超小波边界元理论及其在机械故障机理研究中的应用

In recent years, the investigation of mechanical fault mechanism has become the research focus and big bone problem in fault diagnosis research field. Theorteical analysis, numerical simulation and experimental analysis were developed to prove the mechanical fault mechanism. Amony these, numerical simulation is the best way to investigate the mechanical mechanism for unkown and weak faults. However, the present methods have the bottleneck problems, such as the low solution efficiency, accuracy and reliability, etc. The beyond wavelet boundary element method (BEM) for the investigation of mechanical fault mechanism, is proposed and in depth researched. The key issue is to solving some important and difficult problems, such as how to deal with elemental boundary node of beyond wavelet-based multi-scale elements, the enhancement of geometric approximation ability of BEM and the sparsification of coefficient matrix of BEM, the determination of characteristic parameters for the contanct surface in mechancial systems, the efficency model updating technique, the weak feature signal extraction, and so on. The beyond wavelet BEM will be investigated. And further we will clarify the fast calculation principle for the beyond wavelet BEM and to describe its advantages. On the basis of the above studies, the beyond wavelet BEM models for mechancial systems will be proposed. Rely on the models, the mechanical fault mechanism will be discovered. Through this project investigation, we expect that it will lead to innovations and breakthroughs for some aspects, i.e., the beyond wavelet-based multi-scale elements, the beyond wavelet BEM, the numerical simulation for mechanical systems, the unkown and weak fault characterization, the fault detection indicators in time domain, frequency domain, time-frequency domain, etc. The basic research achievements will be obtained with both academic implication and practical value.

故障机理研究，是近年来故障诊断领域研究的热点和难点，发展了理论分析、数值模拟、实验分析手段，尝试探明故障机理。而数值模拟是研究未知和微弱故障机理的最佳手段，但目前方法存在着求解效率、求解精度和求解可靠性低等瓶颈问题。本项目提出并开展用于故障机理研究的超小波边界元方法研究，重点解决超小波多尺度单元边界节点处理、增强边界元几何逼近能力和稀疏化边界元求解系数矩阵、机械系统接触面的特性参数确定、模型有效修正、微弱特征信息提取等问题。研究超小波边界元，阐明其计算快速性原理和优势。在上述研究基础上，提出机械系统超小波边界元分析模型,开展机械故障机理探索。通过本项目研究，可望在超小波多尺度单元、超小波边界元方法、机械系统数值模拟、未知和微弱故障表征、以及时域、频域和时频域诊断指标等方面有所创新和突破，获得有学术意义和实用价值的基础性成果。

机械故障诊断在工程实践中取得了大量卓有成效应用，该领域尚存在着故障机理研究不足、诊断方法有限和智能诊断系统薄弱等问题，而故障机理研究不足是导致其他问题的根源所在。因此，故障机理研究，是近年来故障诊断领域研究的热点和难点，发展了理论分析、数值模拟、实验分析手段，尝试探明故障机理。而数值模拟是研究未知和微弱故障机理的最佳手段，但目前方法存在着求解效率、求解精度和求解可靠性低等瓶颈问题。本项目提出并开展用于故障机理研究的小波边界元方法研究，重点解决超小波边界元求解格式、机械系统接触面的特性参数确定、模型有效修正、微弱特征信息提取等问题。研究超小波边界元，阐明其计算快速性原理和优势。在上述研究基础上，提出了机械系统超小波边界元分析模型,开展机械故障机理探索。通过本项目研究，在小波边界元方法、机械系统数值模拟、未知和微弱故障表征、以及时域、频域和时频域诊断指标、具体的故障诊断方法等方面有所创新和突破，获得了一些有学术意义和实用价值的基础性成果。发表期刊论文31篇，其中SCI收录期刊论文29篇；国际会议论文9篇；授权发明专利10项、获软件著作权登记2项； 2019年担任SCI期刊专辑(2018IF：1.628)客籍主编、2017年担任SCI期刊专辑(2018IF：1.614)客籍主编。培养硕士研究生10名。项目负责人获得2016年中国侨联创新团队奖、2016年广西科技成果二等奖各1项、2019年浙江省科技成果登记1项。