改进的主成分分析双向变量选择法及应用

改进的主成分分析双向变量选择法及应用吴继曾，吴凯峰ＴｈｅＩｍｐｒｏｖｅｄＭｅｔｈｏｄｏｆＭａｉｎＣｏｍｐｏｎｅｎｔＡｎａｌｙｚｅＴｗｏ－ＷａｙＶａｒｉａｂｌｅＳｅｌｅｃｔｉｏｎａｎｄＩｔ＇ｓＡｐｐｌｉｃａｔｉｏｎ￥ＷｕＪｉｚｅｎｇＷｕＫａｉｆｅｎｇ１...

Data mining and well logging interpretation: application to a conglomerate reservoir

Data mining is the process of extracting implicit but potentially useful information from incomplete, noisy, and fuzzy data. Data mining offers excellent nonlinear modeling and self-organized learning, and it can play a vital role in the interpretation of well logging data of complex reservoirs. We used data mining to identify the lithologies in a complex reservoir. The reservoir lithologies served as the classification task target and were identified using feature extraction, feature selection, and modeling of data streams. We used independent component analysis to extract information from well curves. We then used the branch-and- bound algorithm to look for the optimal feature subsets and eliminate redundant information. Finally, we used the C5.0 decision-tree algorithm to set up disaggregated models of the well logging curves. The modeling and actual logging data were in good agreement, showing the usefulness of data mining methods in complex reservoirs.

Carbonization mechanism of bamboo (phyllostachys) by means of Fourier Transform Infrared and elemental analysis

通过测定在200-600℃炭化竹材得到的固体产物的碳、氢、氧元素的含量及它们的红外光谱,研究了在炭化过程中竹材中半纤维素、纤维素及木素的变化规律。结果表明,结合元素分析,红外光谱分析方法是研究竹材炭化机理的有效手段。在200℃以前,竹材中的半纤维素和纤维素的大量羟基断裂,并结合成水而失去。在200-250℃之间,竹材中的纤维素被降解,其中的吡喃型环也遭到破坏。并且木素中的甲氧基也被脱去。竹材中的木素网状结构在250-400℃之间遭到完全的破坏。竹炭中的碳原子在600℃已基本上完成了芳环化。图3表2参15。

On-line Batch Process Monitoring with Improved Multi-way Independent Component Analysis

In the past decades, on-line monitoring of batch processes using multi-way independent component analysis （MICA） has received considerable attention in both academia and industry. This paper focuses on two troublesome issues concerning selecting dominant independent components without a standard criterion and deter- mining the control limits of monitoring statistics in the presence of non-Gaussian distribution. To optimize the number of key independent components~ we introctuce-anoveiconcept of-system-cleviation, which is ab^e＇io＇evalu[ ate the reconstructed observations with different independent components. The monitored statistics arc transformed to Gaussian distribution data by means of Box-Cox transformation, which helps readily determine the control limits. The proposed method is applied to on-line monitoring of a fed-hatch penicillin fermentation simulator, and the ex- _perimental results indicate the advantages of the improved MICA monitoring compared to the conventional methods.

Numerical study of resting-state fMRI based on kernel ICA

In order to facilitate the extraction of the default mode network（DMN）, reduce the data complexity of the functional magnetic resonance imaging （fMRI）and overcome the restriction of the linearity of the mixing process encountered with the independent component analysis（ICA）, a framework of dimensionality reduction and nonlinear transformation is proposed. First, the principal component analysis（PCA）is applied to reduce the time dimension 153 594×128 of the fMRI data to 153 594×5 for simplifying complexity computation and obtaining 95% of the information. Secondly, a new kernel-based nonlinear ICA method referred as the kernel ICA（KICA）based on the Gaussian kernel is introduced to analyze the resting-state fMRI data and extract the DMN. Experimental results show that the KICA provides a better performance for the resting-state fMRI data analysis compared with the classical ICA. Furthermore, the DMN is accurately extracted and the noise is reduced.

Study of engine noise based on independent component analysis

Independent component analysis was applied to analyze the acoustic signals from diesel engine. First the basic prin-ciple of independent component analysis (ICA) was reviewed. Diesel engine acoustic signal was decomposed into several inde-pendent components (ICs); Fourier transform and continuous wavelet transform (CWT) were applied to analyze the independent components. Different noise sources of the diesel engine were separated, based on the characteristics of different component in time-frequency domain.

Independent component analysis approach for fault diagnosis of condenser system in thermal power plant

A statistical signal processing technique was proposed and verified as independent component analysis(ICA) for fault detection and diagnosis of industrial systems without exact and detailed model.Actually,the aim is to utilize system as a black box.The system studied is condenser system of one of MAPNA's power plants.At first,principal component analysis(PCA) approach was applied to reduce the dimensionality of the real acquired data set and to identify the essential and useful ones.Then,the fault sources were diagnosed by ICA technique.The results show that ICA approach is valid and effective for faults detection and diagnosis even in noisy states,and it can distinguish main factors of abnormality among many diverse parts of a power plant's condenser system.This selectivity problem is left unsolved in many plants,because the main factors often become unnoticed by fault expansion through other parts of the plants.

Batch process monitoring based on multilevel ICA-PCA

In this paper, we describe a new batch process monitoring method based on multilevel independent component analysis and principal component analysis (MLICA-PCA). Unlike the conventional multi-way principal component analysis (MPCA) method, MLICA-PCA provides a separated interpretation for multilevel batch process data. Batch process data are partitioned into two levels: the within-batch level and the between-batch level. In each level, the Gaussian and non-Gaussian components of process information can be separately extracted. I2, T2 and SPE statistics are individually built and monitored. The new method facilitates fault diagnosis. Since the two variation levels are decomposed, the variables responsible for faults in each level can be identified and interpreted more easily. A case study of the Dupont benchmark process showed that the proposed method was more efficient and interpretable in fault detection and diagnosis, compared to the alternative batch process monitoring method.

Blind Separation of Speech Signals Based on Wavelet Transform and Independent Component Analysis

Speech signals in frequency domain were separated based on discrete wavelet transform (DWT) and independent component analysis (ICA). First, mixed speech signals were decomposed into different frequency domains by DWT and the subbands of speech signals were separated using ICA in each wavelet domain; then, the permutation and scaling problems of frequency domain blind source separation (BSS) were solved by utilizing the correlation between adjacent bins in speech signals; at last, source signals were reconstructed from single branches. Experiments were carried out with 2 sources and 6 microphones using speech signals at sampling rate of 40 kHz. The microphones were aligned with 2 sources in front of them, on the left and right. The separation of one male and one female speeches lasted 2.5 s. It is proved that the new method is better than single ICA method and the signal to noise ratio is improved by 1 dB approximately.

Single channel blind source separation based on ICA feature extraction

A new technique is proposed to solve the blind source separation (BSS) given only a single channel observation. The basis functions and the density of the coefficients of source signals learned by ICA are used as the prior knowledge. Based on the learned prior information the learning rules of single channel BSS are presented by maximizing the joint log likelihood of the mixed sources to obtain source signals from single observation, in which the posterior density of the given measurements is maximized. The experimental results exhibit a successful separation performance for mixtures of speech and music signals.

Application of Kernel Independent Component Analysis for Multivariate Statistical Process Monitoring

In this research, a new fault detection method based on kernel independent component analysis （kernel ICA） is developed. Kernel ICA is an improvement of independent component analysis （ICA）, and is different from kernel principal component analysis （KPCA） proposed for nonlinear process monitoring. The basic idea of our approach is to use the kernel ICA to extract independent components efficiently and to combine the selected essential independent components with process monitoring techniques. 12 （the sum of the squared independent scores） and squared prediction error （SPE） charts are adopted as statistical quantities. The proposed monitoring method is applied to Tennessee Eastman process, and the simulation results clearly show the advantages of kernel ICA monitoring in comparison to ICA monitoring.

Nonlinear Statistical Process Monitoring Based on Control Charts with Memory Effect and Kernel Independent Component Analysis

A novel nonlinear combination process monitoring method was proposed based on techniques with memo- ry effect （multivariate exponentially weighted moving average （MEWMA）） and kernel independent component analysis （KICA）. The method was developed for dealing with nonlinear issues and detecting small or moderate drifts in one or more process variables with autocorrelation. MEWMA charts use additional information from the past history of the process for keeping the memory effect of the process behavior trend. KICA is a recently devel- oped statistical technique for revealing hidden, nonlinear statistically independent factors that underlie sets of mea- surements and it is a two-phase algorithm., whitened kernel principal component analysis （KPCA） plus indepen- dent component analysis （ICA）. The application to the fluid catalytic cracking unit （FCCU） simulated process in- dicates that the proposed combined method based on MEWMA and KICA can effectively capture the nonlinear rela- tionship and detect small drifts in process variables. Its performance significantly outperforms monitoring method based on ICA, MEWMA-ICA and KICA, especially for lonu-term performance deterioration.

THE FAST FIXED-POINT ALGORITHM FOR SPECKLE REDUCTION OF POLARIMETRIC SAR IMAGE

In this letter, a simple and efficient method of image speckle reduction for polari- metric SAR is put forward. It is based on the fast fixed-point ICA (Independent Component Analysis) algorithm of orthogonal and symmetric matrix. Simulation experiment is carried out to separate speckle noise from the polarimetric SAR images, and it indicates that this algorithm has high convergency speed and stability, the image speckle noise is reduced effectively and the speckle index is low, and the image quality is improved obviously.

OPTIMAL ANTENNA SUBSET SELECTION AND BLIND DETECTION APPROACH APPLIED TO ORTHOGONAL SPACE-TIME BLOCK CODING

An approach combining optimal antenna subset selection with blind detection scheme for Orthogonal Space-Time Block Coding (OSTBC) is proposed in this paper. The optimal antenna sub- set selection is taken into account at transmitter and/or receiver sides, which chooses the optimal an- tennas to increase the diversity order of OSTBC and improve further its performance. In order to en- hance the robustness of the detection used in the conventional OSTBC scheme, a blind detection scheme based on Independent Component Analysis (ICA) is exploited which can directly extract transmitted signals without channel estimation. Performance analysis shows that the proposed ap- proach can achieve the full diversity and the flexibility of system design by using the antenna selec-tion and the ICA based blind detection schemes.

Processing Human Colonic Pressure Signals by Using Overdetermined ICA

Independent component analysis （ICA） is a widely used method for blind source separation （BSS）. The mature ICA model has a restriction that the number of the sources must equal to that of the sensors used to collect data, which is hard to meet in most practical cases. In this paper, an overdetermined ICA method is proposed and successfully used in the analysis of human colonic pressure signals. Using principal component analysis （PCA）, the method estimates the number of the sources firstly and reduces the dimensions of the observed signals to the same with that of the sources; and then, Fast- ICA is used to estimate all the sources. From 26 groups of colonic pressure recordings, several colonic motor patterns are extracted, which riot only prove the effectiveness of this method, but also greatly facilitate further medical researches.

Removal of jamming using independent component analysis in non-cooperative passive detection system

Due to electronic jamming transmitted by hostile electromc jamming equtpmcnts tional jamming from other illuminating sources in the complex electromagnetic environment, the per- formance of non-cooperative passive detection systems may degrade it significantly. To solve the problem, a receiving frame with multiple channels for signal preprocessing is designed and a theoret- ical analysis to the received signals in the complex electromagnetic environment is provided. Fur- thermore, a scheme for jamming removal using independent component analysis is proposed. Simula- tion results demonstrate the proposed scheme appears as a very appealing solution for removal of jam- ming and an approximate lOdB signal to distortion ratio over traditional schemes is obtained.

High Efficiency Stereo Audio Compression Method Using Polar Coordinate Principle Component Analysis for Wireless Communications

High efficiency audio compression is the basic technology in audio involved multimedia communications. Downmixing and parametric coding is efficient coding scheme with wide applications in some up-to-date audio codecs such as Parametric Stereo (PS) in EAAC+ and MPEG-Surround. Principle Component Analysis (PCA) stereo coding followed this idea to map two channels to one channel with maximum energy and parameterize the secondary channel. This paper investigates the conventional PCA method performance under general stereo model with multiple sound sources and different directions, and then proposes a Polar Coordinate based PCA (PC-PCA) stereo coding method. It has been proved that when multiple sound sources exist with different directions, PC-PCA is better than the conventional PCA method when Mean to Standard deviation Ratio (MSR) is large. A stereo codec based on PC-PCA is proposed to validate the performance improvement of proposed method. Objective and subjective tests show the proposed method achieves a comparative quality and saves 50% parameter bit rate comparing with conventional PCA method, and obtains a 4-8 MUSHRA scores improvement comparing with state-of-the-art stereo codec at the same parameter bit rate.

Detection and Separation of Event-related Potentials from Multi-Artifacts Contaminated EEG by Means of Independent Component Analysis

Event-related potentials (ERP) is an important type of brain dynamics in human cognition research. However, ERP is often submerged by the spontaneous brain activity EEG, for its relatively tiny scale. Further more, the brain activities collected from scalp electrodes are often inevitably contaminated by several kinds of artifacts, such as blinks, eye movements, muscle noise and power line interference. A new approach to correct these disturbances is presented using independent component analysis (ICA). This technique can effectively detect and extract ERP components from the measured electrodes recordings even if they are heavily contaminated. The results compare favorably to those obtained by parametric modeling. Besides, auto-adaptive projection of decomposed results to ERP components was also given. Through experiments, ICA proves to be highly capable of ERP extraction and S/N ratio improving.

An Improved Fixed-point Algorithm for Independent Component Analysis of Functional MRI Data

The fixed-point algorithm and infomax algorithm are two of the most popular algorithms in independent component analysis(ICA).However,it is hard to take both stability and speed into consideration in processing functional magnetic resonance imaging(fMRI)data.In this paper,an optimization model for ICA is presented and an improved fixed-point algorithm based on the model is proposed.In the new algorithms a small step size is added to increase the stability.In order to accelerate the convergence,an improvement on Newton method is made,which makes cubic convergence for the new algorithm.Applying the algorithm and two other algorithms to invivo fMRI data,the results show that the new algorithm separates independent components stably,which has faster convergence speed and less computation than the other two algorithms.The algorithm has obvious advantage in processing fMRI signal with huge data.