Power transformer is one of the most crucial devices in power grid.It is significant to determine incipient faults of power transformers fast and accurately.Input features play critical roles in fault diagnosis accura...Power transformer is one of the most crucial devices in power grid.It is significant to determine incipient faults of power transformers fast and accurately.Input features play critical roles in fault diagnosis accuracy.In order to further improve the fault diagnosis performance of power trans-formers,a random forest feature selection method coupled with optimized kernel extreme learning machine is presented in this study.Firstly,the random forest feature selection approach is adopted to rank 42 related input features derived from gas concentration,gas ratio and energy-weighted dissolved gas analysis.Afterwards,a kernel extreme learning machine tuned by the Aquila optimization algorithm is implemented to adjust crucial parameters and select the optimal feature subsets.The diagnosis accuracy is used to assess the fault diagnosis capability of concerned feature subsets.Finally,the optimal feature subsets are applied to establish fault diagnosis model.According to the experimental results based on two public datasets and comparison with 5 conventional approaches,it can be seen that the average accuracy of the pro-posed method is up to 94.5%,which is superior to that of other conventional approaches.Fault diagnosis performances verify that the optimum feature subset obtained by the presented method can dramatically improve power transformers fault diagnosis accuracy.展开更多
Elevated atmospheric nitrogen(N) deposition has been detected in many regions of China, but its effects on soil N transformation in temperate forest ecosystems are not well known. We therefore simulated N deposition w...Elevated atmospheric nitrogen(N) deposition has been detected in many regions of China, but its effects on soil N transformation in temperate forest ecosystems are not well known. We therefore simulated N deposition with four levels of N addition rate(N0, N30, N60, and N120) for6 years in an old-growth temperate forest in Xiaoxing’an Mountains in Northeastern China. We measured gross N transformation rates in the laboratory usingN tracing technology to explore the effects of N deposition on soil gross N transformations taking advantage of N deposition soils. No significant differences in gross soil N transformation rates were observed after 6 years of N deposition with various levels of N addition rate. For all N deposition soils, the gross NH~+ immobilization rates were consistently lower than the gross N mineralization rates,leading to net N mineralization. Nitrate(NO~-) was primarily produced via oxidation of NH~+(i.e., autotrophic nitrification), whereas oxidation of organic N(i.e., heterotrophic nitrification) was negligible. Differences between the quantity of ammonia-oxidizing bacteria and ammonia-oxidizing archaea were not significant for any treatment, which likely explains the lack of a significant effect on gross nitrification rates. Gross nitrification rates were much higher than the total NO~- consumption rates,resulting in a build-up of NO~-, which highlights the high risk of N losses via NO~- leaching or gaseous N emissions from soils. This response is opposite that of typical N-limited temperate forests suffering from N deposition,suggesting that the investigated old-growth temperate forest ecosystem is likely to approach N saturation.展开更多
Soil samples were taken from an Ermans birch(Betula ermanii)-dark coniferous forest(Picea jezoensis and Abies nephrolepis)ecotone growing on volcanic ejecta in the northern slope of Changbai Mountains of Northeast Chi...Soil samples were taken from an Ermans birch(Betula ermanii)-dark coniferous forest(Picea jezoensis and Abies nephrolepis)ecotone growing on volcanic ejecta in the northern slope of Changbai Mountains of Northeast China,to compare soil carbon(C)and nitrogen(N)transformations in the two forests.The soil type is Umbri-Gelic Cambosols in Chinese Soil Taxonomy.Soil samples were incubated aerobically at 20-C and field capacity of 700 g kg-1 over a period of 27 weeks.The amount of soil microbial biomass and net N mineralization were higher in the Ermans birch than the dark coniferous forest(P<0.05),whereas the cumulative C mineralization(as CO2 emission)in the dark coniferous forest exceeded that in the Ermans birch(P<0.05).Release of the cumulative dissolved organic C and dissolved organic N were greater in the Ermans birch than the dark coniferous forest(P<0.05).The results suggested that differences of forest types could result in considerable change in soil C and N transformations.展开更多
针对水电机组振动信号故障特征提取难,提出一种融合小波变换(Wavelet Transform,WT)和奇异值分解(Singular Value Decomposition,SVD)相结合的故障特征提取方法。首先,通过小波阈值降噪消除强噪声对模型特征提取的干扰,再利用小波变换...针对水电机组振动信号故障特征提取难,提出一种融合小波变换(Wavelet Transform,WT)和奇异值分解(Singular Value Decomposition,SVD)相结合的故障特征提取方法。首先,通过小波阈值降噪消除强噪声对模型特征提取的干扰,再利用小波变换将降噪信号分解成不同频率的模态子序列,应用SVD理论提起子序列的SVD值作为特征,最终将特征输入RF模型中实现水电机组故障的快速识别与诊断。通过在公开数据集和真实机组案例中应用,验证了对水电机组故障诊断的高效性。展开更多
基金support of national natural science foundation of China(No.52067021)natural science foundation of Xinjiang(2022D01C35)+1 种基金excellent youth scientific and technological talents plan of Xinjiang(No.2019Q012)major science and technology special project of Xinjiang Uygur Autonomous Region(2022A01002-2).
文摘Power transformer is one of the most crucial devices in power grid.It is significant to determine incipient faults of power transformers fast and accurately.Input features play critical roles in fault diagnosis accuracy.In order to further improve the fault diagnosis performance of power trans-formers,a random forest feature selection method coupled with optimized kernel extreme learning machine is presented in this study.Firstly,the random forest feature selection approach is adopted to rank 42 related input features derived from gas concentration,gas ratio and energy-weighted dissolved gas analysis.Afterwards,a kernel extreme learning machine tuned by the Aquila optimization algorithm is implemented to adjust crucial parameters and select the optimal feature subsets.The diagnosis accuracy is used to assess the fault diagnosis capability of concerned feature subsets.Finally,the optimal feature subsets are applied to establish fault diagnosis model.According to the experimental results based on two public datasets and comparison with 5 conventional approaches,it can be seen that the average accuracy of the pro-posed method is up to 94.5%,which is superior to that of other conventional approaches.Fault diagnosis performances verify that the optimum feature subset obtained by the presented method can dramatically improve power transformers fault diagnosis accuracy.
基金supported by Grants from the ‘‘973’’ Project(2014CB953803)the Fundamental Research Funds for the Central Universities(2572017EA02)the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD,164320H116)
文摘Elevated atmospheric nitrogen(N) deposition has been detected in many regions of China, but its effects on soil N transformation in temperate forest ecosystems are not well known. We therefore simulated N deposition with four levels of N addition rate(N0, N30, N60, and N120) for6 years in an old-growth temperate forest in Xiaoxing’an Mountains in Northeastern China. We measured gross N transformation rates in the laboratory usingN tracing technology to explore the effects of N deposition on soil gross N transformations taking advantage of N deposition soils. No significant differences in gross soil N transformation rates were observed after 6 years of N deposition with various levels of N addition rate. For all N deposition soils, the gross NH~+ immobilization rates were consistently lower than the gross N mineralization rates,leading to net N mineralization. Nitrate(NO~-) was primarily produced via oxidation of NH~+(i.e., autotrophic nitrification), whereas oxidation of organic N(i.e., heterotrophic nitrification) was negligible. Differences between the quantity of ammonia-oxidizing bacteria and ammonia-oxidizing archaea were not significant for any treatment, which likely explains the lack of a significant effect on gross nitrification rates. Gross nitrification rates were much higher than the total NO~- consumption rates,resulting in a build-up of NO~-, which highlights the high risk of N losses via NO~- leaching or gaseous N emissions from soils. This response is opposite that of typical N-limited temperate forests suffering from N deposition,suggesting that the investigated old-growth temperate forest ecosystem is likely to approach N saturation.
基金Project supported by the National Natural Science Foundation of China(No.90411020)the National Key Basic Research Program(973 Program)of China(No.2002CB412502)
文摘Soil samples were taken from an Ermans birch(Betula ermanii)-dark coniferous forest(Picea jezoensis and Abies nephrolepis)ecotone growing on volcanic ejecta in the northern slope of Changbai Mountains of Northeast China,to compare soil carbon(C)and nitrogen(N)transformations in the two forests.The soil type is Umbri-Gelic Cambosols in Chinese Soil Taxonomy.Soil samples were incubated aerobically at 20-C and field capacity of 700 g kg-1 over a period of 27 weeks.The amount of soil microbial biomass and net N mineralization were higher in the Ermans birch than the dark coniferous forest(P<0.05),whereas the cumulative C mineralization(as CO2 emission)in the dark coniferous forest exceeded that in the Ermans birch(P<0.05).Release of the cumulative dissolved organic C and dissolved organic N were greater in the Ermans birch than the dark coniferous forest(P<0.05).The results suggested that differences of forest types could result in considerable change in soil C and N transformations.
文摘针对水电机组振动信号故障特征提取难,提出一种融合小波变换(Wavelet Transform,WT)和奇异值分解(Singular Value Decomposition,SVD)相结合的故障特征提取方法。首先,通过小波阈值降噪消除强噪声对模型特征提取的干扰,再利用小波变换将降噪信号分解成不同频率的模态子序列,应用SVD理论提起子序列的SVD值作为特征,最终将特征输入RF模型中实现水电机组故障的快速识别与诊断。通过在公开数据集和真实机组案例中应用,验证了对水电机组故障诊断的高效性。
