Model-Based Analysis and Design Optimization of Music Instruments

该研究项目的长期目标是提高我们对乐器空气声学和机械特性的科学理解，这有助于乐器的设计、制造、教学和声音合成。乐器已经通过经验方法发展了几个世纪，并达到了非常高的精致水平。尽管如此，制造商、制造商和玩家仍然遇到各种问题，可以找到解决方案。仪器设计常常涉及妥协。例如，为改善语调而进行的修改可能会无意中产生不太一致的音色。可以应用计算机建模方法来寻找优化或定制的解决方案。我们的重点主要是乐器，尽管进行这项研究时获得的研究方法、结果和技能将适用于广泛的技术、多媒体和声学工程领域。 本研究计划的第一个具体目标是通过测量和计算机建模更好地了解木管乐器吹口的几何、空气声学和演奏参数之间的联系。该提案的这一部分是先前目标的延续，以先前的成就为基础。我们将利用有限元 (FE) 和气动声学建模来表征音质（声学）和演奏性（流体-结构-相互作用）特性。这项研究还将涉及使用热线风速计和压力探头以及机械木管演奏器系统进行测量，以评估流体流动特性并为 3D 空气声学建模提供信息。 该提案的第二个目标是开发用于马林巴琴杆设计的计算机建模方法，该方法可以优化调音，最大限度地减少不良的交叉模式耦合，并支持使用复合材料替代濒临灭绝的红木的探索。该方法将结合有限元和优化方法，并进行扩展以考虑声辐射、与马林巴共鸣管的耦合以及可听性。我们将与天然纤维复合材料设计和制造方面的专家合作，寻找可以替代红木的新材料，同时保持与传统马林巴琴相同的音质。 该提案的第三个目标将涉及算法系统的设计和开发，该系统可以在使用基于虚拟物理的合成模型或机械演奏器系统进行表演时学习模仿人类音乐家的音乐表现力。该方法将利用人工智能和机器学习以及机器人技术和音乐教育学的方法。 这项研究的结果将改善当前乐器设计/分析/评估方面的工作，告知教师和学生正确或更有效的演奏乐器的方法，并且还可以应用于商业声学工程环境。

The long--term goal of this research program is to improve our scientific understanding of the aeroacoustic and mechanical properties of music instruments that can aid in their design, manufacture, pedagogy and sound synthesis. Musical instruments have been developed over centuries through empirical methods and have reached very high levels of refinement. That said, there remain various problems experienced by makers, manufacturers and players for which solutions can be found. Instrument designs often involve compromises. For example, a modification made to improve intonation may inadvertently produce a less consistent timbre. Computer modeling methods can be applied to find optimized or tailored solutions. Our focus is primarily on music instruments, though the research methods, results and skills acquired in conducting this research will be applicable in a wide range of technological, multimedia, and acoustical engineering fields. The first specific objective of this research proposal is to better understand the link between geometric, aeroacoustic and playing parameters of woodwind instrument mouthpieces using measurements and computer modeling. This component of the proposal is a continuation of a previous objective, building on previous achievements. We will make use of Finite Element (FE) and aeroacoustic modeling to characterize the tone quality (acoustic) and playability (fluid-structure-interaction) properties. This research will also involve measurements using a hot-wire anemometer and pressure probes with a mechanical woodwind player system to assess the fluid flow characteristics and inform the 3D aeroacoustic modeling. The second objective of the proposal is to develop computer modeling approaches for the design of marimba bars that can optimize for tuning, minimize undesirable cross-mode coupling, and support explorations with composite materials as a replacement for endangered rosewood. The approach will combine FE and optimization methods, with extensions to account for sound radiation, coupling with marimba resonator pipes, and auralization. We will collaborate with specialists in the design and fabrication of natural fiber composite materials to seek new materials that can replace rosewood while maintaining the same timbral quality as traditional marimbas. The third objective of the proposal will involve the design and development of algorithmic systems that can learn to imitate the musical expressivity of human musicians when performing with virtual physics-based synthesis models or with a mechanical player system. The approach will make use of methods from artificial intelligence and machine learning, as well as robotics and music pedagogy. Results of this research will improve current efforts in the design / analysis / evaluation of music instruments, inform teachers and students about proper or more efficient ways of playing their instruments, and also have application in commercial acoustical engineering contexts.