Real-time voltage stability assessment(VSA)has long been an extensively research topic.In recent years,rapidly mounting deep learning methods have pushed online VSA to a new height that large amounts of learning algor...Real-time voltage stability assessment(VSA)has long been an extensively research topic.In recent years,rapidly mounting deep learning methods have pushed online VSA to a new height that large amounts of learning algorithms are applied for VSA from the perspective of measurement data.Deep learning methods generally require a large dataset which contains measurements in both secure and insecure states,or even unstable state.However,in practice,the data of insecure or unstable state is very rare,as the power system should be guaranteed to operate far away from voltage collapse.Under this circumstance,this paper proposes an autoencoder based method which merely needs data of secure state to evaluate voltage stability of a power system.The principle of this method is that an autoencoder purely trained by secure data is expected to only create precise reconstruction for secure data,while it fails to rebuild data of insecure states.Thus,the residual of reconstruction is effective in indicating VSA.Besides,to develop a more accurate and robust algorithm,long short-term memory(LSTM)networks combined with fully-connected(FC)layers are used to build the autoencoder,and a moving strategy is introduced to bias the features of testing data toward the secure feature domain.Numerous experiments and comparison with traditional machine learning algorithms demonstrate the effectiveness and high accuracy of the proposed method.展开更多
针对影响风电中长期预测的气象、地理等因素众多且复杂,及无法解决长期依赖时间序列的问题,提出一种基于多维特征融合网络(multi-dimensional feature fusion network,MFFN)和长短期记忆(long and short term memory,LSTM)的预测方法—...针对影响风电中长期预测的气象、地理等因素众多且复杂,及无法解决长期依赖时间序列的问题,提出一种基于多维特征融合网络(multi-dimensional feature fusion network,MFFN)和长短期记忆(long and short term memory,LSTM)的预测方法—多维特征提取(feature extraction,FE)-关联函数(copula,CO)-LSTM融合模型(FE-CO-LSTM)。收集来自不同地区4个风电场的特征数据,在研究云贵高原地区风电场的背景下,最大限度扩充模型数据集;使用关联结构函数构造一种提取气象特征的方法,使模型可以在一定程度上量化气象因素和风力发电之间的相关性;基于神经网络模型提出一种特征表示与融合方法,以有效表达风电场气象因素、地理位置等特征;最后提出一种基于LSTM网络的中长期发电量预测模型,以有效解决模型对风电场时间序列数据反向传播时早期月度数据信息缺失的问题。实验结果证明,FE-CO-LSTM表现出最佳的预测性能。展开更多
针对SAR图像检测船舶任务中的目标小、近岸样本目标检测困难等问题,文章提出一种名为长短路特征融合网络(Long and Short path Feature Fusion Network,LSFF-Net)的船舶检测网络。该网络通过长短路特征融合模块有效协调了大目标与小目...针对SAR图像检测船舶任务中的目标小、近岸样本目标检测困难等问题,文章提出一种名为长短路特征融合网络(Long and Short path Feature Fusion Network,LSFF-Net)的船舶检测网络。该网络通过长短路特征融合模块有效协调了大目标与小目标检测,避免小目标特征信息的丢失。网络中应用结构重参数化结构提高了模块学习能力。为了满足多尺度目标检测,加入特征金字塔网络,融合多尺度特征。为了应对近岸样本目标检测,设计数据重分配算法,提高了对近岸样本目标的检测精度。实验结果表明:在公开数据集检测时,算法的平均精度(Average Precision,AP)达到97.50%,优于主流目标检测算法。该方法为提高SAR图像中小目标和近岸样本目标检测精度提供了新的实现方案。展开更多
为了克服卷积神经网络(Convolutional Neural Network,CNN)轴承故障诊断方法特征提取过程困难以及难以捕获时间序列数据之间的长期依赖关系的问题,提出一种改进的卷积-长短时记忆网络(Convolutional Neural Network-Long and Short Term...为了克服卷积神经网络(Convolutional Neural Network,CNN)轴承故障诊断方法特征提取过程困难以及难以捕获时间序列数据之间的长期依赖关系的问题,提出一种改进的卷积-长短时记忆网络(Convolutional Neural Network-Long and Short Term Memory,CNN-LSTM)滚动轴承故障诊断方法。将二维轴承加速度振动信号输入CNN提取局部特征,再将轴承特征信息加载到LSTM长期记忆单元中,引入遗忘机制提取时序数据的全局特征。利用轴承振动信号的局部深层特征和全局时序特征,学习不同区间长度的序列特征,从而提高故障诊断精度。实验结果表明,该方法可用于轴承故障诊断,且具有较高的分类精度和较强的稳定性。展开更多
文摘Real-time voltage stability assessment(VSA)has long been an extensively research topic.In recent years,rapidly mounting deep learning methods have pushed online VSA to a new height that large amounts of learning algorithms are applied for VSA from the perspective of measurement data.Deep learning methods generally require a large dataset which contains measurements in both secure and insecure states,or even unstable state.However,in practice,the data of insecure or unstable state is very rare,as the power system should be guaranteed to operate far away from voltage collapse.Under this circumstance,this paper proposes an autoencoder based method which merely needs data of secure state to evaluate voltage stability of a power system.The principle of this method is that an autoencoder purely trained by secure data is expected to only create precise reconstruction for secure data,while it fails to rebuild data of insecure states.Thus,the residual of reconstruction is effective in indicating VSA.Besides,to develop a more accurate and robust algorithm,long short-term memory(LSTM)networks combined with fully-connected(FC)layers are used to build the autoencoder,and a moving strategy is introduced to bias the features of testing data toward the secure feature domain.Numerous experiments and comparison with traditional machine learning algorithms demonstrate the effectiveness and high accuracy of the proposed method.
文摘针对影响风电中长期预测的气象、地理等因素众多且复杂,及无法解决长期依赖时间序列的问题,提出一种基于多维特征融合网络(multi-dimensional feature fusion network,MFFN)和长短期记忆(long and short term memory,LSTM)的预测方法—多维特征提取(feature extraction,FE)-关联函数(copula,CO)-LSTM融合模型(FE-CO-LSTM)。收集来自不同地区4个风电场的特征数据,在研究云贵高原地区风电场的背景下,最大限度扩充模型数据集;使用关联结构函数构造一种提取气象特征的方法,使模型可以在一定程度上量化气象因素和风力发电之间的相关性;基于神经网络模型提出一种特征表示与融合方法,以有效表达风电场气象因素、地理位置等特征;最后提出一种基于LSTM网络的中长期发电量预测模型,以有效解决模型对风电场时间序列数据反向传播时早期月度数据信息缺失的问题。实验结果证明,FE-CO-LSTM表现出最佳的预测性能。
文摘针对SAR图像检测船舶任务中的目标小、近岸样本目标检测困难等问题,文章提出一种名为长短路特征融合网络(Long and Short path Feature Fusion Network,LSFF-Net)的船舶检测网络。该网络通过长短路特征融合模块有效协调了大目标与小目标检测,避免小目标特征信息的丢失。网络中应用结构重参数化结构提高了模块学习能力。为了满足多尺度目标检测,加入特征金字塔网络,融合多尺度特征。为了应对近岸样本目标检测,设计数据重分配算法,提高了对近岸样本目标的检测精度。实验结果表明:在公开数据集检测时,算法的平均精度(Average Precision,AP)达到97.50%,优于主流目标检测算法。该方法为提高SAR图像中小目标和近岸样本目标检测精度提供了新的实现方案。
文摘为了克服卷积神经网络(Convolutional Neural Network,CNN)轴承故障诊断方法特征提取过程困难以及难以捕获时间序列数据之间的长期依赖关系的问题,提出一种改进的卷积-长短时记忆网络(Convolutional Neural Network-Long and Short Term Memory,CNN-LSTM)滚动轴承故障诊断方法。将二维轴承加速度振动信号输入CNN提取局部特征,再将轴承特征信息加载到LSTM长期记忆单元中,引入遗忘机制提取时序数据的全局特征。利用轴承振动信号的局部深层特征和全局时序特征,学习不同区间长度的序列特征,从而提高故障诊断精度。实验结果表明,该方法可用于轴承故障诊断,且具有较高的分类精度和较强的稳定性。
