Causality- and Passivity-Constrained Nonnegative Attention for Interpretable Structure-Borne Road Noise Prediction in Battery Electric Vehicles
Abstract
In battery electric vehicles (BEVs), structure-borne road noise in the 20 Hz to 300 Hz band becomes more audible because the engine-masking component is largely absent, and conventional transfer-path formulations can be sensitive to suspension nonlinearity and ill-conditioned inversions. This paper presents a physics-informed, non-negative multi-modal fusion network (NN-MMFNet) that predicts in-cabin sound pressure from multipoint chassis excitations while keeping the mapping physically plausible and interpretable. The model combines a dual-stream encoder to separate transient impact signatures from steady resonance content with a strictly
causal fusion/decoding pathway. A passivity-motivated spectral gain cap is applied to prevent non-physical amplification while preserving phase. To enable additive path attribution, the cross-modal attention weights are constrained to be non-negative. Training follows a sim-to-real workflow, using virtual-fleet pretraining and short fine-tuning on measured data. On a production BEV, NN-MMFNet reproduces the 20 Hz to 300 Hz spectrum with a 1.12 dB(A) global root mean square error (RMSE) at 60 km/h and a 0.14 dB error at the 128 Hz boom, outperforming transfer path analysis (TPA), frequency transfer matrix (FTM), and autoregressive moving average (ARMA) baselines. Impulse-response checks show a negligible passivity-violation rate (<0.01 %). The learned attention consistently points to a rear subframe-to-body mounting path near 128 Hz, and a targeted stiffness adjustment at this location reduces the measured cabin noise by 4.2 dB(A).
Keywords:
structure-borne road noise, physics-informed neural networks (PINN), transfer path analysis (TPA), cross-modal attention, battery electric vehicles (BEVs)References
- Cervantes-Madrid G., Peral-Orts R., Campillo-Davo N., Campello-Vincente H. (2021), Inverse transfer path analysis, a different approach to shorten time in NVH assessments, Applied Acoustics, 181: 108178, https://doi.org/10.1016/j.apacoust.2021.108178
- Cheng W. et al. (2022), AR model-based crosstalk cancellation method for operational transfer path analysis, Journal of Mechanical Science and Technology, 36: 1131–1144, https://doi.org/10.1007/s12206-022-0206-7
- Cheng W., Chu Y., Chen X., Zhou G., Blamaud D., Lu J. (2020), Operational transfer path analysis with crosstalk cancellation using independent component analysis, Journal of Sound and Vibration, 473: 115224, https://doi.org/10.1016/j.jsv.2020.115224
- Cheng W., Lu Y., Zhang Z. (2016), Tikhonov regularization-based operational transfer path analysis, Mechanical Systems and Signal Processing, 75: 494–514, https://doi.org/10.1016/j.ymssp.2015.12.025
- de Klerk D., Ossipov A. (2010), Operational transfer path analysis: Theory, guidelines and tire road noise application, Mechanical Systems and Signal Processing, 24(7): 1950–1962, https://doi.org/10.1016/j.ymssp.2010.05.009
- de Klerk D., Rixen D.J. (2010), Component transfer path analysis method with compensation for test bench dynamics, Mechanical Systems and Signal Processing, 24(6): 1693–1710, https://doi.org/10.1016/j.ymssp.2010.01.006
- Gao L. et al. (2024), Operational transfer path analysis with crosstalk cancellation based on least variance spectrum estimation, Journal of Mechanical Science and Technology, 38: 5311–5322, https://doi.org/10.1007/s12206-024-0907-1
- Gustavsen B., Semlyen A. (1999), Rational approximation of frequency domain responses by vector fitting, IEEE Transactions on Power Delivery, 14(3): 1052–1061, https://doi.org/10.1109/61.772353
- Gustavsen B., Semlyen A. (2001), Enforcing passivity for admittance matrices approximated by rational functions, IEEE Transactions on Power Systems, 16(1): 97–104, https://doi.org/10.1109/59.910786
- Huang H., Lim T.C., Wu J., Ding W., Pang J. (2023), Multitarget prediction and optimization of pure electric vehicle tire/road airborne noise sound quality based on a knowledge- and data-driven method, Mechanical Systems and Signal Processing, 197: 110361, https://doi.org/10.1016/j.ymssp.2023.110361
- Jia X., Zhou L., Huang H., Pang J., Yang L. (2024), Improving electric vehicle structural-borne noise based on convolutional neural network-support vector regression, Electronics, 13(1): 113, https://doi.org/10.3390/electronics13010113
- Khan D., Burdzik R. (2023), Measurement and analysis of transport noise and vibration: A review of techniques, case studies, and future directions, Measurement, 220: 113354, https://doi.org/10.1016/j.measurement.2023.113354
- Kiranyaz S., Avci O., Abdeljaber O., Ince T., Gabbouj M., Inman D.J. (2021), 1D convolutional neural networks and applications: A survey, Mechanical Systems and Signal Processing, 151: 107398, https://doi.org/10.1016/j.ymssp.2020.107398
- Kong L. et al. (2025), A novel operational transfer path analysis based on the complex-valued crosstalk elimination method, Measurement, 253(Part D): 117813, https://doi.org/10.1016/j.measurement.2025.117813
- Ma Y., Dai R., Liu T., Liu J., Yang S., Wang J. (2025), Research on vehicle road noise prediction based on AFW-LSTM, Machines, 13(5): 425, https://doi.org/10.3390/machines13050425
- Masri J., Amer M., Salman S., Ismail M., Elsisi M. (2024), A survey of modern vehicle noise, vibration, and harshness: A state-of-the-art, Ain Shams Engineering Journal, 15: 102957, https://doi.org/10.1016/j.asej.2024.102957
- Mohammadi N. (2023), Airborne and structure-borne noise control in the MB truck cabin interior by the noise reduction in the transmission path, Archives of Acoustics, 48(1): 93–101, https://doi.org/10.24425/aoa.2022.142908
- Moorhouse A.T., Elliott A.S., Evans T.A. (2009), In situ measurement of the blocked force of structure-borne sound sources, Journal of Sound and Vibration, 325(4–5): 679–685, https://doi.org/10.1016/j.jsv.2009.04.035
- Munder M., Carbon C.C. (2022), A literature review [2000–2022] on vehicle acoustics: Investigations on perceptual parameters of interior soundscapes in electrified vehicles, Frontiers in Mechanical Engineering, 8: 974464, https://doi.org/10.3389/fmech.2022.974464
- Ortega Almiron J., Bianciardi F., Corbeels P., Pieroni N., Kindt P., Desmet W. (2022), Vehicle road noise prediction using component-based transfer path analysis from tire test-rig measurements on a rolling tire, Journal of Sound and Vibration, 523: 116694, https://doi.org/10.1016/j.jsv.2021.116694
- Park U., Kang Y.J. (2024), Operational transfer path analysis based on neural network, Journal of Sound and Vibration, 579: 118364, https://doi.org/10.1016/j.jsv.2024.118364
- Raissi M., Perdikaris P., Karniadakis G.E. (2019), Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations, Journal of Computational Physics, 378: 686–707, https://doi.org/10.1016/j.jcp.2018.10.045
- Shang Z., Hu F., Zeng F., Wei L., Xu Q., Wang J. (2021), Research of transfer path analysis based on contribution factor of sound quality, Applied Acoustics, 173: 107693, https://doi.org/10.1016/j.apacoust.2020.107693
- van der Seijs M.V., de Klerk D., Rixen D.J. (2016), General framework for transfer path analysis: History, theory and classification of techniques, Mechanical Systems and Signal Processing, 68–69: 217–244, https://doi.org/10.1016/j.ymssp.2015.08.004
- Yang M., Dai P., Yin Y., Wang D., Wang Y., Huang H. (2025), Predicting and optimizing pure electric vehicle road noise via a locality-sensitive hashing transformer and interval analysis, ISA Transactions, 157: 556–572, https://doi.org/10.1016/j.isatra.2024.11.059
- Zhang E., Chen Y., Su L., Zhonglian R., Chen X., Jiang S. (2024), Evaluation modeling of electric bus interior sound quality based on two improved XGBoost algorithms using GS and PSO, Archives of Acoustics, 49(3): 307–317, https://doi.org/10.24425/aoa.2024.148794
- Zhu H., Zhao J., Wang Y., Ding W., Pang J., Huang H. (2024), Improving of pure electric vehicle sound and vibration comfort using a multi-task learning with task-dependent weighting method, Measurement, 233: 114752, https://doi.org/10.1016/j.measurement.2024.114752

