کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
565470 1451859 2016 15 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Introducing passive acoustic filter in acoustic based condition monitoring: Motor bike piston-bore fault identification
ترجمه فارسی عنوان
معرفی فیلتر آکوستیک منفعل در نظارت بر وضعیت آکوستیک: شناسایی خطای موتور پیستون موتور
موضوعات مرتبط
مهندسی و علوم پایه مهندسی کامپیوتر پردازش سیگنال
چکیده انگلیسی


• Implementation of passive acoustic filter in acoustic based condition monitoring.
• GA based optimization methodology to design the desired band-pass acoustic filter.
• Fault identification using machine learning techniques using standard six statistical parameters.
• Demonstration of machine learning improvement using passive acoustic filter.

Requirement of designing a sophisticated digital band-pass filter in acoustic based condition monitoring has been eliminated by introducing a passive acoustic filter in the present work. So far, no one has attempted to explore the possibility of implementing passive acoustic filters in acoustic based condition monitoring as a pre-conditioner. In order to enhance the acoustic based condition monitoring, a passive acoustic band-pass filter has been designed and deployed. Towards achieving an efficient band-pass acoustic filter, a generalized design methodology has been proposed to design and optimize the desired acoustic filter using multiple filter components in series. An appropriate objective function has been identified for genetic algorithm (GA) based optimization technique with multiple design constraints. In addition, the sturdiness of the proposed method has been demonstrated in designing a band-pass filter by using an n-branch Quincke tube, a high pass filter and multiple Helmholtz resonators. The performance of the designed acoustic band-pass filter has been shown by investigating the piston-bore defect of a motor-bike using engine noise signature. On the introducing a passive acoustic filter in acoustic based condition monitoring reveals the enhancement in machine learning based fault identification practice significantly. This is also a first attempt of its own kind.

Figure optionsDownload as PowerPoint slide

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Mechanical Systems and Signal Processing - Volumes 70–71, March 2016, Pages 932–946
نویسندگان
, ,