11 Kalman-Based Virtual Sensing for Improvement of Service Response Replication in Environmental Tests 105 0 5 1015202530 -1 -0.5 0 0.5 1 10-5 (a) 7.64 7.66 7.68 7.7 7.72 -1 -0.5 0 0.5 1 10-5 (b) 0 200 400 600 800 1000 10-15 10-10 (c) 0 5 1015202530 -4 -2 0 2 4 10-5 (d) 7.64 7.66 7.68 7.7 7.72 -4 -2 0 2 4 10-5 (e) 0 200 400 600 800 1000 10-15 10-10 (f) Fig. 11.11 Time history (left), detailed time history (middle) and PSD (right) of estimated (dashed blue line) and measured (solid black line) responses of sensors 11 ((a)-(c)) and 10 ((d)-(f)) 11.6 Conclusions The application of VS techniques to environmental tests on the BARC has been proposed in this work. The BARC is a simple hardware demonstrator used in the framework of the BCC, a challenge that focuses on improving operational environment replication at the component level during environmental tests. In particular, the potential of the AKF for both input and state estimation has been presented in this paper. A first simulated example, which shows good results in terms of input and component response reconstruction, has been reported. Furthermore, a test campaign has been conducted placing the BARC on a monoaxial electrodynamic shaker of comparable size. Results from the AKF application to the acquired data set have been discussed in this paper. It is shown that the AKF estimates different frequency components on each input and therefore fails in identifying the correct time histories. The main reason for the errors in the inputs’ estimation is attributed to the test boundary conditions uncertainties. A possible improvement could consist in adding torque estimation, or in including the shaker in the simulation model in order to take into account its interaction with the BARC. Satisfying results are obtained in terms of component response estimation, even if a low magnitude estimate for the high frequency components of the responses in y direction is provided. Adding acceleration data to the measurements vector represents a future step that can help to recover high frequency components in the estimated quantities. Future investigation aims at applying VS techniques to operational data to obtain the full-field service responses and compare them to the test ones in order to approach the BCC goal. Acknowledgments The author gratefully acknowledge the European Commission for its support of the Marie Sklodowska Curie program through the ITN DyVirt project (GA 764547). References 1. Larsen, W., Blough, J.R., DeClerck, J.P., VanKarsen, C.D., Soine, D.E., Jones, R.: Initial modal results and operating data acquisition of shock/vibration fixture. In: Topics in Modal Analysis & Testing, vol. 9, pp. 363–370. Springer (2019) 2. Musella, U., Blanco, M.A., Mastrodicasa, D., Monco, G., Simone, M., Peeters, B., Mucchi, E., Guillaume, P., et al.: Combining test and simulation to tackle the challenges derived from boundary conditions mismatches in environmental testing. In: Sensors and Instrumentation, Aircraft/Aerospace, Energy Harvesting & Dynamic Environments Testing, vol. 7, pp. 259–269. Springer (2020) 3. Rohe, D.P., Smith, S., Brake, M.R.W., DeClerck, J., Blanco, M.A., Schoenherr, T.F., Skousen, T.J.: Testing summary for the box assembly with removable component structure. In: Sensors and Instrumentation, Aircraft/Aerospace, Energy Harvesting & Dynamic Environments Testing, vol. 7, pp. 167–177. Springer (2020) 4. Simon, D.: Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches. Wiley (2006)
RkJQdWJsaXNoZXIy MTMzNzEzMQ==