摘要
根据托辊出现故障的原因,分析了托辊的三种主要故障形式——筒皮磨损、托辊轴承失效以及托辊弯曲变形。托辊在出现以上故障过程中会产生声音、图像、温度、转速等信号的变化特征,均可反应托辊的故障程度。现有的各类传感器已广泛应用于托辊的故障诊断,但缺少评价托辊故障等级模型,无法根据托辊的运行工况提供智能监测。因此通过将现有的托辊故障诊断技术相结合,构建托辊故障等级预测模型,并采集了某企业的圆管带式输送机实验数据,利用RBF神经网络模型对实验数据进行处理,验证了其模型的可靠性,该模型便于判别托辊的故障程度,为托辊的智能化维护及快速更换提供理论支撑。
According to the causes of the failure of idlers,three main fault forms of idlers-cylinder skin wear,roller bearing failure and roller bending deformation are analyzed.In the process of the above faults,the idlers will produce the changing characteristics of sound,image,temperature,rotational speed and other signals,which can reflect the fault degree of the idlers.All kinds of existing sensors have been widely used in the fault diagnosis of idlers,but there is a lack of fault grade model for evaluating idlers,so it is impossible to provide intelligent monitoring according to the operating conditions of idlers.Therefore,by combining the existing fault diagnosis technology of idlers,the fault grade prediction model of idlers is constructed,and the experimental data of pipe belt conveyor in an enterprise are collected,and the RBF neural network model is used to process the experimental data to verify the reliability of the model.The model is convenient to distinguish the fault degree of idlers and provides theoretical support for intelligent maintenance and rapid replacement of idlers.
作者
高波
袁媛
岳伟
张鑫增
GAO Bo;YUAN Yuan;YUE Wei;ZHANG Xinzeng(School of Transportation and Logistics,Taiyuan University of Science and Technology,Taiyuan 030024,China)
出处
《物流科技》
2023年第13期32-35,共4页
Logistics Sci-Tech
基金
山西省科技成果转化引导专项项目(202104021301062)
太原科技大学研究生教育创新项目(SY2022068)。
