基于FOA-GRNN方法的工业机器人交叉滚子轴承寿命预测
秦亮亮 2024/12/15 18:44:56
江苏联合职业技术学院淮安生物工程分院,江苏淮安 223200
摘要:交叉滚子轴承在工业机器人上得到广泛应用,其使用寿命直接影响到工业机器人经济成本。为保证广义回归神经网络(GRNN)的更高预测精度,采用多种群自适应果蝇优化算法(FOA)对其扩展速度进行优化,构建了基于FOA-GRNN方法的工业机器人交叉滚子轴承寿命预测方法。研究结果表明:通过FOA-GRNN方法预测具有较高的结果。相对于单独的FOA和GRNN方法,采用FOA-GRNN方法各项指标均是最小的,验证了FOA优化GRNN方法的有效性,实现了寻优效率与精度的提升。该研究有助于提高工业机器人的运行寿命,具有很高的节能意义。
关键词:工业机器人;交叉滚子轴承;使用寿命;广义回归神经网络
中图分类号:TH137
Life prediction of industrial robot cross roller bearing based on FOA-GRNN method
Qin Liangliang
Huaian Bioengineering Branch, Jiangsu United Vocational and Technical College, Huaian 223200, China
Abstract: Cross roller bearings are widely used in industrial robots, and their service life directly affects the economic cost of industrial robots. In order to ensure the higher prediction accuracy of generalized regression neural network (GRNN), the multi-population adaptive Fruit Fly optimization algorithm (FOA) was used to optimize its expansion speed, and a FOA-GRNN based life prediction method of industrial robot cross roller bearing was constructed. The results show that the FOA-GRNN method has higher results. Compared with the separate FOA and GRNN methods, the FOA-GRNN method has the smallest indexes, which verifies the effectiveness of FOA optimization GRNN method, and improves the optimization efficiency and accuracy. This research is helpful to improve the operation life of industrial robots and has high energy saving significance.
Key words: industrial robot; Cross roller bearing; Service life; Generalized regression neural networks
1引言
目前,交叉滚子轴承已成为工业机器人传动系统关键部件之一,检测交叉滚子轴承状态能够实现判断工业机器人故障,并进一步排斥潜在的隐患[1-2]。因为构建有效方法难度较大,实际预测结果与真实运行状态存在明显偏差,因此需要进一步开展研究[3-4]。
相关方面的研究吸引了很多的学者,取得了一定的研究成果。邓飞跃等[5]提出多尺度时间卷积网络与Transformer融合模型用于轴承寿命预测,准确学习时序特征与轴承寿命之间的映射关系,验证了所提方法在变工况下所提取的时序特征泛化性较好。潘雪娇等[6]结合时间卷积网络和残差自注意力机制提出端到端的轴承寿命迁移预测,增强模型迁移特征提取能力,获得较好的预测性能。曹胜博等[7]提出基于双通道信息融合与门控单元神经网络的轴承寿命预测,避免传统算法过分依赖专家判断的弊端,双通道数据训练出的门控神经网络模型的预测结果更为准确。
本文为保证GRNN达(未完,下一页)
附件下载:基于FOA-GRNN方法的工业机器人交叉滚子轴承寿命预测
|