A framework for predicting traffic-induced vibrations and structure-borne sound in lightweight buildings using machine learning approaches

Urban densification, despite its economic and environmental opportunities, brings an increase in noise- and vibration-producing sources, which, when coupled with the rising use of lightweight buildings, risks deteriorating living conditions that have a negative impact on health.

The densification of our cities will continue unabated, as will our reliance on lightweight buildings that is fueled by sustainability concerns.We propose to develop a software using novel machine learning methods in combination with recently developed computational methods.

We suggest that timely and informed decision-making in the design of lightweight buildings is necessary to mitigate the negative synergy of urban density, lightweight structures, and rise in noise sources.

This can be achieved via the use of computational prediction tools, which are trained, calibrated and validated to measured data.The research aims at easing vibroacoustic performance prediction during conceptual design of lightweight buildings, thus enabling mitigation of noise and vibration in urban environments and a more efficient structural design using less resourses.

The principal investigator has assembled a team of established international scholars, with complementary technical expertise.

The software will ease the assessment of vibrations and structure-borne noise, which leads to resource-savings and mitigated effects of noise and vibrations, and a fair competitiveness of lightweight buildings in dense urban areas.