Guanghua Xu

Real-coded chaotic quantum-inspired genetic algorithm for training of fuzzy neural networks

In this paper, a novel approach to adjusting the weightings of fuzzy neural networks using a Real-coded Chaotic Quantum-inspired genetic Algorithm (RCQGA) is proposed. Fuzzy neural networks are traditionally trained by using gradient-based methods, which may fall into local minimum during the learning process. To overcome the problems encountered by the conventional learning methods, RCQGA algorithms are adopted because of their capabilities of directed random search for global optimization. It is well known, however, that the searching speed of the conventional quantum genetic algorithms (QGA) is not satisfactory. In this paper, a real-coded chaotic quantum-inspired genetic algorithm (RCQGA) is proposed based on the chaotic and coherent characters of Q-bits. In this algorithm, real chromosomes are inversely mapped to Q-bits in the solution space. Q-bits probability-guided real cross and chaos mutation are applied to the evolution and searching of real chromosomes. Chromosomes consisting of the weightings of the fuzzy neural network are coded as an adjustable vector with real number components that are searched by the RCQGA. Simulation results have shown that faster convergence of the evolution process in searching for an optimal fuzzy neural network can be achieved. Examples of nonlinear functions approximated by using the fuzzy neural network via the RCQGA are demonstrated to illustrate the effectiveness of the proposed method.

Remaining useful life estimation for mechanical systems based on similarity of phase space trajectory

RUL is estimated based on the similarity of phase space trajectory.Nonlinear degradation evolution is revealed by the phase space trajectory.Trajectory matching is not influenced by the scaling and shifting.The estimation accuracy is verified by simulated data and actual data. When evolving from a normal state to failure, mechanical systems undergo a gradual degradation process. Due to the nonlinearity of damage accumulation, degradation data always exhibit a distinctive trend and random fluctuations. It makes the prediction of remaining useful life (RUL) very difficult and inaccurate. The phase space trajectory reconstructed from the time series of degradation data is capable of reliably elucidating the nonlinear degradation behavior. In this paper, a novel method based on the similarity of the phase space trajectory is proposed for estimating the RUL of mechanical systems. First, the reference degradation trajectories are built with historical degradation data using the phase space reconstruction. Second, the similarities between the current degradation trajectory and the reference degradation trajectories are measured with a normalized cross correlation indicator, which is determined solely by the trajectory shape and is not interfered with the scaling and shifting of the trajectory. Trajectory shape and degradation stage matching algorithms are combined to find the optimal segments in the reference degradation trajectories compared with the current degradation trajectory. Finally, the RULs corresponding to the optimal matching segments are subjected to weighted averaging to obtain the RUL of the current degradation process. The proposed method is evaluated utilizing both simulated data in stochastic degradation processes and experimental data measured on an actual pump. The results show that the predicted RULs are very close to the actual RUL.

Feature extraction and recognition for rolling element bearing fault utilizing short-time Fourier transform and non-negative matrix factorization

Due to the non-stationary characteristics of vibration signals acquired from rolling element bearing fault, the time-frequency analysis is often applied to describe the local information of these unstable signals smartly. However, it is difficult to classify the high dimensional feature matrix directly because of too large dimensions for many classifiers. This paper combines the concepts of time-frequency distribution(TFD) with non-negative matrix factorization(NMF), and proposes a novel TFD matrix factorization method to enhance representation and identification of bearing fault. Throughout this method, the TFD of a vibration signal is firstly accomplished to describe the localized faults with short-time Fourier transform(STFT). Then, the supervised NMF mapping is adopted to extract the fault features from TFD. Meanwhile, the fault samples can be clustered and recognized automatically by using the clustering property of NMF. The proposed method takes advantages of the NMF in the parts-based representation and the adaptive clustering. The localized fault features of interest can be extracted as well. To evaluate the performance of the proposed method, the 9 kinds of the bearing fault on a test bench is performed. The proposed method can effectively identify the fault severity and different fault types. Moreover, in comparison with the artificial neural network(ANN), NMF yields 99.3% mean accuracy which is much superior to ANN. This research presents a simple and practical resolution for the fault diagnosis problem of rolling element bearing in high dimensional feature space.

Numerical Simulation of Nonlinear Lamb Waves Used in a Thin Plate for Detecting Buried Micro-Cracks

Compared with conventional linear ultrasonic inspection methods, which are sensitive only to severe defects, nonlinear ultrasonic inspection methods are better for revealing micro-cracks in thin plates. However, most nonlinear ultrasonic inspection methods have only been experimentally investigated using bulk or Rayleigh waves. Numerical studies, especially numerical simulations of Lamb ultrasonic waves, have seldom been reported. In this paper, the interaction between nonlinear S0 mode Lamb waves and micro-cracks of various lengths and widths buried in a thin metallic plate was simulated using the finite element method (FEM). The numerical results indicate that after interacting with a micro-crack, a new wave-packet was generated in addition to the S0 mode wave-packet. The second harmonics of the S0 mode Lamb waves and the new wave-packet were caused by nonlinear acoustic effects at the micro-crack. An amplitude ratio indicator is thus proposed for the early detection of buried micro-cracks.

Analytical and numerical studies of approximate phase velocity matching based nonlinear S0 mode Lamb waves for the detection of evenly distributed microstructural changes

Most previous studies on nonlinear Lamb waves are conducted using mode pairs that satisfying strict phase velocity matching and non-zero power flux criteria. However, there are some limitations in existence. First, strict phase velocity matching is not existed in the whole frequency bandwidth; Second, excited center frequency is not always exactly equal to the true phase-velocity-matching frequency; Third, mode pairs are isolated and quite limited in number; Fourth, exciting a single desired primary mode is extremely difficult in practice and the received signal is quite difficult to process and interpret. And few attention has been paid to solving these shortcomings. In this paper, nonlinear S0 mode Lamb waves at low-frequency range satisfying approximate phase velocity matching is proposed for the purpose of overcoming these limitations. In analytical studies, the secondary amplitudes with the propagation distance considering the fundamental frequency, the maximum cumulative propagation distance (MCPD) with the fundamental frequency and the maximum linear cumulative propagation distance (MLCPD) using linear regression analysis are investigated. Based on analytical results, approximate phase velocity matching is quantitatively characterized as the relative phase velocity deviation less than a threshold value of 1%. Numerical studies are also conducted using tone burst as the excitation signal. The influences of center frequency and frequency bandwidth on the secondary amplitudes and MCPD are investigated. S1–S2 mode with the fundamental frequency at 1.8 MHz, the primary S0 mode at the center frequencies of 100 and 200 kHz are used respectively to calculate the ratios of nonlinear parameter of Al 6061-T6 to Al 7075-T651. The close agreement of the computed ratios to the actual value verifies the effectiveness of nonlinear S0 mode Lamb waves satisfying approximate phase velocity matching for characterizing the material nonlinearity. Moreover, the ratios derived from the primary and secondary horizontal displacements generated from nonlinear S0 mode Lamb waves are closest to the real value, which indicates that using horizontal displacements is more suitable for detecting evenly distributed microstructural changes in large thin plate-like structure. Successful application to evaluating material at different levels of evenly distributed fatigue damage is also numerically conducted.

An automatic patient-specific seizure onset detection method in intracranial EEG based on incremental nonlinear dimensionalityreduction

Epileptic seizure features always include the morphology and spatial distribution of nonlinear waveforms in the electroencephalographic (EEG) signals. In this study, we propose a novel incremental learning scheme based on nonlinear dimensionality reduction for automatic patient-specific seizure onset detection. The method allows for identification of seizure onset times in long-term EEG signals acquired from epileptic patients. Firstly, a nonlinear dimensionality reduction (NDR) method called local tangent space alignment (LTSA) is used to reduce the dimensionality of available initial feature sets extracted with continuous wavelet transform (CWT). One-dimensional manifold which reflects the intrinsic dynamics of seizure onset is obtained. For each patient, IEEG recordings containing one seizure onset is sufficient to train the initial one-dimensional manifold. Secondly, an unsupervised incremental learning scheme is proposed to update the initial manifold when the unlabelled EEG segments flow in sequentially. The incremental learning scheme can cluster the new coming samples into the trained patterns (containing or not containing seizure onsets). Intracranial EEG recordings from 21 patients with duration of 193.8h and 82 seizures are used for the evaluation of the method. Average sensitivity of 98.8%, average uninteresting false positive rate of 0.24/h, average interesting false positives rate of 0.25/h, and average detection delay of 10.8s are obtained. Our method offers simple, accurate training with less human intervening and can be well used in off-line seizure detection. The unsupervised incremental learning scheme has the potential in identifying novel IEEG classes (different onset patterns) within the data.

Effects of Mental Load and Fatigue on Steady-State Evoked Potential Based Brain Computer Interface Tasks: A Comparison of Periodic Flickering and Motion-Reversal Based Visual Attention.

Steady-state visual evoked potentials (SSVEP) based paradigm is a conventional BCI method with the advantages of high information transfer rate, high tolerance to artifacts and the robust performance across users. But the occurrence of mental load and fatigue when users stare at flickering stimuli is a critical problem in implementation of SSVEP-based BCIs. Based on electroencephalography (EEG) power indices α, θ, θ + α, ratio index θ/α and response properties of amplitude and SNR, this study quantitatively evaluated the mental load and fatigue in both of conventional flickering and the novel motion-reversal visual attention tasks. Results over nine subjects revealed significant mental load alleviation in motion-reversal task rather than flickering task. The interaction between factors of “stimulation type” and “fatigue level” also illustrated the motion-reversal stimulation as a superior anti-fatigue solution for long-term BCI operation. Taken together, our work provided an objective method favorable for the design of more practically applicable steady-state evoked potential based BCIs.

Addition of visual noise boosts evoked potential-based brain-computer interface

Although noise has a proven beneficial role in brain functions, there have not been any attempts on the dedication of stochastic resonance effect in neural engineering applications, especially in researches of brain-computer interfaces (BCIs). In our study, a steady-state motion visual evoked potential (SSMVEP)-based BCI with periodic visual stimulation plus moderate spatiotemporal noise can achieve better offline and online performance due to enhancement of periodic components in brain responses, which was accompanied by suppression of high harmonics. Offline results behaved with a bell-shaped resonance-like functionality and 7–36% online performance improvements can be achieved when identical visual noise was adopted for different stimulation frequencies. Using neural encoding modeling, these phenomena can be explained as noise-induced input-output synchronization in human sensory systems which commonly possess a low-pass property. Our work demonstrated that noise could boost BCIs in addressing human needs.

Immune Clonal Selection Optimization Method with Mixed Mutation Strategies

In artificial immune optimization algorithm, the mutation of immune cells has been considered as the key operator that determines the algorithm performance. Traditional immune optimization algorithms have used a single mutation operator, typically a Gaussian. A mixed mutation strategy may be more efficient than a single one. In view of this, a mixed mutation strategy using Gaussian and Cauchy mutations is presented, and a novel clonal selection optimization method based on clonal selection principle is proposed also. The experimental results show the mixed strategy can obtain the same performance as the best of pure strategies or even better in some cases.

Detecting of Driver's Drowsiness Using Multiwavelet Packet Energy Spectrum

Driver drowsiness/fatigue is an important cause of combination-unit truck crashes. The purpose of this study is to detect drowsiness in drivers unobtrusively to prevent accidents and to improve safety on the highways. This paper is the culmination of previous work to determine if steering behavior could be used to unobtrusively detect driver fatigue. In order to investigate the relationship of driver fatigue with steering wheel motion, a fatigue driving model was established. The results of computer simulation demonstrate that the information of driver fatigue hide in steering wheel motion. In order to extract state information contained in it, a multiple scaling functions based multi-wavelet algorithm was proposed which can represent the fatigues characteristic signal and the fatigue characteristic library was be built by using multi wavelet frequency band energy which can represent the fatigue feature of a driver. Finally the paper propose a method of detecting the fatigue driving state by using Support Vector Machines (SVM) which provide the theoretical basis of the development of simple practical driving fatigue monitoring devices. Keywords-component; formatting; style; styling; insert (key words)