Strong Noise Immunity Research Articles

Concurrent fault diagnosis of small pressurized water reactors based on long-short term memory networks

The control systems of small pressurized water reactors (SPWR) with complex structures, compact layouts, and variable operating environment may be involved in various types of signal and concurrent faults. Concurrent faults can be taken as two or more single faults occurring simultaneously, which usually cause much larger damage to the system and are more difficult to be diagnosed than single faults. However, their fault diagnosis methods are rarely studied because of the numerous fault types and tremendous diagnostic difficulty. This paper explores the concurrent fault diagnosis method for sensors and actuators in SPWR control systems. An intelligent current fault diagnosis model is developed using long short-term memory network with the training and test datasets generated based on a fault simulation platform of the target SPWR. The test results show that both single and concurrent faults of the SPWR can be diagnosed rapidly in an average of 1.06 s after their occurrence with the classification and diagnosis accuracies reaching up to 96.61% and 97.27%, respectively. Moreover, by injecting different noise signals on the faulty dataset for training and validation, it is shown that the proposed LSTM network has strong noise immunity. This demonstrate the excellent diagnosis accuracy and efficiency of the model under both single and concurrent fault conditions. This study provides valuable guidance for the accuracy diagnosis of complex concurrent faults of nuclear power plants.

A novel approach to estimating harmonic/interharmonic parameters

The conventional fast Fourier transforms(FFT) approach is susceptible to noise and has a low-frequency resolution with short sampling windows. This is a bottleneck problem for accurately monitoring harmonic and interharmonic. A new method is proposed to detect the parameters of harmonic and interharmonic in this study. At the beginning, the cross-correlation function and FFT estimate the rough values of frequency and phase. Next, the error between the rough and true values is derived. Then, the incoherence of signal and noise is used to create the cost function, and the optimal initial value is computed from the rough values and the error. Finally, the initial value is used in cost function iteration, the so-called Newton–Raphson method, to calculate the frequency and phase of harmonic and interharmonic. The least squares method is used to compute the amplitudes of harmonic and interharmonic. The proposed method allows adaptively selecting the basis function for the cost function instead of using the traditional orthogonal basis. Therefore, it achieves the super-resolution evaluation of harmonic and interharmonic frequency. The simulation results demonstrate that the proposed method provides high accuracy of parameter estimation and strong noise immunity.

Research and Development in the Field of Creating Materials, Technologies and Safety Equipment

Technologies for the manufacture of microwave electromagnetic shields based on powdered carbon, foil and composite multilayer materials are presented. Technological methods for their manufacture and shielding properties are described, and the main mechanisms leading to increased shielding properties are considered depending on the manufacturing technology, composition and alternation of composite layers. It is shown that the values of the reflection coefficient of electromagnetic radiation in the frequency range 2–17 GHz for the shields considered reach a value of (–20) dB. Microwave absorbers with a reflection characteristic lower than (–10.0) dB (down to (–52.0) dB) in the frequency band 8.0–12.0 GHz are developed, and the electrophysical parameters, composition and concentration of components of the composite materials for such microwave absorbers are determined. The systems for active protection of speech information that use combined masking signals consisting of “white” noise and speech-like signals as masking signals, along with “white” noise are considered. Criteria have been developed for the approach to selecting speakers and audiences when assessing speech intelligibility to solve problems of protecting speech information and methods for experimental studies of speech intelligibility against a background of masking acoustic noise. Technologies for the formation of nanocomposites based on carbon nanotubes and nanoparticles of ferromagnetic materials are presented to increase the degree of coherence of the spin texture at macroscopic distances, increase the degree of security of electronic products to strong magnetic fields and noise immunity. It has been shown that at certain concentrations of ferromagnetic nanoparticles, micromagnetic parameters increase, ensuring high performance of active elements in magnetic fields. The role of carbon in this case turns out to be decisive. The presented results seem promising for use in the creation of anechoic chambers, partitions for separating indoor zones intended for the location of electronic devices, products for protecting humans from the effects of electromagnetic radiation in the microwave range, and the formation of combined masking signals in the form of “white” noise for devices protection of speech information, composite coatings to level out the effects of a constant magnetic field on electronic products.

Ice Identification with Error-Accumulation Enhanced Neural Dynamics in Optical Remote Sensing Images

Arctic sea ice plays an important role in Arctic-related research. Therefore, how to identify Arctic sea ice from remote sensing images with high quality in an unavoidable noise environment is an urgent challenge to be solved. In this paper, a constrained energy minimization (CEM) method is applied for Arctic sea ice identification, which only requires the target spectrum. Moreover, an error-accumulation enhanced neural dynamics (EAEND) model with strong noise immunity and high computing accuracy is proposed to aid with the CEM method for Arctic sea ice identification. With the theoretical analysis, the proposed EAEND model possesses a small steady-state error in noisy environments. Finally, compared with other existing models, the proposed EAEND model can not only complete sea ice identification in excellent fashion, but also has the advantages of high efficiency and noise immunity.

A novel roundness error evaluation method for high-speed EMU train axles

This article proposes a novel roundness error evaluation method for high-speed electric multiple units (EMU) train axles to evaluate the roundness error of the minimum zone circle. This method utilizes the adaptive ability of Bayesian linear regression to the data to solve the initial center of the circle, avoiding the multiple search process, which not only improves the data utilization and simplifies the solving steps but also has a strong noise immunity. The method first establishes a Hough space model, maps the collected data points to the Hough space, and forms different conical surfaces; then, by Bayesian linear regression, the probability density function is set with the distance from the space point to the conical surface as a parameter, and through continuous iteration, the point with the highest probability value is the more accurate initial center of the circle. Finally, the final optimal center of the circle is obtained by the principle of minimum inclusion region. We conducted experimental studies on axles of different diameters and compared them with other methods. The results prove the practicality and effectiveness of the technique.

Differential integrating sphere-based photoacoustic spectroscopy gas sensing.

In this Letter, a differential integrating sphere-based photoacoustic spectroscopy (PAS) gas sensor is proposed for the first time to our knowledge. The differential integrating sphere system consists of two integrating spheres and a tube. Based on differential characteristics, the photoacoustic signal of the designed differential integrating sphere was doubly enhanced and the noise was suppressed. Compared with a single channel integrating sphere, the differential integrating sphere sensing system had a 1.86 times improvement in signal level. An erbium-doped fiber amplifier (EDFA) was adopted to amplify the output of diode laser to enhance the optical excitation. The second harmonic (2f) signal of differential integrating sphere-based acetylene (C2H2) PAS sensor with an amplified 1000 mW optical output power was 104.67 mV, which was 22.80 times improved compared to the sensing system without EDFA. When the integration time was 100 s, the minimum detection limit (MDL) of the differential integrating sphere-based C2H2 PAS sensor was 416.7 ppb. The differential integrating sphere provides a new method, to the best of our knowledge, for the development of PAS sensor, which has the advantages of photoacoustic signal enhancement, strong noise immunity, and no need for optical adjustment.

Design of memristor hyperchaotic circuit with burst oscillation and infinite attractor coexistence and its application

A 4D hyperchaotic system with symmetry based on memristors is proposed and the chaotic properties of the circuit are verified by phase diagrams, Poincaré cross sections, bifurcation, Lyapunov exponents, and 0–1 tests. In addition, the system is found to have rich dynamical behaviors such as transient chaos, intermittent chaos, offset enhancement, burst oscillation, and infinite attractor coexistence. The complexity comparison analysis reveals that this hyperchaotic system has good symmetry as well as the initial values of the system that are more suitable for image encryption. Circuit simulation of the system using simulation software proved the usability of the system. A field programmable gate array (FPGA) is used to implement the actual digital circuit of the system. The final FPGA results agree with the numerical simulation and prove the practical value of the constructed memristor hyperchaotic system. Finally, chaotic circuits are applied to DNA image encryption. It is discovered to have a very large key space, strong key sensitivity, outstanding anti-cropping, noise immunity and high security performance. Thus, the proposed hyperchaotic system has good advantages for application in the field of image encryption.

MDGN: Circuit design of memristor‐based denoising autoencoder and gated recurrent unit network for lithium‐ion battery state of charge estimation

AbstractDue to the highly complex and non‐linear physical dynamics of lithium‐ion batteries, it is unfeasible to measure the state of charge (SOC) directly. Designing systems capable of accurate SOC estimation has become a key technology for battery management systems (BMS). Existing mainstream SOC estimation approaches still suffer from the limitations of low efficiency and high‐power consumption, owing to the great number of samples required for training. To address these gaps, this paper proposes a memristor‐based denoising autoencoder and gated recurrent unit network (MDGN) for fast and accurate SOC estimation of lithium‐ion batteries. Specifically, the DAE circuit module is designed to extract useful feature representation with strong generalization and noise immunity. Then, the gated recurrent unit (GRU) circuit module is designed to learn the long‐term dependencies between high‐dimensional input and output data. The overall performance is evaluated by root mean square error (RMSE) and mean absolute error (MAE) at 0, 25, and 45°C, respectively. Compared with the current state‐of‐the‐art methods, the entire scheme shows its superior performance in accuracy, robustness, and operation cost (referring to time cost).

An Improved MSCNN and GRU Model for Rolling Bearing Fault Diagnosis

In this paper, a novel fault diagnosis method based on the fusion of squeeze and excitation-multiscale convolutional neural networks (SENet-MSCNN) and gate recurrent unit (GRU) is proposed to address the problem of low diagnosis rate caused by the fact that normal samples are much larger than fault samples in the vibration big data. The method takes the time-domain vibration signal as input and fuses the spatial features extracted by SENet-MSCNN. The temporal features extracted by GRU in order to bring them into the fully connected layer for identification so as to realize the intelligent diagnosis of rolling bearing adaptive feature extraction. Finally, the method is applied to the simulated signal and experimental data for testing and analysis. The results reveal that the model can reach 98.98 % and 76.44 % migration diagnostic accuracy in bearing and gearbox datasets. At the same time, it has strong noise immunity, adaptivity, and robustness, providing an effective way for intelligent diagnosis of rolling bearing vibration big data.

Probability-Based Diagnostic Imaging of Fatigue Damage in Carbon Fiber Composites Using Sparse Representation of Lamb Waves

Carbon fiber composites are commonly used in aerospace and other fields due to their excellent properties, and fatigue damage will occur in the process of service. Damage imaging can be performed using damage probability imaging methods to obtain the fatigue damage condition of carbon fiber composites. At present, the damage factor commonly used in the damage probability imaging algorithm has low contrast and poor anti-noise performance, which leads to artifacts in the imaging and misjudgment of the damaged area. Therefore, this paper proposes a fatigue damage probability imaging method for carbon fiber composite materials based on the sparse representation of Lamb wave signals. Based on constructing the Lamb wave dictionary, a fast block sparse Bayesian learning algorithm is used to represent the Lamb wave signals sparsely, and the definition of Lamb wave sparse representing the damage factor calculates the damage probability of the monitoring area and then images the fatigue damage of the carbon fiber composite materials. The imaging research was carried out using the fatigue monitoring experiment data of NASA’s carbon fiber composite materials. The results show that the proposed damage factor can clearly distinguish the damaged area from the undamaged area and has strong noise immunity. Compared with the energy damage factor and the cross-correlation damage factor, the error percentages are reduced by at least 58.63%, 28.11%, and 8.43% for signal-to-noise ratios of 6 dB, 3 dB, and 0.1 dB, respectively, after adding noise to the signal. The results can more accurately reflect the real location and area of fatigue damage in carbon fiber composites.

Research on Fault Diagnosis of Rolling Bearing Based on SEMSCNN and GRU Model

Abstract In this paper, a deep neural network model fault diagnosis method based on the fusion of SENet-MSCNN and GRU is proposed to address the problem of low diagnosis rate caused by the fact that the normal samples are much larger than the fault samples in the collected rolling bearing vibration big data. The method takes the time-domain vibration signal as input and fuses the spatial features extracted by SE-MSCNN and the temporal features extracted by GRU to bring into the fully connected layer for identification, to realize the intelligent diagnosis of rolling bearing adaptive feature extraction. Finally, the method is applied to the simulated signal and experimental data for testing and analysis respectively. The results show that the model can reach 100% fault recognition rate under variable speed conditions, and at the same time has strong noise immunity, adaptivity and robustness. Thus, providing an effective new way for intelligent diagnosis of rolling bearing vibration big data.

Multiplication sampling moire method for full-field deformation measurement of composite materials.

A multiplication sampling moire (MSM) method was proposed for robust deformation distribution measurement by performing phase analysis of the second harmonic (second-order frequency) of a single grating pattern. The MSM method has a very strong noise immunity because the second harmonic spectrum is far from the low-frequency region of the background noise in the frequency domain. Phase analysis of an experimental grid image on a carbon fiber-reinforced plastic (CFRP) specimen indicated that the MSM method effectively solved the problem of non-negligible phase measurement errors of conventional methods that extract the fundamental frequency of the grating, in the case of severe local noise. The displacement and strain distributions of CFRP in a tensile test were successfully measured. This method is suitable for deformation measurement of various composite materials.

A fault diagnosis method for small pressurized water reactors based on long short-term memory networks

This paper proposes a sensor and actuator fault diagnosis method for small pressurized water reactors (SPWRs), with an innovative labeled fault dictionary established to map complex fault modes, using long short-term memory (LSTM) networks. It can directly learn features from multivariable time-series data and capture long-term dependencies through the cyclic behavior and gate mechanism of LSTM to realize the end-to-end fault diagnosis of SPWRs. Experimental results on a SPWR fault dataset show that the method can effectively diagnose the location, type, and extent of sensor and actuator faults from raw time-series signals with an average accuracy of 92.06% and outperforms three other widely-used fault diagnosis methods. Furthermore, the diagnosis results on the SPWR fault dataset injected with different noise signals demonstrate the strong noise immunity capability of the established LSTM network. Therefore, the proposed method is expected to achieve satisfactory fault diagnosis performances in actual operating environments of SPWRs.

Charging Load Pattern Extraction for Residential Electric Vehicles: A Training-Free Nonintrusive Method

Extracting the charging load pattern of residential electric vehicle (REV) will help grid operators make informed decisions in terms of scheduling and demand-side response management. Due to the multistate and high-frequency characteristics of integrated residential appliances from the residential perspective, it is difficult to achieve accurate extraction of the charging load pattern. To deal with that, this article presents a novel charging load extraction method based on residential smart meter data to noninvasively extract REV charging load pattern. The proposed algorithm harnesses the low-frequency characteristics of the charging load pattern and applies a two-stage decomposition technique to extract the characteristics of the charging load. The two-stage decomposition technique mainly includes: the trend component of the charging load being decomposed by seasonal and trend decomposition using loess method, and the low-frequency approximate component being decomposed by discrete wavelet technology. Furthermore, based on the extracted characteristics, event monitoring, and dynamic time warping is applied to estimate the closest charging interval and amplitude. The key features of the proposed algorithm include 1) significant improvement in extraction accuracy; 2) strong noise immunity; 3) online implementation of extraction. Experiments based on ground truth data validate the superiority of the proposed method compared to the existing ones.

A reconfigurable bidirectional associative memory network with memristor bridge

Bidirectional associative memories (BAMs) have been extensively applied in auto and heteroassociative learning. However, the research on the hardware implementation of BAM neural networks is relatively few. Memristor, a nanoscale synaptic-like element, provides a new perspective for the circuit implementation of neural networks. In this work, we improve a threshold memristor bridge circuit (T-MBC) and design a novel memristive bidirectional associative memory (MBAM) network circuit on this basis. T-MBC is capable to achieve positive, negative, and zero synaptic weights without any switches, inverters, transistors, and current–voltage conversion processes. The validity of T-MBC is illustrated by PSPICE. MBAM has a simpler structure and gains excellent effects in flexibility, and reconfigurability. Besides, it is shown that MBAM can be trained to recall both disturbed binary images and incomplete/noisy grey-scale images robustly. Meanwhile, a more complex application of emotion classification based on MBAM is also accomplished. Experimental results verify the effectiveness of the MBAM neural network and demonstrate that the MBAM neural network has high recognition accuracy and strong noise immunity.

Multiple Kernel Graph Cut for SAR Image Change Detection

Complementary information between two difference images (DI’s) has great contribution to improve change detection performances. Based on the effectiveness and flexibility of the multiple kernel learning (MKL) in information fusion, we develop a multiple kernel graph cut (MKGC) algorithm for synthetic aperture radar (SAR) image change detection. An energy function containing a weighted summation kernel is proposed for fusing the complementary information between the subtraction image and the ratio image. By iteratively minimizing the energy function, the kernel weights, region parameters and region labels are estimated automatically and optimally. Besides of it avoids modeling, MKGC also has a complete description of the changed areas and the strong noise immunity. Experiments on real GaoFen-3 SAR data set demonstrate the effectiveness of the MKGC algorithm, and illustrate that it is a good candidate for SAR image change detection.

Wide-view and accurate deformation measurement at microscales by phase extraction of scanning moiré pattern with a spatial phase-shifting technique.

A scanning-based second-order moiré method is proposed for high-accuracy deformation measurement in a large field of view (FOV) by analyzing the phase distribution of a single-shot scanning moiré fringe image using a spatial phase-shifting technique. In this method, the grating pitch can be as small as around one pixel in the scanning moiré image to ensure a wide FOV, while high-precision phase measurement is achievable. The strain measurement accuracy has been verified from simulations at different grating pitches, applied strains, and noise levels. The simulation results show that the closer the grating pitch is to the scanning pitch, the smaller the strain measurement error, and the recommended pitch ratio is 0.9∼1.1. Furthermore, the feasibility of this method has been verified from a tensile experiment on an aluminum specimen under a laser scanning microscope with scanning moiré images recorded. The microscale strains of aluminum measured at different tensile loads agree well with the strain gauge results. As an integration of the scanning and sampling moiré methods, this method has the advantages of a large FOV, high accuracy, strong noise immunity, and visualization of magnified deformation. Compared with the traditional phase-shifting scanning moiré method, this method only needs to record a single scanning moiré image and is suitable for dynamic deformation analysis.

Use of Bispectrum Analysis to Inspect the Non-Linear Dynamic Characteristics of Beam-Type Structures Containing a Breathing Crack.

A breathing crack is a typical form of structural damage attributed to long-term dynamic loads acting on engineering structures. Traditional linear damage identification methods suffer from the loss of valuable information when structural responses are essentially non-linear. To deal with this issue, bispectrum analysis is employed to study the non-linear dynamic characteristics of a beam structure containing a breathing crack, from the perspective of numerical simulation and experimental validation. A finite element model of a cantilever beam is built with contact elements to simulate a breathing crack. The effects of crack depth and location, excitation frequency and magnitude, and measurement noise on the non-linear behavior of the beam are studied systematically. The result demonstrates that bispectral analysis can effectively identify non-linear damage in different states with strong noise immunity. Compared with existing methods, the bispectral non-linear analysis can efficiently extract non-linear features of a breathing crack, and it can overcome the limitations of existing linear damage detection methods used for non-linear damage detection. This study’s outcome provides a theoretical basis and a paradigm for damage identification in cracked structures.

Design and optimization of photoacoustic CO gas sensor for fault diagnosis of SF<sub>6</sub> gas insulated equipment

Trace gas analysis for SF<sub>6</sub> decomposition is a powerful diagnostic method to identify partial discharge problem occurring in electrical equipment. In particular, it is recognized that the SF<sub>6</sub> decomposition gases (such as CO, H<sub>2</sub>S, SO<sub>2</sub> and CF<sub>4</sub>) can effectively determine the inner insulation condition of the electrical equipment. Currently, most of researches of diagnostic methods cannot meet the online high-precision detection of gas derivatives in SF<sub>6</sub> electrical insulation equipment. Therefore, there is a need of developing a sensitive, selective and cost-effective sensor system for CO detection in an SF<sub>6</sub> buffer gas environment due to the fact that the power system is filled with pure SF<sub>6</sub> as the dielectric gas, which means that the concentration of SF<sub>6</sub> is usually > 99.8%. The traditional photoacoustic CO gas sensors cannot be directly used in power system, since several SF<sub>6</sub> physical constants strongly differ from those of N<sub>2</sub> or air. In addition, SF<sub>6</sub> molecule reveals uninterrupted and strong absorption lines in the mid-infrared spectral region. In this work, a CO gas sensor working in high concentration SF<sub>6</sub> background gas is designed by using a distributed feedback (DFB) laser as an excitation source with a center wavelength of 2.3 μm. The absorption line strength of 2.3 μm is ~ two orders of magnitude higher than the absorption line strength around 1.56 μm, which can improve the sensor detection performance. A background-gas-induced high-<i>Q</i> differential photoacoustic cell is simulated numerically and tested experimentally. The quality factor of the designed photoacoustic cell in pure SF<sub>6</sub> gas is 85, which is ~ 4 times higher than that in N<sub>2</sub> carrier gas. The experimental results show that the maximum gas flow rate of the differential structure photoacoustic cell is ~ 6 times higher than that of the single resonant cavity photoacoustic cell. After optimizing the resonance frequency, gas velocity and working pressure of the sensor system, the detection sensitivity of the volume fraction of 1.85 × 10<sup>–6</sup> is achieved. In the case of the volume fraction of 50 × 10<sup>–6</sup> CO/SF<sub>6</sub> gas mixture, the maximum photoacoustic signal amplitude of 19.6 μV is obtained, the corresponding normalized noise equivalent concentration (1σ) is 3.68 × 10<sup>–8</sup> cm<sup>–1</sup>·W·Hz<sup>1/2</sup> in 1 s integration time. A linear fitting is implemented to evaluate the response of the sensor from 50 × 10<sup>–6</sup> to 1000 × 10<sup>–6</sup>, resulting in an <i>R</i> square value of 0.9997. The CO photoacoustic gas sensor has high sensitivity, good selectivity and strong noise immunity, which can provide an on-line detection technology for potential insulation fault diagnosis in the power system. The capability of CO gas sensor can be improved by using a higher excitation optical output power and/or reducing the PAC resonator volume to increase the cell constant.

A multiscale reconstructed attractors-based method for identification of structural damage under impact excitations

Vibrational responses-based damage identification has been widely investigated in recent decades. Different from existing studies that process vibrational responses in time, frequency, or time-frequency domains, a newly emerged reconstructed attractor-based damage identification is becoming a novel strategy for damage identification from vibrational responses. In this aspect, reported studies to date demonstrate the dependence of the efficacy of this method on the tuned chaotic excitation quantified by the Lyapunov exponents. Nevertheless, the tuned chaotic excitation is not always actualized for a complicated engineering structure in reality, seriously hindering the application of the method. To conquer this problem, this study retrofits the conventional reconstructed attractor by introducing multiscale analysis offered by the stationary wavelet transform into phase space reconstruction of attractors. Such a retrofit leads to a novel quantity of multiscale reconstructed attractors (MRAs) that consist of a set of attractors reconstructed on individual scales of vibrational responses. The MRAs characterize structural damage from the compound variations of a set of attractors topological trajectories, termed the MRAs-based prediction error. Compared with the conventional reconstructed attractor, the MRAs-based prediction error features detecting structural damage with great accuracy and strong noise immunity, and especially no need for tuned chaotic excitation. The feasibility of the MRAs in damage identification are numerically verified and its effectiveness are experimentally validated on cantilever beams. The proposed method expands the applications of reconstructed attractors from vibrational responses on structural health monitoring.