基于AMESim的齿轮泵内泄漏故障预测技术研究

Research on the Gear Pump Fault Prediction Based on AMESim Simulation Platform

采用AMESim软件构建了齿轮泵内泄漏故障仿真平台,通过运行仿真平台得到齿轮泵泄漏的故障仿真数据。运用故障仿真数据,采取基于支持向量机的故障预测方法,在Matlab中建立了故障预测模型,将预测数据与仿真数据进行对比,计算出预测的误差,并对该模型的故障预测精度进行了评估验证。讨论了基于不同数量样本建模对故障预测精度的影响,为选择合理的样本数量构建预测模型提供了参考。结果表明:建模数据的量越大,预测的误差越小,对齿轮泵的故障预测具有很好的应用价值。提出的思路和方法对齿轮泵内泄漏故障预测技术应用具有一定的参考价值。

Firstly,the gear pump internal leakage fault simulation platform is constructed by the AMEblm software, and the fault simulation data can be obtained through the simulation platform. With the fault prediction method based on support vector machine （SVM）,a fault prediction model is set up in Matlab,and the fault simulation data are predicted by the fault prediction model. Then, the predicted data are acquired, and the fault prediction precision validation of the model is evaluated through the simulation data. In addi- tion, the paper also elaborates the influence on the fault prediction accuracy by modeling based on different numbers of samples,thus offering the reference for selecting the reasonable number of samples. The findings indicate that the high quantity of modeling data is, the lower error of the prediction is, which is of the significant application value for fault prediction of gear pumps. The idea and method presented serve as the reference for the application of the pump gear leakage failure prediction technology.