When a high impedance fault(HIF)occurs in a distribution network,the detection efficiency of traditional protection devices is strongly limited by the weak fault information.In this study,a method based on S-transform...When a high impedance fault(HIF)occurs in a distribution network,the detection efficiency of traditional protection devices is strongly limited by the weak fault information.In this study,a method based on S-transform(ST)and average singular entropy(ASE)is proposed to identify HIFs.First,a wavelet packet transform(WPT)was applied to extract the feature frequency band.Thereafter,the ST was investigated in each half cycle.Afterwards,the obtained time-frequency matrix was denoised by singular value decomposition(SVD),followed by the calculation of the ASE index.Finally,an appropriate threshold was selected to detect the HIFs.The advantages of this method are the ability of fine band division,adaptive time-frequency transformation,and quantitative expression of signal complexity.The performance of the proposed method was verified by simulated and field data,and further analysis revealed that it could still achieve good results under different conditions.展开更多
The complexities of multi-wing chaotic systems based on the modified Chen system and a multi-segment quadratic function are investigated by employing the statistical complexity measure （SCM） and the spectral entropy...The complexities of multi-wing chaotic systems based on the modified Chen system and a multi-segment quadratic function are investigated by employing the statistical complexity measure （SCM） and the spectral entropy （SE） algorithm. How to choose the parameters of the SCM and SE algorithms is discussed. The results show that the complexity of the multi-wing chaotic system does not increase as the number of wings increases, and it is consistent with the results of the Grassberger-Procaccia （GP） algorithm and the largest Lyapunov exponent （LLE） of the multi-wing chaotic system.展开更多
风电机组的环境恶劣和工况多变导致风电机组故障频发,为了保障风电机组的可靠运行,基于数据的机组异常状态检测尤为重要。该研究提出一种基于级联深度学习预测模型的风电机组状态检测方法,首先对风电场数据采集与监视控制(supervisory c...风电机组的环境恶劣和工况多变导致风电机组故障频发,为了保障风电机组的可靠运行,基于数据的机组异常状态检测尤为重要。该研究提出一种基于级联深度学习预测模型的风电机组状态检测方法,首先对风电场数据采集与监视控制(supervisory control and data acquisition,SCADA)系统的数据进行预处理,并通过距离相关系数(distance correlation coefficient,DCC)分析选取输入参数;然后结合卷积神经网络(convolution neural network,CNN)和长短期神经网络(long short-term memory,LSTM)建立观测参数与目标参数之间的逻辑关系,通过均方根误差(root mean square error,RMSE)和样本熵(sample entropy,SE)对齿轮箱轴承温度预测残差进行分析,监测齿轮箱轴承温度异常变化;最后以华北某风场的SCADA数据进行算例验证,结果表明该方法能够准确检测到齿轮箱轴承温度异常,提前发现风电机组的早期故障,为风电机组安全可靠运行提供重要价值。展开更多
坝肩边坡变形在外部因素影响下呈现出不确定性和随机性,从而不易预测。基于聚类模态分解(EEMD)、样本熵(SE)和改进型粒子群算法优化的最小二乘支持向量机(IPSO LSSVM)方法,提出一种名为EEMD SE IPSO LSSVM的坝肩边坡变形预测模型。首先...坝肩边坡变形在外部因素影响下呈现出不确定性和随机性,从而不易预测。基于聚类模态分解(EEMD)、样本熵(SE)和改进型粒子群算法优化的最小二乘支持向量机(IPSO LSSVM)方法,提出一种名为EEMD SE IPSO LSSVM的坝肩边坡变形预测模型。首先,利用EEMD将原始坝肩边坡变形时间序列分解为若干个不同复杂度的子序列,并基于SE判定各子序列的复杂度,将相近的子序列进行合并重组以减少计算规模;然后,分别对各重组子序列建立IPSO LSSVM预测模型;最后,将各预测分量进行叠加重构,得到最终的大坝变形预测值。以澜沧江苗尾水电站左岸坝肩边坡为例,将BPNN、RBFNN、LSSVM、EEMD SE LSSVM与EEMD SE PSO LSSVM进行对比研究。结果表明,该模型的计算精度优于其他神经网络模型,具有较好的适宜性和稳定性,是一种可靠的坝肩边坡变形预测方法,能够为大坝安全监测提供有价值的参考。展开更多
基金financial supported by the Natural Science Foundation of Fujian,China(2021J01633).
文摘When a high impedance fault(HIF)occurs in a distribution network,the detection efficiency of traditional protection devices is strongly limited by the weak fault information.In this study,a method based on S-transform(ST)and average singular entropy(ASE)is proposed to identify HIFs.First,a wavelet packet transform(WPT)was applied to extract the feature frequency band.Thereafter,the ST was investigated in each half cycle.Afterwards,the obtained time-frequency matrix was denoised by singular value decomposition(SVD),followed by the calculation of the ASE index.Finally,an appropriate threshold was selected to detect the HIFs.The advantages of this method are the ability of fine band division,adaptive time-frequency transformation,and quantitative expression of signal complexity.The performance of the proposed method was verified by simulated and field data,and further analysis revealed that it could still achieve good results under different conditions.
基金Project supported by the National Natural Science Foundation of China (Grant Nos. 61161006 and 61073187)
文摘The complexities of multi-wing chaotic systems based on the modified Chen system and a multi-segment quadratic function are investigated by employing the statistical complexity measure （SCM） and the spectral entropy （SE） algorithm. How to choose the parameters of the SCM and SE algorithms is discussed. The results show that the complexity of the multi-wing chaotic system does not increase as the number of wings increases, and it is consistent with the results of the Grassberger-Procaccia （GP） algorithm and the largest Lyapunov exponent （LLE） of the multi-wing chaotic system.
文摘风电机组的环境恶劣和工况多变导致风电机组故障频发,为了保障风电机组的可靠运行,基于数据的机组异常状态检测尤为重要。该研究提出一种基于级联深度学习预测模型的风电机组状态检测方法,首先对风电场数据采集与监视控制(supervisory control and data acquisition,SCADA)系统的数据进行预处理,并通过距离相关系数(distance correlation coefficient,DCC)分析选取输入参数;然后结合卷积神经网络(convolution neural network,CNN)和长短期神经网络(long short-term memory,LSTM)建立观测参数与目标参数之间的逻辑关系,通过均方根误差(root mean square error,RMSE)和样本熵(sample entropy,SE)对齿轮箱轴承温度预测残差进行分析,监测齿轮箱轴承温度异常变化;最后以华北某风场的SCADA数据进行算例验证,结果表明该方法能够准确检测到齿轮箱轴承温度异常,提前发现风电机组的早期故障,为风电机组安全可靠运行提供重要价值。
文摘坝肩边坡变形在外部因素影响下呈现出不确定性和随机性,从而不易预测。基于聚类模态分解(EEMD)、样本熵(SE)和改进型粒子群算法优化的最小二乘支持向量机(IPSO LSSVM)方法,提出一种名为EEMD SE IPSO LSSVM的坝肩边坡变形预测模型。首先,利用EEMD将原始坝肩边坡变形时间序列分解为若干个不同复杂度的子序列,并基于SE判定各子序列的复杂度,将相近的子序列进行合并重组以减少计算规模;然后,分别对各重组子序列建立IPSO LSSVM预测模型;最后,将各预测分量进行叠加重构,得到最终的大坝变形预测值。以澜沧江苗尾水电站左岸坝肩边坡为例,将BPNN、RBFNN、LSSVM、EEMD SE LSSVM与EEMD SE PSO LSSVM进行对比研究。结果表明,该模型的计算精度优于其他神经网络模型,具有较好的适宜性和稳定性,是一种可靠的坝肩边坡变形预测方法,能够为大坝安全监测提供有价值的参考。
