首页|期刊导航|计量学报|基于低采样AD量化的标准振动台失真度测试方法

基于低采样AD量化的标准振动台失真度测试方法OA

Distortion Measurement Method Based on Low Sampling AD Quantization for Standard Vibrator

中文摘要英文摘要

为实现低采样率AD量化下标准电磁振动台输出振动激励信号中谐波成分的准确分析,以满足振动校准工作中高精度总谐波失真度(THD)参数获取需求.首先,基于AD7606芯片搭建了标准振动台输出振动激励信号的振动数据低速采样模块,并通过实验分析了低采样AD量化过程对振动激励信号THD测试误差的影响.其次,提出一种基于局部离群因子(LOF)均匀采样的最小二乘支持向量机(LSSVM)误差校正方法,通过辨识到的误差校正模型对低采样AD量化下的误差数据进行修正,进而得到更加准确的各次谐波及THD等关键参数.最终仿真及实验结果表明:基于LOF均匀采样的LSSVM误差校正方法辨识到的误差校正模型相较于常规误差校正具有更高的拟合准确度,能够将低速采样模块的THD分析结果与NI采集卡之间的误差控制在0.45%以内.

To accurately analyze the harmonic components in the output excitation signal of the standard vibrator under low-sampling AD quantization and meet the high-precision total harmonic distortion(THD)acquisition requirements in vibration calibration.First,a low-speed sampling module for the vibration data of the standard vibrator's output excitation signal was built using the AD7606 chip.The influence of low-sampling AD quantization on the resulting THD test error of the excitation signal was analyzed experimentally.Second,an error correction method based on a least squares support vector machine(LSSVM)combined with a local outlier factor(LOF)uniform sampling was proposed.The error data caused by low-sampling AD quantization were corrected using the identified error correction model,leading to more accurate extraction of key parameters such as individual harmonics and THD.The final simulation and experimental results show that the LSSVM error correction model identified using the LOF-based uniform sampling strategy achieves higher fitting accuracy compared to conventional error correction methods,and is able to limit the error between the THD analysis results of the low-speed sampling module and the NI data acquisition card to within 0.45%.

张旭飞;魏鑫;曹承乾;王硕;武艺凡

太原理工大学 机械工程学院,山西 太原 030024||新型传感器与智能控制教育部和山西省重点实验室,山西 太原 030024太原理工大学 机械工程学院,山西 太原 030024太原理工大学 机械工程学院,山西 太原 030024太原理工大学 机械工程学院,山西 太原 030024太原理工大学 机械工程学院,山西 太原 030024

通用工业技术

力学计量标准振动台总谐波失真度AD采样最小二乘支持向量机局部离群因子

mechanical metrologystandard vibratortotal harmonic distortionAD samplingleast squares support vector machinelocal outlier factor

《计量学报》 2026 (3)

374-379,6

国家自然科学基金(52375544)山西省基础研究计划(202203021211152)山西省研究生实践创新项目(2025SJ125)

10.3969/j.issn.1000-1158.2026.03.08

评论