外部随机激励干扰下机械齿轮折断故障模糊性诊断系统

Fuzzy diagnosis system of mechanical gearbreaking fault under external random excitation

在外部随机激励干扰环境下,为及早发现机械齿轮中的潜在故障,通过模糊性判断,设计了一种机械齿轮折断故障诊断系统。系统分为硬、软件两部分。传感器与采集器是硬件部分的关键组件,为软件流程提供支持。在软件部分,根据随机噪声激励干扰环境中齿轮折断故障的特征选取时变参数,在优化加速度振动传感器后,利用模糊算法设计了模糊预测模型,在对模型实施动态优化、更新修整后,设计训练程序用于训练模型参数,设计监测程序实时监测机械齿轮运行状态,及时发出故障预警。仿真试验中,针对直列式六缸发动机的齿轮传动结构建立仿真模型,分别从有效性、适用性、稳定性以及自适应性4个方面,对系统进行了验证,发现所提系统能够精准预警不同种类齿轮样本的折断故障。

In order to detect potential faults in mechanical gears as early as possible under external random excitation interference environment,a mechanical gear fracture fault diagnosis system is designed through fuzzy judgment.The system is divided into two parts:hardware and software.Sensors and collectors are key components of the hardware part,providing support for software processes.In the software section,time-varying parameters are selected based on the characteristics of gear fracture faults in random noise excitation interference environments.After optimizing the acceleration vibration sensor,a fuzzy prediction model is designed using fuzzy algorithms.After dynamic optimization and updating of the model,a training program is designed to train model parameters.A monitoring program is designed to monitor the operating status of mechanical gears in real time and issue fault warnings in a timely manner.The system is validated from four aspects:effectiveness,applicability,stability,and adaptability.It is found that the proposed system can accurately warn of fracture faults of different types of gear samples.