DrMuhammad Gulzari

The advancement of European society is dependent on safe, efficient, and environment-friendly technologies. Radically new approaches in transportation and energy production will substantially modify the urban and interurban acoustic landscapes. Noise mitigation will be one of the major challenges of the next decades. Breakthrough technologies will be essential to cope with this challenge in the mid/long term. The most likely candidate to provide these radically new technologies are acoustic metamaterials. These artificially engineered structures have revealed fantastic dynamic mechanical properties and wave phenomena that are unachievable from naturally occurring materials. The remaining challenges that prevent their uptake are optimized design tools and the manufacture of complex precise microstructures. This research intends to overcome these challenges by developing deep learning methods for optimized, low-cost design tools and developing post-processing steps applied to the latest additive manufacturing technologies to achieve designs that, until recently, would have been considered unmanufacturable. High fidelity numerical modelling tools will be used to develop a training dataset suitable for deep learning networks. This dataset will include the topological features required to achieve the frequency bandgaps where sound wave propagation is prohibited. From the trained networks, optimized and highly accurate acoustic metamaterials will be produced without the need for high cost modelling. The output designs will be produced by additive manufacturing technology and experimental findings will be compared with theoretical and numerical results. The impact of the manufacturing will be assessed to identify issues which lead to degraded performance of the acoustic metamaterial. This information will then be included in a second training phase of the deep learning networks allowing the networks to produce designs which have been optimized for practical manufacture. Our approaches, research methodology and findings will be pioneering for acoustic metamaterial research and will pave the way for practical acoustic metamaterial design and applications.