Feature extraction of range images provided by ranging sensor is a key issue of pattern recognition. To automatically extract the environmental feature sensed by a 2D ranging sensor laser scanner, an improved method b...Feature extraction of range images provided by ranging sensor is a key issue of pattern recognition. To automatically extract the environmental feature sensed by a 2D ranging sensor laser scanner, an improved method based on genetic clustering VGA-clustering is presented. By integrating the spatial neighbouring information of range data into fuzzy clustering algorithm, a weighted fuzzy clustering algorithm （WFCA） instead of standard clustering algorithm is introduced to realize feature extraction of laser scanner. Aimed at the unknown clustering number in advance, several validation index functions are used to estimate the validity of different clustering algorithms and one validation index is selected as the fitness function of genetic algorithm so as to determine the accurate clustering number automatically. At the same time, an improved genetic algorithm IVGA on the basis of VGA is proposed to solve the local optimum of clustering algorithm, which is implemented by increasing the population diversity and improving the genetic operators of elitist rule to enhance the local search capacity and to quicken the convergence speed. By the comparison with other algorithms, the effectiveness of the algorithm introduced is demonstrated.展开更多
To improve the recognition rate of signal modulation recognition methods based on the clustering algorithm under the low SNR, a modulation recognition method is proposed. The characteristic parameter of the signal is ...To improve the recognition rate of signal modulation recognition methods based on the clustering algorithm under the low SNR, a modulation recognition method is proposed. The characteristic parameter of the signal is extracted by using a clustering algorithm, the neural network is trained by using the algorithm of variable gradient correction （Polak-Ribiere） so as to enhance the rate of convergence, improve the performance of recognition under the low SNR and realize modulation recognition of the signal based on the modulation system of the constellation diagram. Simulation results show that the recognition rate based on this algorithm is enhanced over 30% compared with the methods that adopt clustering algorithm or neural network based on the back propagation algorithm alone under the low SNR. The recognition rate can reach 90% when the SNR is 4 dB, and the method is easy to be achieved so that it has a broad application prospect in the modulating recognition.展开更多
Constrained by complex imaging mechanism and extraordinary visual appearance,change detection with synthetic aperture radar(SAR)images has been a difficult research topic,especially in urban areas.Although existing st...Constrained by complex imaging mechanism and extraordinary visual appearance,change detection with synthetic aperture radar(SAR)images has been a difficult research topic,especially in urban areas.Although existing studies have extended from bi-temporal data pair to multi-temporal datasets to derive more plentiful information,there are still two problems to be solved in practical applications.First,change indicators constructed from incoherent feature only cannot characterize the change objects accurately.Second,the results of pixel-level methods are usually presented in the form of the noisy binary map,making the spatial change not intuitive and the temporal change of a single pixel meaningless.In this study,we propose an unsupervised man-made objects change detection framework using both coherent and incoherent features derived from multi-temporal SAR images.The coefficients of variation in timeseries incoherent features and the man-made object index(MOI)defined with coherent features are first combined to identify the initial change pixels.Afterwards,an improved spatiotemporal clustering algorithm is developed based on density-based spatial clustering of applications with noise(DBSCAN)and dynamic time warping(DTW),which can transform the initial results into noiseless object-level patches,and take the cluster center as a representative of the man-made object to determine the change pattern of each patch.An experiment with a stack of 10 TerraSAR-X images in Stripmap mode demonstrated that this method is effective in urban scenes and has the potential applicability to wide area change detection.展开更多
针对目前海上风电出力预测方法精度较低的问题,提出一种基于词向量化和长短期记忆网络(word to vector long short-term memory,Word2vec-LSTM)与聚类修正的海上风电出力预测方法。对Word2vec方法进行改进来提取时间序列数据特征,实现...针对目前海上风电出力预测方法精度较低的问题,提出一种基于词向量化和长短期记忆网络(word to vector long short-term memory,Word2vec-LSTM)与聚类修正的海上风电出力预测方法。对Word2vec方法进行改进来提取时间序列数据特征,实现数据信息的高效利用;在长短期记忆神经网络的预测模型基础上,研究了一种基于k-shape聚类结果的预测结果修正算法,对预测结果距离聚类中心超过阈值的数值判定为预测误差偏大的数据并向簇中心进行修正。最后,基于江苏某海上风电场的真实数据进行测试,结果表明,基于Word2vec-LSTM与聚类修正的海上风电出力预测方法的平均绝对误差(mean absolute error,MAE)和均方根误差(root mean square error,RMSE)达到5.04和5.42,相比传统LSTM预测模型的误差平均降低了11.10%和12.25%,为海上风电并网与电网调控提供了技术支持。展开更多
随着向新型能源体系的转型加速,亟待开展对多元负荷用户的复杂用能特性分析的深入研究。提出了一种综合考量电、冷、热多元负荷耦合特性的用户用能特性标签库构建技术及用户画像方法。首先运用快速相关性滤波算法剔除高冗余低相关特征,...随着向新型能源体系的转型加速,亟待开展对多元负荷用户的复杂用能特性分析的深入研究。提出了一种综合考量电、冷、热多元负荷耦合特性的用户用能特性标签库构建技术及用户画像方法。首先运用快速相关性滤波算法剔除高冗余低相关特征,并通过随机森林和递归式特征消除算法精选出具有强区分能力的用能特征。在聚类阶段,改进的自适应三支密度峰值聚类算法(three-way adaptive density peak clustering,3W-ADPC)通过结合自适应近邻搜索和三支聚类算法提升负荷聚类效果。实证结果表明,所提方法具备在计算效率和聚类精度上的双重优势,能够精准揭示多元负荷用户综合用能特性和深层次信息,证实所提方法在多元负荷用户行为研究中的实用价值。展开更多
为了提高电力负荷监控的准确性,研究融合主成分含噪密度聚类(density-based spatial clustering of applications with noise with principal component analysis,PCADBSCAN)的混合非侵入式负荷辨识方法。首先,针对原始负荷特征维度较...为了提高电力负荷监控的准确性,研究融合主成分含噪密度聚类(density-based spatial clustering of applications with noise with principal component analysis,PCADBSCAN)的混合非侵入式负荷辨识方法。首先,针对原始负荷特征维度较高的问题,采用主成分分析算法对原始特征数据降维,构建负荷特征模板库,同时,获取负荷电流波形,构建负荷电流模板库。其次,采用基于密度的聚类算法对负荷特征模板库内的样本进行非监督聚类,提取各聚类簇中心。然后,计算待辨识负荷与各特征模板库聚类中心的欧式距离,完成负荷特征匹配,并计算待辨识负荷的电流波形与电流模板库内各电流波形的综合关联度,完成负荷电流波形匹配。最后,混合两次匹配结果,综合判断待辨识负荷,从而实现高可靠辨识。基于用电数据测试数据集的仿真结果显示,该方法各项指标均超过96%。展开更多
基金the National Natural Science Foundation of China (60234030)the Natural Science Foundationof He’nan Educational Committee of China (2007520019, 2008B520015)Doctoral Foundation of Henan Polytechnic Universityof China (B050901, B2008-61)
文摘Feature extraction of range images provided by ranging sensor is a key issue of pattern recognition. To automatically extract the environmental feature sensed by a 2D ranging sensor laser scanner, an improved method based on genetic clustering VGA-clustering is presented. By integrating the spatial neighbouring information of range data into fuzzy clustering algorithm, a weighted fuzzy clustering algorithm （WFCA） instead of standard clustering algorithm is introduced to realize feature extraction of laser scanner. Aimed at the unknown clustering number in advance, several validation index functions are used to estimate the validity of different clustering algorithms and one validation index is selected as the fitness function of genetic algorithm so as to determine the accurate clustering number automatically. At the same time, an improved genetic algorithm IVGA on the basis of VGA is proposed to solve the local optimum of clustering algorithm, which is implemented by increasing the population diversity and improving the genetic operators of elitist rule to enhance the local search capacity and to quicken the convergence speed. By the comparison with other algorithms, the effectiveness of the algorithm introduced is demonstrated.
基金supported by the National Natural Science Foundation of China(6107207061301179)the National Science and Technology Major Project(2010ZX03006-002-04)
文摘To improve the recognition rate of signal modulation recognition methods based on the clustering algorithm under the low SNR, a modulation recognition method is proposed. The characteristic parameter of the signal is extracted by using a clustering algorithm, the neural network is trained by using the algorithm of variable gradient correction （Polak-Ribiere） so as to enhance the rate of convergence, improve the performance of recognition under the low SNR and realize modulation recognition of the signal based on the modulation system of the constellation diagram. Simulation results show that the recognition rate based on this algorithm is enhanced over 30% compared with the methods that adopt clustering algorithm or neural network based on the back propagation algorithm alone under the low SNR. The recognition rate can reach 90% when the SNR is 4 dB, and the method is easy to be achieved so that it has a broad application prospect in the modulating recognition.
基金supported by the National Natural Science Foundation of China(41774006)the Comparative Study of Geo-environment and Geohazards in the Yangtze River Delta and the Red River Delta Projectthe Shanghai Science and Technology Development Foundation(20dz1201200)。
文摘Constrained by complex imaging mechanism and extraordinary visual appearance,change detection with synthetic aperture radar(SAR)images has been a difficult research topic,especially in urban areas.Although existing studies have extended from bi-temporal data pair to multi-temporal datasets to derive more plentiful information,there are still two problems to be solved in practical applications.First,change indicators constructed from incoherent feature only cannot characterize the change objects accurately.Second,the results of pixel-level methods are usually presented in the form of the noisy binary map,making the spatial change not intuitive and the temporal change of a single pixel meaningless.In this study,we propose an unsupervised man-made objects change detection framework using both coherent and incoherent features derived from multi-temporal SAR images.The coefficients of variation in timeseries incoherent features and the man-made object index(MOI)defined with coherent features are first combined to identify the initial change pixels.Afterwards,an improved spatiotemporal clustering algorithm is developed based on density-based spatial clustering of applications with noise(DBSCAN)and dynamic time warping(DTW),which can transform the initial results into noiseless object-level patches,and take the cluster center as a representative of the man-made object to determine the change pattern of each patch.An experiment with a stack of 10 TerraSAR-X images in Stripmap mode demonstrated that this method is effective in urban scenes and has the potential applicability to wide area change detection.
文摘针对目前海上风电出力预测方法精度较低的问题,提出一种基于词向量化和长短期记忆网络(word to vector long short-term memory,Word2vec-LSTM)与聚类修正的海上风电出力预测方法。对Word2vec方法进行改进来提取时间序列数据特征,实现数据信息的高效利用;在长短期记忆神经网络的预测模型基础上,研究了一种基于k-shape聚类结果的预测结果修正算法,对预测结果距离聚类中心超过阈值的数值判定为预测误差偏大的数据并向簇中心进行修正。最后,基于江苏某海上风电场的真实数据进行测试,结果表明,基于Word2vec-LSTM与聚类修正的海上风电出力预测方法的平均绝对误差(mean absolute error,MAE)和均方根误差(root mean square error,RMSE)达到5.04和5.42,相比传统LSTM预测模型的误差平均降低了11.10%和12.25%,为海上风电并网与电网调控提供了技术支持。
文摘在自组织映射(Self-organizing Map,SOM)模型的训练过程中,不同类数据对权重矩阵的更新有不同作用,某一类数据对权重矩阵的更新会对其他类获胜神经元特征向量产生偏离其数据特征的影响,从而降低算法聚类精度。针对以上问题,提出一种改进的基于置信度SOM模型(Improved Confidence-based SOM Model,icSOM)。样本数据首先由K-means算法初步分类,为模型训练提供更多的数据信息;然后将预分类后的数据分别训练相互独立的SOM模型,以消除不同类之间的影响;最后在传统SOM模型基础上提出置信度矩阵概念,通过综合判断获胜神经元的置信度及其与输入数据间的欧氏距离最终得到置信神经元,根据置信神经元所属类别给数据分配聚类标签。在鸢尾花数据集(Iris)及葡萄酒数据集(Wine)上利用icSOM进行聚类分析,实验结果表明,所提算法可以更好地处理样本数据,取得了较好的聚类效果。
文摘随着向新型能源体系的转型加速,亟待开展对多元负荷用户的复杂用能特性分析的深入研究。提出了一种综合考量电、冷、热多元负荷耦合特性的用户用能特性标签库构建技术及用户画像方法。首先运用快速相关性滤波算法剔除高冗余低相关特征,并通过随机森林和递归式特征消除算法精选出具有强区分能力的用能特征。在聚类阶段,改进的自适应三支密度峰值聚类算法(three-way adaptive density peak clustering,3W-ADPC)通过结合自适应近邻搜索和三支聚类算法提升负荷聚类效果。实证结果表明,所提方法具备在计算效率和聚类精度上的双重优势,能够精准揭示多元负荷用户综合用能特性和深层次信息,证实所提方法在多元负荷用户行为研究中的实用价值。
文摘为了提高电力负荷监控的准确性,研究融合主成分含噪密度聚类(density-based spatial clustering of applications with noise with principal component analysis,PCADBSCAN)的混合非侵入式负荷辨识方法。首先,针对原始负荷特征维度较高的问题,采用主成分分析算法对原始特征数据降维,构建负荷特征模板库,同时,获取负荷电流波形,构建负荷电流模板库。其次,采用基于密度的聚类算法对负荷特征模板库内的样本进行非监督聚类,提取各聚类簇中心。然后,计算待辨识负荷与各特征模板库聚类中心的欧式距离,完成负荷特征匹配,并计算待辨识负荷的电流波形与电流模板库内各电流波形的综合关联度,完成负荷电流波形匹配。最后,混合两次匹配结果,综合判断待辨识负荷,从而实现高可靠辨识。基于用电数据测试数据集的仿真结果显示,该方法各项指标均超过96%。
