Abstract With the continuous development of information technology, data resources have become increasingly rich, but the knowledge contained in data resources has not been fully explored and utilized. In order to find an effective computer testing technology, the paper introduces the partial differential equation (PDE) into the denoising process of rotor vibration signal through data mining technology, and generalizes the unified model of PDE filtering. Several filtering methods are compared through simulation experiments. The effect is that the flexible rotor is balanced by different dynamic balancing methods, and satisfactory results are obtained. From the simulation results, it can be concluded that the integration method is not suitable to extract the unbalanced signal with strong noise background, but it provides a way to calculate the amplitude and phase of sinusoidal signal without noise; The processing is simple and suitable for the calculation of the dynamic balance test system with fewer sampling points; both the DFT method and the FFT method use the principle of Fourier transform spectrum analysis, but the FFT method calculates the speed much faster than the DFT method. Experiments on the classification of fault data prove that the improved Apriori algorithm is greatly improved compared with the original Apriori algorithm, and the speed of acquiring fault rules is improved.

中文翻译：

---

机械工程计算机测试系统数据挖掘技术

摘要 随着信息技术的不断发展，数据资源日益丰富，但数据资源中所包含的知识还没有得到充分的挖掘和利用。为了寻找一种有效的计算机测试技术，本文通过数据挖掘技术将偏微分方程(PDE)引入转子振动信号的去噪过程，概括了偏微分方程滤波的统一模型。通过仿真实验比较了几种滤波方法。其效果是通过不同的动平衡方法对柔性转子进行平衡，取得了满意的效果。从仿真结果可以得出结论，积分方法不适合提取具有强噪声背景的不平衡信号，但它提供了一种计算无噪声正弦信号幅度和相位的方法；处理简单，适用于采样点较少的动平衡测试系统的计算；DFT方法和FFT方法都采用傅里叶变换频谱分析的原理，但FFT方法的计算速度比DFT方法快得多。通过对故障数据分类的实验证明，改进后的Apriori算法较原Apriori算法有很大的改进，提高了获取故障规则的速度。DFT方法和FFT方法都采用傅里叶变换频谱分析的原理，但FFT方法的计算速度比DFT方法快得多。通过对故障数据分类的实验证明，改进后的Apriori算法较原Apriori算法有很大的改进，提高了获取故障规则的速度。DFT方法和FFT方法都采用傅里叶变换频谱分析的原理，但FFT方法的计算速度比DFT方法快得多。通过对故障数据分类的实验证明，改进后的Apriori算法较原Apriori算法有很大的改进，提高了获取故障规则的速度。