A hybrid numerical-experimental methodology for robust assessment of automotive brake system noise

几十年来，刹车噪音一直是汽车制造商面临的一个问题，并导致消费者对质量状况的感知，而这反过来又会通过保修索赔对制造商产生成本影响。这是一个复杂的多方面的问题。尽管对刹车噪声的广泛研究已经进行了几十年，但在早期设计阶段还没有全面的预测方法。相反，制造商使用测功机测试和有限元建模技术相结合的方法来确定可能影响制动噪声恶化的因素。这些方法是采用在一个阶段时，制动系统已经完成。这就导致了噪声控制方法通过复杂变化到变量组合在一个迭代过程中来减少或消除噪声的条件。这些对策对某一特定制动系统非常敏感，不适用于其他制动系统；因此，该过程必须反复重复不同的制动系统，导致一个昂贵和耗时的过程。为了在达到最终设计之前缓解这种情况，FCA（菲亚特克莱斯勒汽车公司）正在寻求开发一种更强大的方法，能够提供可靠和可重复的结果，基于对制动噪声背后的科学的更好理解。这就是这项研究背后的动机。其主要目标是开发一个强大的预测工具，以便更好地理解与刹车噪音有关的现象，重点是尖叫噪音。主要规范是改善实验和数值预测之间的相关性，以便该工具可以在早期设计阶段和过程中使用，而不是作为现有系统的诊断工具。这将通过在建模过程中添加不确定性和可变性来实现，并借助由受控实验室实验和测功机测试平台获得的实验数据库。更重要的是，该工具还可以预测盘式制动器的声辐射，这是当前工具和实践中经常忽略的任务。我们将为合作伙伴提供一个工程工具，以及相关的方法，用于制动噪声预测和设计挑战。该知识工具箱将允许合作伙伴在其产品设计周期中更好地集成制动系统噪声。

Brake noise has been an issue for automotive manufacturers for decades and has led to customer perceived quality conditions, which in turn has a cost implication to manufacturers through warranty claims. It is a complex multifaceted problem. Although extensive research on brake noise has been conducted over decades, no comprehensive predictive methodology at the early design stage exists. Rather, a combination of dynamometer testing and finite element modeling techniques are used by the manufacturers to determine factors that may be influential in the aggravation of brake noise. These methods are employed at a stage when the brake system has already been finalized. This has led to noise control methods through complex changes to a combination of variables in an iterative process to reduce or eliminate noise conditions. These countermeasures are very keen to a particular brake system and are not applicable to other brake systems; hence, the process has to be repeated repeatedly for different brake systems resulting in a costly and time-consuming process. With the aim of mitigating this condition before reaching the final design, FCA (Fiat Chrysler Automobiles) is seeking to develop a more robust methodology, which has the ability to provide reliable and repeatable results, based on a better understanding of the sciences behind brake noise. This is the motivation behind this research. Its main objective is to develop a robust predictive tool that allows for a better understanding of the phenomena pertaining to brake noise with an emphasis on squeal noise. The main specification is to improve the correlation between the experimental and the numerical predictions so that the tool can be used at early design phases and processes rather than a diagnostic tool for existing systems. This will be achieved by adding uncertainty and variability in the modeling process with the help of experimental databases obtained with controlled laboratory experiments and dynamometer test rigs. More importantly, the tool will also predict disc brake sound radiation, a task often ignored in current tools and practices. The partner will be provided with an engineering tool together with the relevant methodologies dedicated to their brake noise prediction and design challenges. This knowledge toolbox will allow the partner a better engineering integration of brake system noise all along their products design cycle.