A multiscale principal component analysis method is proposed for sensor fault detection and identification. After decomposition of sensor signal by wavelet transform, the coarse-scale coefficients from the sensors wit...A multiscale principal component analysis method is proposed for sensor fault detection and identification. After decomposition of sensor signal by wavelet transform, the coarse-scale coefficients from the sensors with strong correlation are employed to establish the principal component analysis model. A moving window is designed to monitor data from each sensor using the model. For the purpose of sensor fault detection and identification, the data in the window is decomposed with wavelet transform to acquire the coarse-scale coefficients firstly, and the square prediction error is used to detect the failure. Then the sensor validity index is introduced to identify faulty sensor, which provides a quantitative identifying index rather than qualitative contrast given by the approach with contribution. Finally, the applicability and effectiveness of the proposed method is illustrated by sensors of industrial boiler.展开更多
In this paper,an improved zerotree structure and a new coding procedure are adopted,which improve the reconstructed image qualities.Moreover,the lists in SPIHT are replaced by flag maps,and lifting scheme is adopted t...In this paper,an improved zerotree structure and a new coding procedure are adopted,which improve the reconstructed image qualities.Moreover,the lists in SPIHT are replaced by flag maps,and lifting scheme is adopted to realize wavelet transform,which lowers the memory requirements and speeds up the coding process.Experimental results show that the algorithm is more effective and efficient compared with SPIHT.展开更多
Too many sensors and data information in structural health monitoring system raise the problem of how to realize multi-sensor information fusion. An experiment on a three-story frame structure was conducted to obtain ...Too many sensors and data information in structural health monitoring system raise the problem of how to realize multi-sensor information fusion. An experiment on a three-story frame structure was conducted to obtain vibration test data in 36damage cases. A coupling neural network (NN) based on multi-sensor information fusion is proposed to achieve identification of damage occurrence, damage localization and damage quantification, respectively. First, wavelet packet transform (WPT) is used to extract features of vibration test data from structure with different damage extent. Then, data fusion is conducted by assembling feature vectors of different type sensors. Finally, three sets of coupling NN are constructed to implement decision fusion and damage identification. The results of experimental study proved the validity and feasibility of the proposed methodology.展开更多
Based on the good localization characteristic of the wavelet transform both in time and frequency domain, a de-noising method based on wavelet transform is presented, which can make the extraction of visual evoked pot...Based on the good localization characteristic of the wavelet transform both in time and frequency domain, a de-noising method based on wavelet transform is presented, which can make the extraction of visual evoked potentials in single training sample from the EEG background noise in favor of studying the changes between the single sample response happen. The information is probably related with the different function, appearance and pathologies of the brain. At the same time this method can also be used to remove those signal’s artifacts that do not appear with EP within the same scope of time or frequency. The traditional Fourier filter can hardly attain the similar result. This method is different from other wavelet de-noising methods in which different criteria are employed in choosing wavelet coefficient. It has a biggest virtue of noting the differences among the single training sample and making use of the characteristics of high time frequency resolution to reduce the effect of interference factors to a maximum extent within the time scope that EP appear. The experiment result proves that this method is not restricted by the signal-to-noise ratio of evoked potential and electroencephalograph (EEG) and even can recognize instantaneous event under the condition of lower signal-to-noise ratio, as well as recognize the samples which evoked evident response more easily. Therefore, more evident average evoked response could be achieved by de-nosing the signals obtained through averaging out the samples that can evoke evident responses than de-nosing the average of original signals. In addition, averaging methodology can dramatically reduce the number of record samples needed, thus avoiding the effect of behavior change during the recording process. This methodology pays attention to the differences among single training sample and also accomplishes the extraction of visual evoked potentials from single trainings sample. As a result, system speed and accuracy could be improved to a great extent if this methodology is applied to brain-computer interface system based on evoked responses.展开更多
基金Supported by National Natural Science Foundation of P. R. China (60572010)
文摘A multiscale principal component analysis method is proposed for sensor fault detection and identification. After decomposition of sensor signal by wavelet transform, the coarse-scale coefficients from the sensors with strong correlation are employed to establish the principal component analysis model. A moving window is designed to monitor data from each sensor using the model. For the purpose of sensor fault detection and identification, the data in the window is decomposed with wavelet transform to acquire the coarse-scale coefficients firstly, and the square prediction error is used to detect the failure. Then the sensor validity index is introduced to identify faulty sensor, which provides a quantitative identifying index rather than qualitative contrast given by the approach with contribution. Finally, the applicability and effectiveness of the proposed method is illustrated by sensors of industrial boiler.
基金Supported by Korea ETRI cooperationfoundation(12003121192202) .
文摘In this paper,an improved zerotree structure and a new coding procedure are adopted,which improve the reconstructed image qualities.Moreover,the lists in SPIHT are replaced by flag maps,and lifting scheme is adopted to realize wavelet transform,which lowers the memory requirements and speeds up the coding process.Experimental results show that the algorithm is more effective and efficient compared with SPIHT.
文摘Too many sensors and data information in structural health monitoring system raise the problem of how to realize multi-sensor information fusion. An experiment on a three-story frame structure was conducted to obtain vibration test data in 36damage cases. A coupling neural network (NN) based on multi-sensor information fusion is proposed to achieve identification of damage occurrence, damage localization and damage quantification, respectively. First, wavelet packet transform (WPT) is used to extract features of vibration test data from structure with different damage extent. Then, data fusion is conducted by assembling feature vectors of different type sensors. Finally, three sets of coupling NN are constructed to implement decision fusion and damage identification. The results of experimental study proved the validity and feasibility of the proposed methodology.
文摘Based on the good localization characteristic of the wavelet transform both in time and frequency domain, a de-noising method based on wavelet transform is presented, which can make the extraction of visual evoked potentials in single training sample from the EEG background noise in favor of studying the changes between the single sample response happen. The information is probably related with the different function, appearance and pathologies of the brain. At the same time this method can also be used to remove those signal’s artifacts that do not appear with EP within the same scope of time or frequency. The traditional Fourier filter can hardly attain the similar result. This method is different from other wavelet de-noising methods in which different criteria are employed in choosing wavelet coefficient. It has a biggest virtue of noting the differences among the single training sample and making use of the characteristics of high time frequency resolution to reduce the effect of interference factors to a maximum extent within the time scope that EP appear. The experiment result proves that this method is not restricted by the signal-to-noise ratio of evoked potential and electroencephalograph (EEG) and even can recognize instantaneous event under the condition of lower signal-to-noise ratio, as well as recognize the samples which evoked evident response more easily. Therefore, more evident average evoked response could be achieved by de-nosing the signals obtained through averaging out the samples that can evoke evident responses than de-nosing the average of original signals. In addition, averaging methodology can dramatically reduce the number of record samples needed, thus avoiding the effect of behavior change during the recording process. This methodology pays attention to the differences among single training sample and also accomplishes the extraction of visual evoked potentials from single trainings sample. As a result, system speed and accuracy could be improved to a great extent if this methodology is applied to brain-computer interface system based on evoked responses.