engleski

Optimization of noise barrier efficiency using genetic algorithms

Although noise barriers are the most common technical solution for decreasing the excess noise levels at a receiver point, there is a constant need for increasing their efficiency. Insertion loss, the acoustic efficiency parameter, largely depends on the barrier’s geometry, mostly its height. Nevertheless, the barrier’s top and its overall shape can have a substantial influence on the insertion loss as well. This paper presents the use of genetic algorithms in the design phase of noise barriers by using BEM as the numerical tool for calculating its insertion loss. The optimization was not carried out only by taking into account its acoustical performance, but also having in mind the economic feasibility of using different materials and shapes of the barrier modules. An economic feasibility coefficient is defined and used as a final numerical value for comparing the overall efficiency of barrier design. The barrier performance was evaluated for a point sound source in both 2D and 3D simulation space, and its total rating based on the economic feasibility coefficient and its acoustical performance were compared to a reference plain barrier of same height. Scale model measurements were performed as well and compared to the simulation results.

Noise barriers; Insertion loss; Genetic algorithms

nije evidentirano