基于函数连接型神经网络的非线性主动噪声控制系统研究
发布时间：2024-02-26 18:03
随着工业化和城市化的快速发展,严重的噪声污染的出现,诞生了一个有趣的噪声控制研究领域。从技术上讲,噪声控制领域大致可以分为两种方式:被动和主动。被动噪声控制(PNC)技术使用特殊材料来吸收或/和隔离不需要的噪声。然而这些材料设备通常体积大,安装困难,成本高,对低频噪声下消噪效果低差。相比之下,主动噪声控制(ANC)技术可以克服PNC方法的缺点,能有效消除或降低低频噪声。随着电子技术和自适应处理理论的发展,由于在重量、尺度和成本等方面都有潜在的优势,ANC噪声控制技术得到了越来越多的重视。本文主要研究在系统非线性较强的情况下,采用FLANN对ANC系统进行新的非线性自适应控制器设计,以提高噪声消除性能,降低计算复杂度。主要包括以下几个方面:首先,研究基于FLANN的ANC系统的性能,包括分析ANC系统中非线性的特性,分析FLANN对ANC系统的非线性建模能力;其次,基于这些分析,提出多种含有交叉项的FLANN控制器结构。此外,基于滤波误差技术和数据相关的部分更新策略设计新的算法,从而进一步减少计算负担。本论文的主要贡献如下:(1)基于对ANC系统组件中的非线性特性的分析,讨论了在实际AN...
【文章页数】：178 页
【学位级别】：博士
【文章目录】：
摘要
abstract
List of Abbreviations
Chapter1 Introduction
1.1 Significance and background of the research
1.2 Overview of the ANC system
1.2.1 Basic principle
1.2.2 Development of ANC system
1.3 The research situation
1.4 Motivation
1.5 The main research results of the dissertation
1.6 Organization of the Dissertation
Chapter2 Analysis of nonlinear influence and FLANN model in ANC system
2.1 Introduction
2.2 Analysis of nonlinear influence on the ANC system
2.2.1 Nonlinear influence in the reference noise
2.2.2 Nonlinear influence in the primary path
2.2.3 Nonlinear influence in the secondary path
2.3 Type of nonlinearity in ANC systems
2.3.1 Memory and Memoryless nonlinearity
2.3.2 Chaotic nonlinearity
2.4 Functional link artificial neural networks model
2.4.1 Structure
2.4.2 Nonlinear adaptive FLANN filter
2.5 The ANC system based on FLANN
2.5.1 Structure
2.5.2 The Filtered-S least mean square(FsLMS)algorithm
2.6 Analysis of the nonlinear modeling capability of FLANN for the ANC system
2.7 The performance evaluation of the FLANN-based ANC system
2.7.1 Evaluation of noise-canceling performance
2.7.2 Evaluation of computational resources performance
2.8 Conclusion
Chapter3 Simplified generalized FLANN filter for nonlinear active noise control
3.1 Introdution
3.2 Filter bank implementation of a class of nonlinear filters
3.3 The nonlinear adaptive simplified generalized FLANN(SG-FLANN)controller for ANC system
3.3.1 The generalized FLANN nonlinear filter with simplified diagonal-structure
3.3.2 The simplified generalized Fs-LMS(SGFs-LMS)algorithm
3.3.3 M-max simplified generalized filtered error least mean square (Mm SGFE-LMS) algorithm
3.4 The analysis of adaptive SG-FLANN filter in nonlinear ANC systems
3.5 Stability conditions of adaptive algorithms
3.6 Computational complexity analysis
3.6.1 Computational Complexity for NANC/LSP
3.6.2 Computational Complexity for NANC/NSP
3.7 Simulation
3.7.1 Experiment 3.1
3.7.2 Experiment 3.2
3.8 Conclusion
Chapter4 Nonlinear adaptive bilinear FLANN filter for active noise control
4.1 Introduction
4.2 The nonlinear adaptive bilinear filter
4.3 The nonlinear adaptive bilinear FLANN filter for ANC
4.3.1 The bilinear FLANN structure
4.3.2 Leaky bilinear filter x-least mean square(LBFx-LMS)algorithm
4.3.3 M-max partial update Leaky bilinear filter-error least mean square (Mm LBFE-LMS) algorithm
4.4 The bounded-input bounded-output(BIBO)stability condition of bilinear FLANN
4.5 Computational complexity analysis
4.5.1 Computational Complexity for NANC/LSP
4.5.2 Computational Complexity for NANC/NSP
4.6 Simulation
4.6.1 The nonlinear ANC with nonlinear secondary path
4.6.2 The nonlinear ANC with linear secondary path
4.7 Conclusion
Chapter5 Generalized exponential FLANN filter with channel-reduced diagonal structure for nonlinear active noise control
5.1.Introduction
5.2.Nonlinear adaptive exponential FLANN filter
5.3.The generalized E-FLANN with channel-reduced diagonal(GE-FLANN-CRD)filter for ANC
5.3.1.The generalized E-FLANN filter and its multichannel implementation
5.3.2.Generalized exponential filtered-s least mean square(GEFs-LMS)algorithm
5.3.3 M-max generalized exponential filtered-error least mean square(MmGEFE-LMS)algorithm
5.4 Convergence analysis and stability conditions
5.5 Computational complexity analysis
5.6 Simulation
5.6.1 Experiment5.1
5.6.2 Experiment5.2
5.7 Conclusion
Chapter6 Computationally efficient pipelined architecture-based adaptive generalized FLANN filter for nonlinear active noise control
6.1.Introduction
6.2 Nonlinear adaptive pipelined generalized FLANN(P-GFLANN)filter
6.2.1 P-GFLANN structure
6.2.2 Adaptive algorithm of the P-GFLANN filter
6.2.3 Stability conditions analysis
6.2.4 Computational complexity analysis
6.2.5 The performance evaluation of the P-GFLANN
6.3 The nonlinear adaptive P-GFLANN filter for ANC
6.3.1 Structure of the ANC system based on P-GFLANN
6.3.2 Pipelined generalized filtered-s least mean squre(PGFs-LMS)algorithm
6.4 Nonlinear adaptive hierarchical update P-GFLANN (HUP-GFLANN) filter for ANC
6.4.1 Structure of the NANC system based on the HUP-GFLANN filter
6.4.2 Hierarchical M-max generalized filtered-error least mean square (HMm GFE-LMS) algorithm
6.5 Computational complexity analysis for the pipelined architecture-based ANC system
6.6 Simulation
6.6.1 Experiment6.1
6.6.2 Expriment6.2
6.6.3 Experiment6.3
6.6.4 Experiment6.4
6.6.5 Experiment6.5
6.7 Conclusion
Conclusion and Future Work
Conclusions
Recommendations for Future Work
Acknowledgements
Reference
Appendix
List of Publication
本文编号：3911647
【文章页数】：178 页
【学位级别】：博士
【文章目录】：
摘要
abstract
List of Abbreviations
Chapter1 Introduction
1.1 Significance and background of the research
1.2 Overview of the ANC system
1.2.1 Basic principle
1.2.2 Development of ANC system
1.3 The research situation
1.4 Motivation
1.5 The main research results of the dissertation
1.6 Organization of the Dissertation
Chapter2 Analysis of nonlinear influence and FLANN model in ANC system
2.1 Introduction
2.2 Analysis of nonlinear influence on the ANC system
2.2.1 Nonlinear influence in the reference noise
2.2.2 Nonlinear influence in the primary path
2.2.3 Nonlinear influence in the secondary path
2.3 Type of nonlinearity in ANC systems
2.3.1 Memory and Memoryless nonlinearity
2.3.2 Chaotic nonlinearity
2.4 Functional link artificial neural networks model
2.4.1 Structure
2.4.2 Nonlinear adaptive FLANN filter
2.5 The ANC system based on FLANN
2.5.1 Structure
2.5.2 The Filtered-S least mean square(FsLMS)algorithm
2.6 Analysis of the nonlinear modeling capability of FLANN for the ANC system
2.7 The performance evaluation of the FLANN-based ANC system
2.7.1 Evaluation of noise-canceling performance
2.7.2 Evaluation of computational resources performance
2.8 Conclusion
Chapter3 Simplified generalized FLANN filter for nonlinear active noise control
3.1 Introdution
3.2 Filter bank implementation of a class of nonlinear filters
3.3 The nonlinear adaptive simplified generalized FLANN(SG-FLANN)controller for ANC system
3.3.1 The generalized FLANN nonlinear filter with simplified diagonal-structure
3.3.2 The simplified generalized Fs-LMS(SGFs-LMS)algorithm
3.3.3 M-max simplified generalized filtered error least mean square (Mm SGFE-LMS) algorithm
3.4 The analysis of adaptive SG-FLANN filter in nonlinear ANC systems
3.5 Stability conditions of adaptive algorithms
3.6 Computational complexity analysis
3.6.1 Computational Complexity for NANC/LSP
3.6.2 Computational Complexity for NANC/NSP
3.7 Simulation
3.7.1 Experiment 3.1
3.7.2 Experiment 3.2
3.8 Conclusion
Chapter4 Nonlinear adaptive bilinear FLANN filter for active noise control
4.1 Introduction
4.2 The nonlinear adaptive bilinear filter
4.3 The nonlinear adaptive bilinear FLANN filter for ANC
4.3.1 The bilinear FLANN structure
4.3.2 Leaky bilinear filter x-least mean square(LBFx-LMS)algorithm
4.3.3 M-max partial update Leaky bilinear filter-error least mean square (Mm LBFE-LMS) algorithm
4.4 The bounded-input bounded-output(BIBO)stability condition of bilinear FLANN
4.5 Computational complexity analysis
4.5.1 Computational Complexity for NANC/LSP
4.5.2 Computational Complexity for NANC/NSP
4.6 Simulation
4.6.1 The nonlinear ANC with nonlinear secondary path
4.6.2 The nonlinear ANC with linear secondary path
4.7 Conclusion
Chapter5 Generalized exponential FLANN filter with channel-reduced diagonal structure for nonlinear active noise control
5.1.Introduction
5.2.Nonlinear adaptive exponential FLANN filter
5.3.The generalized E-FLANN with channel-reduced diagonal(GE-FLANN-CRD)filter for ANC
5.3.1.The generalized E-FLANN filter and its multichannel implementation
5.3.2.Generalized exponential filtered-s least mean square(GEFs-LMS)algorithm
5.3.3 M-max generalized exponential filtered-error least mean square(MmGEFE-LMS)algorithm
5.4 Convergence analysis and stability conditions
5.5 Computational complexity analysis
5.6 Simulation
5.6.1 Experiment5.1
5.6.2 Experiment5.2
5.7 Conclusion
Chapter6 Computationally efficient pipelined architecture-based adaptive generalized FLANN filter for nonlinear active noise control
6.1.Introduction
6.2 Nonlinear adaptive pipelined generalized FLANN(P-GFLANN)filter
6.2.1 P-GFLANN structure
6.2.2 Adaptive algorithm of the P-GFLANN filter
6.2.3 Stability conditions analysis
6.2.4 Computational complexity analysis
6.2.5 The performance evaluation of the P-GFLANN
6.3 The nonlinear adaptive P-GFLANN filter for ANC
6.3.1 Structure of the ANC system based on P-GFLANN
6.3.2 Pipelined generalized filtered-s least mean squre(PGFs-LMS)algorithm
6.4 Nonlinear adaptive hierarchical update P-GFLANN (HUP-GFLANN) filter for ANC
6.4.1 Structure of the NANC system based on the HUP-GFLANN filter
6.4.2 Hierarchical M-max generalized filtered-error least mean square (HMm GFE-LMS) algorithm
6.5 Computational complexity analysis for the pipelined architecture-based ANC system
6.6 Simulation
6.6.1 Experiment6.1
6.6.2 Expriment6.2
6.6.3 Experiment6.3
6.6.4 Experiment6.4
6.6.5 Experiment6.5
6.7 Conclusion
Conclusion and Future Work
Conclusions
Recommendations for Future Work
Acknowledgements
Reference
Appendix
List of Publication
本文编号：3911647
本文链接：https://www.wllwen.com/shoufeilunwen/xxkjbs/3911647.html
