论文部分内容阅读
传感器状态对于凿岩台车的作业有着极其重要的影响,对其展开故障诊断十分必要.核主成分分析(KPCA)方法通过集成算子与非线性核函数计算高维特征空间的主元成分,有效捕捉过程变量中的非线性关系,将其用于传感器4种常见故障的诊断,先用Q统计量进行故障监测,再用T2贡献量百分比变化来识别故障.仿真和实际应用结果表明:KPCA方法具有很好的故障监测与诊断能力.
Sensor status is very important for the operation of rock drilling rig and it is necessary to diagnose it.Kernel Principal Component Analysis (KPCA) method calculates the principal component of high-dimensional feature space by integrating operator and nonlinear kernel function, Effectively capture the nonlinear relationship among process variables and use it to diagnose four kinds of common faults of sensors, firstly use Q statistics to monitor the faults and then identify the faults with the percentage change of T2 contribution.The simulation and practical application show that KPCA The method has good fault monitoring and diagnostic capabilities.