摘要
机械设备中的主轴SVPWM逆变器存在死区效应,造成故障输出电流波形严重畸变和零电流钳位现象。通过分析由设备信号死区时间、功率器件导通关断时间引起的误差电压矢量,在传统的机械设备早期故障信号提纯优化算法基础上,提出了一种通过补偿EMD电压前馈死区信号的改进型机械设备早期故障信号提纯算法。基于希尔伯特变换法,可以从满足限制条件的固有模态函数中得到有意义的瞬间故障频率特征,并将其分解成各个固有模态函数分量。然后判断各分量所在扇区并通过EMD进行补偿,进而完成信号的有效提纯,并解决瞬间形变问题。实验结果表明,该算法可以有效地改善输出故障信号的波形,提高故障检测准确度。
There is a dead zone effect in the spindle SVPWM inverter of the mechanical equipment, which leads to a serious distortion in the fault signal output current waveform and the phenomenon of zero current clamp. After analyzing the error voltage vector caused by the dead time of equipment signal and the power device turn - off time, an improved early fault signal purification algorithm for the mechanical equipment was proposed, which works by com- pensating EMD voltage feedforward dead zone signal. Based on the Hilbert transform algorithm, the meaningful tran- sient fault frequency characteristics were obtained from the satisfactory limited intrinsic mode functions, and divided into multiple intrinsic mode function components. To purify the signals and solve the instantaneous deformation effec- tively, the sector of each intrinsic mode function component was judged and compensated by EMD. The experimental results show that the proposed algorithm can improve the output waveform of the fault signal and the accuracy of the fault detection effectively.
出处
《计算机仿真》
CSCD
北大核心
2014年第2期293-296,449,共5页
Computer Simulation
关键词
故障检测
死区效应
死区补偿
信号提纯
Fault detection
Dead zone effect
Dead zone compensation
Purification of signal
