The measured parameters change with the machine degradation, where the trends of the parameters can be applied in condition prediction. The deterioration of the parameters usually present characteristics in a monotoni...
详细信息
The measured parameters change with the machine degradation, where the trends of the parameters can be applied in condition prediction. The deterioration of the parameters usually present characteristics in a monotonic or low-frequency way. As the measure signal are merged with instant signals, noise and measurement error in different frequency bands, model decomposition approaches are widely applied to separate the different bands for feature extraction. Among the model decomposition methods, Variational model decomposition is recently proposed by K. Dragomiretskiy and D. Zosso as a non-recursive model which can extract modes concurrently. This work will analyze its effectiveness in degradation feature extraction and application in condition prediction in prognosis. Firstly, variational model decomposition is analyzed, where the characteristics of variational model decomposition are compared with another mode decomposition method ensemble empirical mode decomposition. simulation data is applied to test the effectiveness of the mode decomposition methods. Secondly, field data with component degradation is collected, and the degradation feature extraction is applied with the three methods mentioned above.
In order to maintain the separability of different fault onsets, more characteristic parameters need to be used to establish the feature space with multiple possible faults. Usually, the dimensions of feature space ha...
详细信息
In order to maintain the separability of different fault onsets, more characteristic parameters need to be used to establish the feature space with multiple possible faults. Usually, the dimensions of feature space have a positive correlation with the number of faults. Describing multiple faults in visualization is more difficult and the classification bounds with hyperplanes are less convenient to realize. Therefore implementing dimensionality reduction approaches in original feature space is beneficial for fault diagnosis and prognostics. Multidimensional Scaling (MDS) has been widely studied and used for dimensionality reduction as well as Principal Component Analysis (PCA). When these methods are applied in the prognostics process, the effectiveness needs validation with the reason that the deterioration features in degradations are different from the faults onsets that in diagnostics. The paper analyzes the nonlinear dimensionality reduction methods including Self-Organizing Map (SOM) and Isometric feature mapping (Isomap). Based on the deterioration data in an engine, the effects of these methods are analyzed.
An active distribution systempower-supply planning model considering cooling,heating and power load balance is proposed in this paper.A regional energy service company is assumed to be in charge of the investment and...
详细信息
An active distribution systempower-supply planning model considering cooling,heating and power load balance is proposed in this paper.A regional energy service company is assumed to be in charge of the investment and operation for the system in the *** expansion of substations,building up distributed combined cooling,heating and power(CCHP),gas heating boiler(GHB)and air conditioner(AC)are included as investment planning *** terms of operation,the load scenarios are divided into heating,cooling and transition ***,the extreme load scene is included to assure the power supply reliability of the *** results demonstrate the effectiveness of the proposed model and illustrate the economic benefits of applying distributed CCHP in regional power supply on investment and operation.
The paper analyses the existed protection of microgrid. And in order to meet the requirements of microgrid protection, a novel protection is presented in this paper. The protection based on local information without t...
详细信息
ISBN:
(纸本)9781849192125
The paper analyses the existed protection of microgrid. And in order to meet the requirements of microgrid protection, a novel protection is presented in this paper. The protection based on local information without the need of communication, uses current traveling waves to identify faulted feeder, and the bus bar voltages to determine a event caused by fault or switch operation. In the protection scheme, current traveling waves are measured by current transducers in N line and phase lines, and wavelet multi-resolution analysis is used to decompose traveling wave signals. The initial traveling waves are compared in magnitude and polarity with each other to determine the faulted feeder. The protection is independent of microgrid operation connected with the main power grid or in island.
Advanced applications of energy storage system comprise load shifting, isolated island operation, frequency regulation and reactive support, etc. There are already algorithms and control strategies for advanced applic...
详细信息
ISBN:
(纸本)9781849197588
Advanced applications of energy storage system comprise load shifting, isolated island operation, frequency regulation and reactive support, etc. There are already algorithms and control strategies for advanced applications, but still lack an independent supporting software platform. This paper concentrates on the structure and realization of software platform for advanced application of large-scale energy storage system, including demand analysis, software architecture and module design, user interface and functionality development. An easy to operate and expand software platform, with a beautiful graphical user interface and support for visualization of data, was designed and implemented.
作者:
Liu, JiyeZheng, ZedongTsinghua University
State Key Lab of Control and Simulation of Power Systems and Generation Equipments Department of Electrical Engineering Beijing China
Paralleling SiC MOSFETs is a popular way to increase the current capacity. However, due to asymmetric layout, device parameter mismatch and other factors, imbalanced switching or dynamic current is observed among the ...
详细信息
This paper uses the least squares method,BP neural network and gray GM(1,1) model for trend forecasts and predicting the remaining life of the aero-engine fault *** comparing the results of the three methods,the least...
详细信息
ISBN:
(纸本)9781510806474
This paper uses the least squares method,BP neural network and gray GM(1,1) model for trend forecasts and predicting the remaining life of the aero-engine fault *** comparing the results of the three methods,the least squares method and BP neural network method predicting results are more exact than the gray GM(1,1) *** three methods are also used for predicting the actual data,and the result is close to the real.
The vibration signals of wind turbines are highly nonlinear and non-stationary due to wind turbine operation conditions that are very complicated. The signals will be more complex when a fault occurs. Aiming at these ...
详细信息
Under the skin effect caused by the alternating electromagnetic field, the transmission line parameters changes with frequency, and different line resistivity, surge impedance, wave speed are presented in one line und...
详细信息
This paper analyzes the general process of power flow transferring and divides it into four parts: 1) the process of first phase selection after fault occurs;2) the process of selective action at the first time;3) the...
详细信息
暂无评论