There are multiple operating modes in the real industrial process, and the collected data follow the complex multimodal distribution, so most traditional process monitoring methods are no longer applicable because the...There are multiple operating modes in the real industrial process, and the collected data follow the complex multimodal distribution, so most traditional process monitoring methods are no longer applicable because their presumptions are that sampled-data should obey the single Gaussian distribution or non-Gaussian distribution. In order to solve these problems, a novel weighted local standardization(WLS) strategy is proposed to standardize the multimodal data, which can eliminate the multi-mode characteristics of the collected data, and normalize them into unimodal data distribution. After detailed analysis of the raised data preprocessing strategy, a new algorithm using WLS strategy with support vector data description(SVDD) is put forward to apply for multi-mode monitoring process. Unlike the strategy of building multiple local models, the developed method only contains a model without the prior knowledge of multi-mode process. To demonstrate the proposed method's validity, it is applied to a numerical example and a Tennessee Eastman(TE) process. Finally, the simulation results show that the WLS strategy is very effective to standardize multimodal data, and the WLS-SVDD monitoring method has great advantages over the traditional SVDD and PCA combined with a local standardization strategy(LNS-PCA) in multi-mode process monitoring.展开更多
针对齿轮箱振动信号复杂多变,导致现有的齿轮箱故障诊断方法诊断精度不高、较弱故障特征容易被噪声淹没等问题,提出了一种基于向量加权平均优化算法(weighted mean of vectors,INFO)、变分模态分解(variational mode decomposition,VMD...针对齿轮箱振动信号复杂多变,导致现有的齿轮箱故障诊断方法诊断精度不高、较弱故障特征容易被噪声淹没等问题,提出了一种基于向量加权平均优化算法(weighted mean of vectors,INFO)、变分模态分解(variational mode decomposition,VMD)和卷积神经网络(convolutional neural network,CNN)的齿轮故障诊断方法。该方法首先采用熵权法将不同位置的振动传感器信号信息进行融合,利用INFO对VMD算法中参数进行优化,并设计一个复合评价指标作为参数优化的评价标准,使用奇异峭度差分谱的方法对敏感分量进行重构;其次,从重构的信号中提取时域、频域特征并输入到CNN模型中进行分类;最后通过Shap(Shapley additive explanations)值法对模型输入特征的重要性进行排序,分析不同特征组合对模型分类和特定故障识别的影响。在东南大学行星齿轮数据集上进行验证,结果表明,利用所提特征组合进行故障诊断,CNN模型故障诊断准确率为98.24%,高于其他特征组合,为行星齿轮箱的故障诊断提供了一组有效的特征指标。展开更多
针对畜禽疫病文本中特征项权重分配不准导致诊断准确率较低的问题,利用提出的TF-IIGM-NW(Term Frequency-Improved Inverse Gravity Moment With Normalization and Weighting)改进算法结合Word2vec词向量进行文本向量化表示。该方法在T...针对畜禽疫病文本中特征项权重分配不准导致诊断准确率较低的问题,利用提出的TF-IIGM-NW(Term Frequency-Improved Inverse Gravity Moment With Normalization and Weighting)改进算法结合Word2vec词向量进行文本向量化表示。该方法在TF-IIGM(Term Frequency-Improved Inverse Gravity Moment)算法的基础之上,对其进行归一化处理并结合基于关键词抽取算法设定的规则,进一步提升文本内核心关键词权重,然后将其与结合Word2vec词向量获取的文本向量化表示结果输入支持向量机(Support Vector Machine,SVM)进行畜禽疫病诊断。为了验证算法的有效性,基于自建的羊疫病文本数据集,将改进算法与现有词向量常见处理方式进行对比分析。结果表明,基于TF-IIGM-NW算法的macro-F1值与micro-F1值分别达到96.73%,96.76%;与传统经典算法TF-IDF(Term Frequency-Inverse Document Frequency)相比,分别提升2.25%,2.26%;与TF-IIGM算法相比,分别提高0.90%,0.97%。改进算法能够有效提升疫病诊断性能。通过SVM在每类疫病上的实验结果分析表明,羊口疮疫病类别最易被错判。展开更多
针对二次风机状态监测故障的预警问题,提出了一种基于混沌向量加权平均(chaos weighted mean of vectors, CINFO)算法的长短期记忆(long short-term memory, LSTM)网络多输出回归方法。首先,使用Spearman相关系数分析方法筛选出与二次...针对二次风机状态监测故障的预警问题,提出了一种基于混沌向量加权平均(chaos weighted mean of vectors, CINFO)算法的长短期记忆(long short-term memory, LSTM)网络多输出回归方法。首先,使用Spearman相关系数分析方法筛选出与二次风机轴承温度、二次风机轴承振动相关性系数较高的特征参数,对输入数据进行降维。然后,通过CINFO确定多输出LSTM网络的最优超参数,提高了神经网络的预测精度。随后,根据序贯概率比检验(sequential probability ratio test, SPRT)法确定了设备的故障阈值。最后,将选定的特征参数作为CINFO-LSTM网络的输入,使用序贯概率比检验法实现了二次风机的故障预警。实验结果验证了该方法的可行性和有效性。展开更多
电力系统作为实时信息与能源高度融合的电力信息物理融合系统(cyber-physical power system,CPPS),虚假数据注入攻击(false data injection attacks,FDIAs)的准确辨识将有效保证CPPS安全稳定运行。为准确、高效地完成日前负荷预测,首先...电力系统作为实时信息与能源高度融合的电力信息物理融合系统(cyber-physical power system,CPPS),虚假数据注入攻击(false data injection attacks,FDIAs)的准确辨识将有效保证CPPS安全稳定运行。为准确、高效地完成日前负荷预测,首先使用肯德尔相关系数(Kendall's tau-b)量化日期类型的取值,引入加权灰色关联分析选取相似日,再建立基于最小二乘支持向量机(least squares support vector machine,LSSVM)的日前负荷预测模型。将预测负荷通过潮流计算求解的系统节点状态量与无迹卡尔曼滤波(unscented Kalman filter,UKF)动态状态估计得到的状态量进行自适应加权混合,最后基于混合预测值和静态估计值间的偏差变量提出了攻击检测指数(attack detection index,ADI),根据ADI的分布检测FDIAs。若检测到FDIAs,使用混合预测状态量对该时刻的量测量进行修正。使用IEEE-14和IEEE-39节点系统进行仿真,结果验证了所提方法的有效性与可行性。展开更多
针对复杂环境下四旋翼无人机三维航迹规划问题,提出了一种改进的事件触发灰狼优化算法(event triggered grey wolf optimization,ETGWO)。引入球面矢量刻画飞行路径的生成,通过减少搜索空间提升搜索能力;设计自适应权重动态调整飞行航...针对复杂环境下四旋翼无人机三维航迹规划问题,提出了一种改进的事件触发灰狼优化算法(event triggered grey wolf optimization,ETGWO)。引入球面矢量刻画飞行路径的生成,通过减少搜索空间提升搜索能力;设计自适应权重动态调整飞行航迹成本适应度函数,以提高航迹规划效率和准确性;在灰狼优化算法(grey wolf optimization,GWO)基础上,选择使用改进的非线性收敛因子,提升算法的鲁棒性;为了更好地平衡算法的全局搜索和局部搜索能力,通过引入基于事件触发机制的灰狼个体位置更新速度来改进GWO算法的位置更新策略。仿真对比实验表明,所提出ETGWO算法在四旋翼无人机(quadrotor unmanned aerial vehicles,QUAV)飞行航迹规划方面具有更优越的性能。展开更多
A new threshold secret sharing scheme is constructed by introducing the concept of share vector, in which the number of shareholders can be adjusted by randomly changing the weights of them. This kind of scheme overco...A new threshold secret sharing scheme is constructed by introducing the concept of share vector, in which the number of shareholders can be adjusted by randomly changing the weights of them. This kind of scheme overcomes the limitation of the static weighted secret sharing schemes that cannot change the weights in the process of carrying out and the deficiency of low efficiency of the ordinary dynamic weighted sharing schemes for its resending process. Thus, this scheme is more suitable to the case that the number of shareholders needs to be changed randomly during the scheme is carrying out.展开更多
基金Project(61374140)supported by the National Natural Science Foundation of China
文摘There are multiple operating modes in the real industrial process, and the collected data follow the complex multimodal distribution, so most traditional process monitoring methods are no longer applicable because their presumptions are that sampled-data should obey the single Gaussian distribution or non-Gaussian distribution. In order to solve these problems, a novel weighted local standardization(WLS) strategy is proposed to standardize the multimodal data, which can eliminate the multi-mode characteristics of the collected data, and normalize them into unimodal data distribution. After detailed analysis of the raised data preprocessing strategy, a new algorithm using WLS strategy with support vector data description(SVDD) is put forward to apply for multi-mode monitoring process. Unlike the strategy of building multiple local models, the developed method only contains a model without the prior knowledge of multi-mode process. To demonstrate the proposed method's validity, it is applied to a numerical example and a Tennessee Eastman(TE) process. Finally, the simulation results show that the WLS strategy is very effective to standardize multimodal data, and the WLS-SVDD monitoring method has great advantages over the traditional SVDD and PCA combined with a local standardization strategy(LNS-PCA) in multi-mode process monitoring.
文摘针对齿轮箱振动信号复杂多变,导致现有的齿轮箱故障诊断方法诊断精度不高、较弱故障特征容易被噪声淹没等问题,提出了一种基于向量加权平均优化算法(weighted mean of vectors,INFO)、变分模态分解(variational mode decomposition,VMD)和卷积神经网络(convolutional neural network,CNN)的齿轮故障诊断方法。该方法首先采用熵权法将不同位置的振动传感器信号信息进行融合,利用INFO对VMD算法中参数进行优化,并设计一个复合评价指标作为参数优化的评价标准,使用奇异峭度差分谱的方法对敏感分量进行重构;其次,从重构的信号中提取时域、频域特征并输入到CNN模型中进行分类;最后通过Shap(Shapley additive explanations)值法对模型输入特征的重要性进行排序,分析不同特征组合对模型分类和特定故障识别的影响。在东南大学行星齿轮数据集上进行验证,结果表明,利用所提特征组合进行故障诊断,CNN模型故障诊断准确率为98.24%,高于其他特征组合,为行星齿轮箱的故障诊断提供了一组有效的特征指标。
文摘针对二次风机状态监测故障的预警问题,提出了一种基于混沌向量加权平均(chaos weighted mean of vectors, CINFO)算法的长短期记忆(long short-term memory, LSTM)网络多输出回归方法。首先,使用Spearman相关系数分析方法筛选出与二次风机轴承温度、二次风机轴承振动相关性系数较高的特征参数,对输入数据进行降维。然后,通过CINFO确定多输出LSTM网络的最优超参数,提高了神经网络的预测精度。随后,根据序贯概率比检验(sequential probability ratio test, SPRT)法确定了设备的故障阈值。最后,将选定的特征参数作为CINFO-LSTM网络的输入,使用序贯概率比检验法实现了二次风机的故障预警。实验结果验证了该方法的可行性和有效性。
文摘电力系统作为实时信息与能源高度融合的电力信息物理融合系统(cyber-physical power system,CPPS),虚假数据注入攻击(false data injection attacks,FDIAs)的准确辨识将有效保证CPPS安全稳定运行。为准确、高效地完成日前负荷预测,首先使用肯德尔相关系数(Kendall's tau-b)量化日期类型的取值,引入加权灰色关联分析选取相似日,再建立基于最小二乘支持向量机(least squares support vector machine,LSSVM)的日前负荷预测模型。将预测负荷通过潮流计算求解的系统节点状态量与无迹卡尔曼滤波(unscented Kalman filter,UKF)动态状态估计得到的状态量进行自适应加权混合,最后基于混合预测值和静态估计值间的偏差变量提出了攻击检测指数(attack detection index,ADI),根据ADI的分布检测FDIAs。若检测到FDIAs,使用混合预测状态量对该时刻的量测量进行修正。使用IEEE-14和IEEE-39节点系统进行仿真,结果验证了所提方法的有效性与可行性。
文摘针对复杂环境下四旋翼无人机三维航迹规划问题,提出了一种改进的事件触发灰狼优化算法(event triggered grey wolf optimization,ETGWO)。引入球面矢量刻画飞行路径的生成,通过减少搜索空间提升搜索能力;设计自适应权重动态调整飞行航迹成本适应度函数,以提高航迹规划效率和准确性;在灰狼优化算法(grey wolf optimization,GWO)基础上,选择使用改进的非线性收敛因子,提升算法的鲁棒性;为了更好地平衡算法的全局搜索和局部搜索能力,通过引入基于事件触发机制的灰狼个体位置更新速度来改进GWO算法的位置更新策略。仿真对比实验表明,所提出ETGWO算法在四旋翼无人机(quadrotor unmanned aerial vehicles,QUAV)飞行航迹规划方面具有更优越的性能。
基金supported by the National Preeminent Youth Foundation(70225002)the Doctor Foundation of North China Electric Power University(200822029).
文摘A new threshold secret sharing scheme is constructed by introducing the concept of share vector, in which the number of shareholders can be adjusted by randomly changing the weights of them. This kind of scheme overcomes the limitation of the static weighted secret sharing schemes that cannot change the weights in the process of carrying out and the deficiency of low efficiency of the ordinary dynamic weighted sharing schemes for its resending process. Thus, this scheme is more suitable to the case that the number of shareholders needs to be changed randomly during the scheme is carrying out.
