 MN m torque transducer of PTB. The data gained from the experiments is at the moment being analysed. 5. Engagement with industries that utilise the MN m range for torque measurements. This includes a workshop for stakeholders as well as contributions to written standards. Up to now, test bench operators have been involved in the project to establish the overview mentioned above, a newsletter for stakeholders with up to now three issues was created and distributed and the project objectives were announced at standardisation meetings. All information is gathered on the project webpage and interested parties can join the project as stakeholder or collaborator at any time. Furthermore, a project meeting with the involvement of stakeholders has taken place in order to obtain information on the normal operation and control of nacelle test benches to derive recommendations for a torque calibration procedure. Impact As the project progresses, several impact activities have taken place. A newsletter was established and three issues have been sent to stakeholders to keep them in touch with the consortium. Furthermore, the first standardisation organisations have been informed about the project during regular meetings. This is the case for EURAMET TC-M and ISO TC 164. A stakeholder meeting has been held in Aachen in March 2017. One publication has been made in a peer-reviewed online journal. A reference is given at the end of this summary. Eight publications in conference proceedings have been made so far and nine conference contributions are planned. Many of them are joint publications. The design of the force lever system has been submitted for a German patent application (DE 10 2017 109 479.2). Moreover, a press release on the measurements on the test bench of RWTH has been published by PTB (in German) (https://www.ptb.de/cms/presseaktuelles/journalisten/nachrichten-presseinformationen/ presseinfo.html?tx_news_pi1%5bnews%5d=8484&tx_news_pi1%5bcontroller%5d=news&tx_news_pi1%5 Baction%5D=detail&tx_news_pi1%5Bday%5D=24&tx_news_pi1%5Bmonth%5D=11&tx_news_pi1%5Byear %5D=2017&cHash=aa44c7dbe5ab02646d377b6fb1fc9e52). This release also presents the newly established 5 MN m torque transfer standard of PTB and mentions the development of the new torque calibration procedure which has been used for the tests. Impact on relevant standards Currently, there are no standards for the traceable torque calibration of nacelle test benches. The results of this project will be input into different international committees so that guidelines and standards for the traceable calibration of nacelle test benches can be developed. This project will directly influence IEA Task 35 3/5

which focuses on the development of guidelines and recommendations for the in-system testing of test benches. Two partners represent the consortium in the IEA Task 35. The results of this project will also be of great importance for the EURAMET Technical Committee Mass and related quantities (EURAMET TC-M), which deals with the metrological aspects of torque measurements, and for ISO Technical Committee 164, subcommittee 1, which is concerned with standards for mechanical testing and metals. The results of this project will also be used in facilities which are similar to nacelle test benches (see examples below). The consortium has four representatives in the EURAMAT TC-M and one representative in the ISO subcommittee. Impact on industrial and other user communities The impact created will be improved torque measurements on nacelle test benches and in the wind energy industry. These impacts will reach different levels of users in industry: Nacelle test bench operators: Traceability of torque in the MN m range will highly improve the quality of test results and this will enable nacelle test benches to be certified. Three test bench operators are part of the consortium and will directly obtain all results. Other operators were already contacted and are also interested in attending stakeholder meetings to receive the output of the project. A first test of a torque calibration was performed within the project on the nacelle test bench of the Center for Wind Power Drives in Aachen. Wind park operators will benefit from more refined systems (nacelles, gearboxes etc.) indirectly by using the test bench facilities. Wind turbine manufacturers will benefit from better test results when using the facilities to test new nacelle prototypes but also for development purposes. Better test results, especially in European nacelle test benches, will also strengthen the European position in the wind industry worldwide. The potential users are kept up to date with the project newsletter which contains the newest information. New interested parties are always welcome. The first publication from the project is available and several others are underway and will be listed on the project webpage as soon as they are published. Impact on the metrological and scientific communities Progress will be reported to the EURAMET TC-M during its annual meetings. A project workshop is planned for the TC-M meeting in 2018. Further impact on the scientific community will be generated by providing an evaluated method for the torque calibration of nacelle test benches for the first time. Through the generated traceability for high torques, future research projects using nacelle test benches will benefit from known uncertainty. Validated simulation models can also be used in future projects, for example, on drive trains and torque transducers. List of publications 1. Schlegel, C., Kahmann, H. & Kumme, R. MN m torque calibration for nacelle test benches using transfer standards. ACTA IMEKO 5, 12 18 (2016). 2. Kock, S., Jacobs, G., Bosse, D., Weidinger, P. Torque measurement uncertainty in multi-mw nacelle test benches. Conference for Wind Power Drives : conference proceedings. 1-14 (2017). 3. Weidinger, P., Foyer, G., Schlegel, C., Kumme, R. Extending the torque calibration range - necessity and outline of a mathematical approach. Proceedings of VII international competition of COOMET "Best Young Metrologist" (2017). 4. Foyer, G., Kock, S. Measurement uncertainty evaluation of torque measurements in nacelle test benches. Proceedings of 23rd IMEKO TC3 Conference (2017). 5. Schlegel, C., Kahmann, H., Weidinger, P., Kumme, R. New perspectives for MN m torque measurement at PTB. Proceedings of 23rd IMEKO TC3 Conference (2017). 6. Lorente, R., Medina, N., Sáenz, M., Sebastián, M. Torque traceability for nacelle's test benches: a design proposal. Proceedings of 23rd IMEKO TC3 Conference (2017). 7. Weidinger, P., Schlegel, C., Foyer, G., Kumme, R. Characterisation of a 5 MN m torque transducer by combining traditional calibration and finite element method simulations. Proceedings Sensor 2017, 516-521 (2017). 4/5

8. Kock, S., Jacobs, G., Bosse, D., Gnauert, J. Conception of 5 MN m torque transducer for wind turbine test benches. Proceedings of 5 th Innovation Messtechnik (2017). 9. Kock, S., Jacobs, G., Bosse, D., Foyer, G. Influences on MN m torque measurement in multi-mw nacelle test benches. Proceedings of DEWEK 2017. Project start date and duration: 01 September 2015, 36 months Coordinator: Dr. Rolf Kumme, PTB, Tel: +49 531 592 1200 E-mail: rolf.kumme@ptb.de Project website address: http://www.ptb.de/emrp/torquemetrology.html Internal Funded Partners: Partner 1 PTB, Germany Partner 2 CEM, Spain Partner 3 CMI, Czech Republic Partner 4 VTT, Finland External Funded Partners: Partner 5 CENER, Spain Partner 6 FhG, Germany Partner 7 RWTH, Germany 5/5