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
分布式显示控制系统(以下简称分布式)被广泛应用于指挥调度、视频会议、信息发布等领域,是高清视频信号传输和调度的核心,也担负着大量音频信号的传输和调度任务。分布式在与音频扩声系统之间相互传输音频信号时,通常需要通过非平衡模...分布式显示控制系统(以下简称分布式)被广泛应用于指挥调度、视频会议、信息发布等领域,是高清视频信号传输和调度的核心,也担负着大量音频信号的传输和调度任务。分布式在与音频扩声系统之间相互传输音频信号时,通常需要通过非平衡模拟信号线缆实现,存在多次数模(Digital to Analog,D/A)和模数(Analog to Digital,A/D)转换,造成音频信号质量显著下降和精度损失。针对该问题,提出将AES67音频网关作为分布式和音频扩声系统之间的桥梁,在2个系统之间通过AES67实现网际互连协议(Internet Protocol,IP)全数字方式转换与传输,保证高质量的音频传输和性能。展开更多
Acoustic emission(AE)localization algorithms based on homogeneous media or single-velocity are less accurate when applied to the triaxial localization experiments.To the end,a robust triaxial localization method of AE...Acoustic emission(AE)localization algorithms based on homogeneous media or single-velocity are less accurate when applied to the triaxial localization experiments.To the end,a robust triaxial localization method of AE source using refraction path is proposed.Firstly,the control equation of the refraction path is established according to the sensor coordinates and arrival times.Secondly,considering the influence of time-difference-of-arrival(TDOA)errors,the residual of the governing equation is calculated to estimate the equation weight.Thirdly,the refraction points in different directions are solved using Snell’s law and orthogonal constraints.Finally,the source coordinates are iteratively solved by weighted correction terms.The feasibility and accuracy of the proposed method are verified by pencil-lead breaking experiments.The simulation results show that the new method is almost unaffected by the refraction ratio,and always holds more stable and accurate positioning performance than the traditional method under different ratios and scales of TDOA outliers.展开更多
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。
文摘分布式显示控制系统(以下简称分布式)被广泛应用于指挥调度、视频会议、信息发布等领域,是高清视频信号传输和调度的核心,也担负着大量音频信号的传输和调度任务。分布式在与音频扩声系统之间相互传输音频信号时,通常需要通过非平衡模拟信号线缆实现,存在多次数模(Digital to Analog,D/A)和模数(Analog to Digital,A/D)转换,造成音频信号质量显著下降和精度损失。针对该问题,提出将AES67音频网关作为分布式和音频扩声系统之间的桥梁,在2个系统之间通过AES67实现网际互连协议(Internet Protocol,IP)全数字方式转换与传输,保证高质量的音频传输和性能。
基金the National Natural Science Foundation of China (Nos.52304123 and 52104077)the Postdoctoral Fellowship Program of CPSF (No.GZB20230914)+1 种基金the China Postdoctoral Science Foundation (No.2023M730412)the National Key Research and Development Program for Young Scientists (No.2021YFC2900400)。
文摘Acoustic emission(AE)localization algorithms based on homogeneous media or single-velocity are less accurate when applied to the triaxial localization experiments.To the end,a robust triaxial localization method of AE source using refraction path is proposed.Firstly,the control equation of the refraction path is established according to the sensor coordinates and arrival times.Secondly,considering the influence of time-difference-of-arrival(TDOA)errors,the residual of the governing equation is calculated to estimate the equation weight.Thirdly,the refraction points in different directions are solved using Snell’s law and orthogonal constraints.Finally,the source coordinates are iteratively solved by weighted correction terms.The feasibility and accuracy of the proposed method are verified by pencil-lead breaking experiments.The simulation results show that the new method is almost unaffected by the refraction ratio,and always holds more stable and accurate positioning performance than the traditional method under different ratios and scales of TDOA outliers.
