Optimizing source placement for noise minimization using hybrid acoustic simulation. (March 2018)

• Record Type: Journal Article
• Authors: Morales, Nicolas; Manocha, Dinesh
• Is Part Of: Computer aided design. Volume 96(2018)
• Page Start: 1
• Page End: 12
• Publication Date: 2018-03
• Subjects: Acoustic design -- Automatic design optimization -- Architectural acoustics -- Environmental noise reduction; Computer-aided design -- Periodicals; Engineering design -- Data processing -- Periodicals; Computer graphics -- Periodicals; Conception technique -- Informatique -- Périodiques; Infographie -- Périodiques; Computer graphics; Engineering design -- Data processing; Periodicals; Electronic journals; 620.00420285
• DOI: 10.1016/j.cad.2017.09.007
• Languages: English
• ISSNs: 0010-4485

Abstract

Abstract: Environmental noises can have a large impact on workplace productivity and health. We address the problem of designing large CAD structures that automatically meet the noise guidelines and standards. We present a novel approach that determines the optimal placement of sound sources such as generators, machinery, fans and HVAC appliances that can conform to these design requirements in indoor scenes. Our approach uses a hybrid sound propagation algorithm that combines elements of both geometric and wave-based techniques for handling higher and lower frequencies, respectively. We use an efficient simulated annealing approach for acoustic optimization that uses impulse response caching and sound source clustering to improve the convergence time. Finally, we present our approach in the context of a traditional CAD pipeline, where noise conformance requirements can be integrated during the early stages of the digital design phase. We highlight the approach's performance on many complex CAD models corresponding to an office and a warehouse. Highlights: We present an algorithm for acoustic design optimization to reduce environmental noise. An accurate and efficient hybrid sound propagation technique for noise reduction. Source clustering and IR caching are used to reduce iteration cost. We show the accuracy of our hybrid technique over standard geometric methods.