针对额外提取数据特征的方法需要花费大量时间和人力成本、轴承退化的线性预测精度低等问题,以及时序数据具有时间依赖关系的特点,提出了端到端的结合长短时记忆网络的深度变分自编码器模型(E2E Deep VAE-LSTM)用于轴承退化预测。通过改...针对额外提取数据特征的方法需要花费大量时间和人力成本、轴承退化的线性预测精度低等问题,以及时序数据具有时间依赖关系的特点,提出了端到端的结合长短时记忆网络的深度变分自编码器模型(E2E Deep VAE-LSTM)用于轴承退化预测。通过改进VAE的结构,并结合LSTM,该模型可以在含有异常值的数据集上直接进行训练和预测;使用系统重建误差表征轴承退化趋势,实现了轴承退化的非线性预测。在三个真实数据集上的实验结果表明,E2E Deep VAE-LSTM模型可以得到满意的预测结果,预测精度均高于现有的几种AE类模型及其他几种方法,且具有良好的泛化能力和抗过拟合能力。展开更多
异常检测是保障飞机运行安全的重要手段,现有的固定阈值异常检测方法对数据时序特征利用较少,提取特征的能力较差。为提高飞机运行安全,提出了一种基于LSTM_AE神经网络的无监督离线异常检测的模型,对实际飞行数据进行异常检测。首先使用...异常检测是保障飞机运行安全的重要手段,现有的固定阈值异常检测方法对数据时序特征利用较少,提取特征的能力较差。为提高飞机运行安全,提出了一种基于LSTM_AE神经网络的无监督离线异常检测的模型,对实际飞行数据进行异常检测。首先使用LSTM(Long Short Term Memory)网络提取正常飞行数据的深度时序特征,再基于AE(Auto Encoder)对提取到的时序特征进行训练,利用模型收敛后得到重构误差确定自适应阈值,最后根据训练好的模型和自适应阈值进行异常检测。试验利用NASA公开的ALFA数据集。结果表明:基于LSTM_AE方法优于传统的固定阈值检测方法,可以实现对异常的检测,准确率为0.8717,召回率为0.9872,F1分数为0.9258。展开更多
空调热交换器性能异常检测技术是快速判断民机空调系统运行状态并合理安排维修任务的关键,传统的异常检测方法难以有效处理高维时序数据,无法实现系统早期故障预警。为此,本文提出了一种基于长短期记忆网络(LSTM,long-short term memory...空调热交换器性能异常检测技术是快速判断民机空调系统运行状态并合理安排维修任务的关键,传统的异常检测方法难以有效处理高维时序数据,无法实现系统早期故障预警。为此,本文提出了一种基于长短期记忆网络(LSTM,long-short term memory)与自编码器(AE,autoencoder)模型的无监督异常检测方法,用以识别民机空调系统异常运行状态。首先,基于民机空调系统原始传感器参数构建表征空调热交换器性能的特征监测参数;其次,构建LSTM-AE模型进行数据特征重构并计算重构误差;最后,使用孤立森林(iForest, isolation forest)进行无监督异常监测。将本文构建的无监督异常检测方法与传统方法对比,并建立模型评估指标,验证结果表明,所构建的模型方法可以对民机空调热交换器性能异常状态进行有效检测。展开更多
Currently,the inexorable trend toward the electrification of automobiles has heightened the prominence of road noise within overall vehicle noise.Consequently,an in-depth investigation into automobile road noise holds...Currently,the inexorable trend toward the electrification of automobiles has heightened the prominence of road noise within overall vehicle noise.Consequently,an in-depth investigation into automobile road noise holds substantial practical importance.Previous research endeavors have predominantly centered on the formulation of mechanism models and data-driven models.While mechanism models offer robust controllability,their application encounters challenges in intricate analyses of vehicle body acoustic-vibration coupling,and the effective utilization of accumulated data remains elusive.In contrast,data-driven models exhibit efficient modeling capabilities and can assimilate conceptual vehicle knowledge,but they impose stringent requirements on both data quality and quantity.In response to these considerations,this paper introduces an innovative approach for predicting vehicle road noise by integrating mechanism-driven and data-driven methodologies.Specifically,a series model is devised,amalgamating mechanism analysis with data-driven techniques to predict vehicle interior noise.The simulation results from dynamic models serve as inputs to the data-driven model,ultimately generating outputs through the utilization of the Long Short-Term Memory with Autoencoder(AE-LSTM)architecture.The study subsequently undertakes a comparative analysis between different dynamic models and data-driven models,thereby validating the efficacy of the proposed series vehicle road noise prediction model.This series model,encapsulating the rigid-flexible coupling dynamic model and AE-LSTM series model,not only demonstrates heightened computational efficiency but also attains superior prediction accuracy.展开更多
火箭地面测发控系统流程复杂,为自动检测系统运行流程中的异常,本文提出一种基于多变量长短时记忆自编码器(LSTM-AE,Long-short Term Memory-Autoencoder)方法的测发控系统运行异常检测方法。首先对正常系统运行数据进行分析,对系统中...火箭地面测发控系统流程复杂,为自动检测系统运行流程中的异常,本文提出一种基于多变量长短时记忆自编码器(LSTM-AE,Long-short Term Memory-Autoencoder)方法的测发控系统运行异常检测方法。首先对正常系统运行数据进行分析,对系统中开关量和数字量数据进行归一化处理,并系统中模拟量信息进行统计特征提取。然后将提取的模拟量特征和其他数字量信息组成系统运行特征向量,并作为LSTM-AE的样本输入,经自编码学习自动识别系统状态,并建立系统状态迁移异常检测模型。通过对某火箭地面测发控流程监测数据的自动建模,实现了对仿真注入故障的高效实时检测。仿真结果表明,本文所提的自编码方法优于传统的BP神经网络方法。展开更多
文摘针对额外提取数据特征的方法需要花费大量时间和人力成本、轴承退化的线性预测精度低等问题,以及时序数据具有时间依赖关系的特点,提出了端到端的结合长短时记忆网络的深度变分自编码器模型(E2E Deep VAE-LSTM)用于轴承退化预测。通过改进VAE的结构,并结合LSTM,该模型可以在含有异常值的数据集上直接进行训练和预测;使用系统重建误差表征轴承退化趋势,实现了轴承退化的非线性预测。在三个真实数据集上的实验结果表明,E2E Deep VAE-LSTM模型可以得到满意的预测结果,预测精度均高于现有的几种AE类模型及其他几种方法,且具有良好的泛化能力和抗过拟合能力。
文摘异常检测是保障飞机运行安全的重要手段,现有的固定阈值异常检测方法对数据时序特征利用较少,提取特征的能力较差。为提高飞机运行安全,提出了一种基于LSTM_AE神经网络的无监督离线异常检测的模型,对实际飞行数据进行异常检测。首先使用LSTM(Long Short Term Memory)网络提取正常飞行数据的深度时序特征,再基于AE(Auto Encoder)对提取到的时序特征进行训练,利用模型收敛后得到重构误差确定自适应阈值,最后根据训练好的模型和自适应阈值进行异常检测。试验利用NASA公开的ALFA数据集。结果表明:基于LSTM_AE方法优于传统的固定阈值检测方法,可以实现对异常的检测,准确率为0.8717,召回率为0.9872,F1分数为0.9258。
文摘空调热交换器性能异常检测技术是快速判断民机空调系统运行状态并合理安排维修任务的关键,传统的异常检测方法难以有效处理高维时序数据,无法实现系统早期故障预警。为此,本文提出了一种基于长短期记忆网络(LSTM,long-short term memory)与自编码器(AE,autoencoder)模型的无监督异常检测方法,用以识别民机空调系统异常运行状态。首先,基于民机空调系统原始传感器参数构建表征空调热交换器性能的特征监测参数;其次,构建LSTM-AE模型进行数据特征重构并计算重构误差;最后,使用孤立森林(iForest, isolation forest)进行无监督异常监测。将本文构建的无监督异常检测方法与传统方法对比,并建立模型评估指标,验证结果表明,所构建的模型方法可以对民机空调热交换器性能异常状态进行有效检测。
基金funded by the SWJTU Science and Technology Innovation Project,Grant Number 2682022CX008the Natural Science Foundation of Sichuan Province,Grant Number 2022NSFSC1892.
文摘Currently,the inexorable trend toward the electrification of automobiles has heightened the prominence of road noise within overall vehicle noise.Consequently,an in-depth investigation into automobile road noise holds substantial practical importance.Previous research endeavors have predominantly centered on the formulation of mechanism models and data-driven models.While mechanism models offer robust controllability,their application encounters challenges in intricate analyses of vehicle body acoustic-vibration coupling,and the effective utilization of accumulated data remains elusive.In contrast,data-driven models exhibit efficient modeling capabilities and can assimilate conceptual vehicle knowledge,but they impose stringent requirements on both data quality and quantity.In response to these considerations,this paper introduces an innovative approach for predicting vehicle road noise by integrating mechanism-driven and data-driven methodologies.Specifically,a series model is devised,amalgamating mechanism analysis with data-driven techniques to predict vehicle interior noise.The simulation results from dynamic models serve as inputs to the data-driven model,ultimately generating outputs through the utilization of the Long Short-Term Memory with Autoencoder(AE-LSTM)architecture.The study subsequently undertakes a comparative analysis between different dynamic models and data-driven models,thereby validating the efficacy of the proposed series vehicle road noise prediction model.This series model,encapsulating the rigid-flexible coupling dynamic model and AE-LSTM series model,not only demonstrates heightened computational efficiency but also attains superior prediction accuracy.
文摘火箭地面测发控系统流程复杂,为自动检测系统运行流程中的异常,本文提出一种基于多变量长短时记忆自编码器(LSTM-AE,Long-short Term Memory-Autoencoder)方法的测发控系统运行异常检测方法。首先对正常系统运行数据进行分析,对系统中开关量和数字量数据进行归一化处理,并系统中模拟量信息进行统计特征提取。然后将提取的模拟量特征和其他数字量信息组成系统运行特征向量,并作为LSTM-AE的样本输入,经自编码学习自动识别系统状态,并建立系统状态迁移异常检测模型。通过对某火箭地面测发控流程监测数据的自动建模,实现了对仿真注入故障的高效实时检测。仿真结果表明,本文所提的自编码方法优于传统的BP神经网络方法。
