首页> 外文会议>International Conference on Mechanical Design >A Novel Auto Design Method of Acoustic Filter Based on Genetic Programming
【24h】

A Novel Auto Design Method of Acoustic Filter Based on Genetic Programming

机译:基于遗传编程的声学滤波器新型自动设计方法

获取原文

摘要

A novel auto design method of acoustic filter based on genetic programming (GP) is proposed in this paper. Unlike the model-based auto optimization method by genetic algorithms that optimize acoustic filters only in the parameter domain, the proposed method can optimizes the acoustic filter in both the structure region and the parameter region simultaneously thanks to its tree representation of individuals. In the proposed method, the widely used acoustic components have been equivalent to the basic elements in the topology. The acoustic filter individuals in GP are composed of these elements. Transfer matrix method is employed to calculate the transmission loss of the acoustic filter individuals. By a certain number of generations, the GP finds the individuals with good fitness value. The feasibility of the method is validated by two numerical experiments. The first case study is to validate the searching ability of the method. The result shows that the proposed method has excellent ability to search the optimal solution with transmission loss curve close to the objective transmission loss curve. The second experiment is based on a typical automobile muffler design strategies that requires a high noise reduction magnitude at two specified frequency band and a certain noise reduction magnitude at the whole frequency band, the evolution result meets the design requirements well. The results of these two numerical experiments demonstrate that the proposed method has excellent searching ability and is feasible in the auto design of acoustic filter.
机译:本文提出了一种基于遗传编程(GP)的声学滤波器的新型自动设计方法。与基于模型的自动优化方法不同,该遗传算法仅在参数域中优化声学滤波器,所提出的方法可以同时优化结构区域和参数区域中的声学滤波器,同时相应于其的树形表示。在所提出的方法中,广泛使用的声学组件相当于拓扑中的基本元素。 GP中的声学过滤器由这些元件组成。使用传输矩阵方法来计算声学滤波器的传输损耗。通过一定数量的几代人,GP找到具有良好健身值的个人。通过两个数值实验验证了该方法的可行性。第一种案例研究是验证方法的搜索能力。结果表明,该方法具有优异的能力,其能够与靠近目标传输损​​耗曲线的传输损耗曲线搜索最佳解决方案。第二个实验基于典型的汽车消声器设计策略,该设计策略需要在两个指定的频带处具有高降噪幅度和整个频带的某些降噪幅度,进化结果符合设计要求。这两个数值实验的结果表明,所提出的方法具有优异的搜索能力,并且在声学滤波器的自动设计中是可行的。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号