Large Parameter Variations Research Articles

Envelope spectrum neural network with adaptive domain weight harmonization for intelligent bearing fault diagnosis under cross-machine scenarios

Accurate bearing fault diagnosis technology is highly important for ensuring the safe operation of mechanical equipment. Fault diagnosis methods can be roughly divided into signal processing-based methods (SPM) and data-driven methods (DDMs), which rely on physical knowledge and data knowledge, respectively. However, SPM cannot adapt to large data samples and dynamic parameter variations. DDMs did not consider physical representation consistency. To address the above issues, an envelope spectrum neural network (ESNN) is proposed for cross-machine bearing fault diagnosis. First, a deep transfer convolution network (DTCN) is constructed as the basic diagnostic framework. Second, a knowledge-to-model conversion strategy (KMC) is built to create ESNN, which utilizes equivalent convolution functions to construct the first two layers of DTCN, guiding the model to learn physical knowledge. Subsequently, an adaptive domain weight harmonization (ADWH) mechanism is proposed that can dynamically combine marginal distributions and joint distributions and automatically learn hidden features contained in physical and data knowledge, thus alleviating domain shift issues. Experimental evaluations are conducted using fault datasets from three different machines. The results show that, compared to models without knowledge guidance, ESNN achieves a 6.4% improvement in diagnostic accuracy. Compared to many advanced models, ESNN can achieve a maximum diagnostic accuracy improvement of 35.94%. The code of the ESNN can be found at https://github.com/John-520/ESNN.

PV Panel Model Parameter Estimation by Using Particle Swarm Optimization and Artificial Neural Network.

Photovoltaic (PV) panels are one of the popular green energy resources and PV panel parameter estimations are one of the popular research topics in PV panel technology. The PV panel parameters could be used for PV panel health monitoring and fault diagnosis. Recently, a PV panel parameters estimation method based in neural network and numerical current predictor methods has been developed. However, in order to further improve the estimation accuracies, a new approach of PV panel parameter estimation is proposed in this paper. The output current and voltage dynamic responses of a PV panel are measured, and the time series of the I-V vectors will be used as input to an artificial neural network (ANN)-based PV model parameter range classifier (MPRC). The MPRC is trained using an I-V dataset with large variations in PV model parameters. The results of MPRC are used to preset the initial particles' population for a particle swarm optimization (PSO) algorithm. The PSO algorithm is used to estimate the PV panel parameters and the results could be used for PV panel health monitoring and the derivation of maximum power point tracking (MMPT). Simulations results based on an experimental I-V dataset and an I-V dataset generated by simulation show that the proposed algorithms can achieve up to 3.5% accuracy and the speed of convergence was significantly improved as compared to a purely PSO approach.

Prediction of the undrained base heave and post uplift in braced excavation of a typical subway station in thick soft clay

Construction of a typical subway station usually involves a narrow long excavation with a width of approximately 20 m. An undrained base heave is a common problem in deep excavation areas with soft clay, which may induce considerable post uplift and damage the safety of the retaining system. Post uplift involves complex interactions between soil, pile post foundation, and supported struts, and an accurate prediction of post uplift is challenging. This paper aimed to establish a practical method for predicting the base heave and post uplift in the construction of typical subway stations in areas with thick soft clay. Series soil–water coupled finite element method (FEM) analysis was conducted to investigate the characteristics of the base heave with different soil properties and wall stiffnesses. The results showed that the calculated base heave can be fitted by a normalized function, regardless of large variations in soil properties and wall design parameters. The normalized function was then applied to a soil–post–strut interaction analysis model to predict the post uplift at different excavation depths. The application of the proposed method to a real subway station project showed that the predicted results were in good agreement with the field observations.

Influence of environmental conditions on mass rearing parameters of tsetse flies at the Bobo-Dioulasso insectary (Burkina Faso): Retrospective study

With the goal of eradicating tsetse flies and trypanosomiasis in Africa, several control methods have been developed. One of this is the biological control through the application of the sterile insect technique (SIT) as a part of an area-wide integrated pest management. This method required the mass production of sterile males with a high competitiveness. The success of the tsetse mass rearing is strongly linked to the rearing conditions. The objective of this study was to do a retrospective evaluation of the impact of ambient conditions on the production performances of Glossina palpalis gambiensis at Insectary of Bobo Dioulasso. Data of productivity, fecundity, adult emergence and environmental conditions (temperature and relative humidity) of the rearing rooms were recorded from March to June 2020 and were analysed. The results showed a reduction of the normal pupae produced and the increase of soft pupae during this period. In addition, the relative humidity has shown a significant positive correlation with the production of soft pupae, whereas it had a significant negative correlation with the number of pupae produced per female per 10 days. However, the temperature variation during the period of the data recording didn’t have any impact on the production parameters. Rearing rooms need a better management to avoid large variations in environmental parameters.

Microphysical characteristics of black carbon from various emission sources

The complex microphysical characteristics of black carbon aerosol (BC) results in large uncertainties in the evaluation of BC's climatic and environmental effects. In this study, the microphysical properties were investigated in fresh BC emitted from traffic sources (gasoline vehicle and diesel) and aged BC emitted from solid fuel burning sources (coal, cotton stem, rice stem and mulberry). For the BC core size distribution, the number size distribution of traffic sources peaks at a lower value than that of aged solid fuel burning sources. Although traffic sources emit a greater number of small BC cores, traffic sources could also emit BC aerosols with large BC cores (diameter larger than 400 nm) not found in solid fuel sources. The coating thickness of BC from traffic sources is significantly lower than that of aged BC from solid fuel sources. Such differences in size distribution and coating thickness cause large variations in other microphysical parameters, such as light absorption and hygroscopicity, for these two types of BC. The mass absorption cross section (MAC) of aged BC from cotton stems reached 11.69 m2/g, nearly 2 times higher than the MAC (6.08 m2/g) of fresh BC from gasoline vehicles. The hygroscopicity of aged BC from solid fuel burning is also much higher than that of fresh BC from traffic. The critical supersaturation needed to activate BC to become cloud droplets is 0.03% for the former and 0.42% for the latter. This study reports the BC emission characteristics of traffic sources and emphasizes the large property difference between fresh traffic emission BC and aged solid fuel burning BC. The microphysical properties of BC should be carefully considered to better resolve BC concentrations and optical properties in climatic models.

Power System Stabilizer Design using Genetic Algorithms and Particle Swarm Optimization

in this paper, Meta-heuristic techniques using genetic algorithms GA and particle swarm optimization PSO to tuning optimal design of power system stabilizer PSS proposed. This latter have been used for many years to add damping to electromechanical oscillations of power system, Based on this idea we have proposed multiobjective function composed with tow function, first maximize stability margin by increasing the damping factors while minimizing the real parts of the eigenvalues. Simulation results to comparative study between genetic algorithms and particle swarm optimization obtained by our realized graphical user interface (GUI) proved the efficiency of PSS optimized by genetic algorithms in comparison with Particle Swarm Optimization, showing stable system responses almost insensitive to large parameter variations and under different operating regime (under-excited, nominal and over excited regime).

Instabilities in the wake of an isolated cylindrical roughness element

The instability mechanism behind a geometrically simple cylindrical roughness element continues to be a challenging topic in fluid mechanics. Considerable progress has been made in understanding the phenomena in recent years, but more research is needed to predict the temporal nature and spatial structure of the dominant instability in a given flow configuration. This is of particular interest, as these instabilities dictate the transition to turbulence and thus are significant for large-scale effects such as skin friction drag. A smoke-flow visualization study with a large variation of parameters, featuring a cylindrical roughness element connected to a linear traverse, has been performed. Results show good agreement with previous investigations and provide further insights into the stability properties, revealing several unexpected effects. For a low roughness aspect ratio $\eta$ , no global instability is detected even at the highest roughness Reynolds number $Re_{kk}$ , whereas a high aspect ratio indicates a delay in the onset of instability. From the acquired visualizations, we constructed the, so far, richest instability diagram of the wake behind an isolated roughness element in the $Re_{kk}\unicode{x2013}\eta$ space, sampled in the same measurement campaign. Furthermore, information regarding the dominant frequency in the wake can be extracted from the visualization images. Our results suggest a new scaling of the frequency as the velocity is increased. Finally, it is shown that the dominant frequency in a certain flow regime can be well predicted using a Strouhal number based on the cylinder diameter and the roughness velocity.

Computation of parameter dependent robust invariant sets for LPV models with guaranteed performance

This paper presents an iterative algorithm to compute a Robust Control Invariant (RCI) set, along with an invariance-inducing control law, for Linear Parameter-Varying (LPV) systems. As real-time measurements of the scheduling parameters are typically available, we allow the RCI set description and the invariance-inducing controller to be scheduling parameter dependent. Thus, the considered formulation leads to parameter-dependent conditions for the set invariance, which are replaced by sufficient Linear Matrix Inequalities (LMIs) via Polya’s relaxation. These LMI conditions are then combined with a novel volume maximization approach in a Semidefinite Programming (SDP) problem, which aims at computing the desirably large RCI set. Besides ensuring invariance, it is also possible to guarantee performance within the RCI set by imposing a chosen quadratic performance level as an additional constraint in the SDP problem. Using numerical examples, we show that the presented iterative algorithm can generate RCI sets for large parameter variations where commonly used robust approaches fail.

Enhanced Torque Estimation in Variable Leakage Flux PMSM Combining High and Low Frequency Signal Injection

Torque measurement/estimation in Variable Leakage Flux Permanent Magnet Synchronous Motors (VLF-PMSMs) is preferred for accurate control. Torque measurement systems are expensive, require extra room, add weight, can introduce resonances into the system and can be sensitive to electromagnetic interference. Alternatively, torque can be estimated, accurate knowledge of machine parameters being often required in this case, which for the case of VLF-PMSMs can be a serious drawback due to the large variation of machine parameters with the operating condition. This paper proposes a torque estimation method for VLF-PMSMs based on stator flux linkage estimation. Machine parameters involved in stator flux linkage estimation are obtained from the response of the machine to a pulsating high frequency current signal and a low-frequency and small-magnitude, square-wave current signal. Both signals are injected on top of the fundamental excitation without interfering with the normal operation of the machine.

Complex-Variable Sliding-Mode Control of Instantaneous Complex Energy and Power for Grid-Tied Inverter*

A complex-variable sliding-mode control (SMC) algorithm is proposed to govern inverters interfacing renewable energy sources (RESs) with the electrical grid. It is conceived to control the instantaneous energy stored in the passive components of the system and its rate of change, as well as the instantaneous reactive energy and power exchanged with the grid. The stability of the resulting closed-loop system is analyzed, and tuning of the designed SMC algorithm is also tackled. In addition, the design and tuning of a nonlinear observer estimating the renewable power supplied to the DC link are addressed. The overall control scheme obtained by combining the proposed complex-variable SMC algorithm with such observer is assessed in simulation, demonstrating its outstanding tracking performance and high robustness in presence of large parameter variations.

Ultra-broadband and highly efficient silicon nitride bi-layer grating couplers

The silicon nitride (Si3N4) waveguide platform becomes prominent in silicon photonics due to its ultralow propagation loss, negligible two-photon absorption, tolerance to phase error, and so on. Particularly, Si3N4 photonic circuits provide nonlinear opportunities as thick waveguides enable dispersion engineering, which is fundamental for frequency comb and supercontinuum generation. However, grating couplers based on thick Si3N4 waveguides, as the building block for any device, are prone to higher-order Bloch mode excitation, restricting the coupling efficiency dramatically, thus precluding practical applications that large bandwidth and are power restricted. We propose and demonstrate an 800 nm-thick Si3N4bi-layer fully etched grating coupler which is ultra-broadband and highly efficient. By applying the apodization technique, the bi-layer grating shows a conversion efficiency of −1.44 dB and a 1 dB bandwidth of 135 nm. Further integrated with a bottom reflector, a conversion efficiency of −1.01 dB with a 1-dB bandwidth of 117 nm is achieved. The proposed structures are highly fabrication tolerant and remain great performance regardless large structural parameter variations. It paves the way for a large-volume high-yield production towards fully-integrated frequency comb chip, optical beam forming network, Lidar, etc.

Meta-analysis of genetic parameters for growth traits in meat, wool and dual-purpose sheep breeds in the world using a random-effects model.

There is large variation in genetic parameters in literature for growth traits in sheep. Reliable estimation of genetic parameters is required for developing breeding programmes. The aim of this study was to aggregate results of different studies by meta-analysis to improve reliability of estimated parameters. In the current study, 221 papers that have been published between 1995 and 2021 were reviewed. Using a random-effects model in the Comprehensive Meta-Analysis software, direct and maternal heritabilities, as well as, genetic and phenotypic correlations between growth traits were estimated in meat (M), wool (W) and dual-purpose (D) sheep breeds. The growth traits in this study were birth weight, 3-month weight, 6-month weight, 9-month weight and yearling weight. The combined direct heritability was the lowest for birth weight (0.190 ± 0.004, 0.198 ± 0.003 and 0.196 ± 0.004 for M, W and D breeds, respectively) and the highest for yearling weight (0.264 ± 0.010, 0.304 ± 0.005 and 0.285 ± 0.020 for M, W and D breeds, respectively). The maternal heritability was the lowest for yearling weight (0.085 ± 0.003, 0.055 ± 0.002 and 0.052 ± 0.005 for M, W and D breeds, respectively) and the highest for 6-month weight (0.240 ± 0.088, 0.164 ± 0.001 and 0.162 ± 0.006 for M, W and D breeds, respectively). The phenotypic and genetic correlations were lower between the weights measured at more distant intervals. The lowest genetic correlation was observed between birth weight and yearling weight (0.290 ± 0.051 for W breeds). The small standard errors could indicate that the aggregation of results from different studies improved the reliability of estimated parameters and reduced range of 95% confidence intervals. Hence, the results could be used with greater level of confidence in sheep breeding programmes.

Influence of Nb addition and process parameters on the microstructure and phase transformation behavior of NiTiNb ternary shape memory alloys fabricated by laser powder bed fusion

NiTiNbx (x=1,3,5) alloys were fabricated by laser powder bed fusion (L-PBF) of powder mixtures containing pre-alloyed NiTi and elemental Nb powders. Upon Nb addition, the dependence of the martensite transformation temperatures (MTTs) on the L-PBF process parameters was significantly altered. In case of the NiTiNb1 alloy, the MTTs increased as the laser energy density was increased, whereas it decreased in case of the NiTiNb5 alloy. It is suggested that the net effect of Nb addition is to lower the MTTs, while Ni evaporation has the opposite effect. The latter contribution is predominant in alloys with low Nb addition (1 at.%), while the Nb-related effects become the major influencing factor in the high-Nb alloys (5 at.%). In the case of the NiTiNb3 alloys, the factors that promote and suppress MTTs are almost completely compensated, and the MTT is rather stable despite of the large variation of L-PBF process parameters.

Nanotwinned (inter)martensite transformation interfaces in Ni50Mn25Ga20Fe5 magnetic shape memory single crystal foil

Using common and high resolution transmission electron microscopy (TEM), we study the magnetic and crystal phase structures as well as their evolution in Ni50Mn25Ga20Fe5 magnetic shape memory material with high Curie point. The particular alloying by Fe and changing thickness of the TEM foil enable us to observe all martensitic phases known in Ni-Mn-Ga Heusler alloy system and their respective transitions simultaneously. Starting from cubic austenite at about 10 nm foil thickness, the structure evolves via peculiar interleaved stripes of austenite and five-layered modulated 10M martensite to pure 10M phase with a low density of stacking faults at 40 nm thickness. With further increasing thickness, the 10M phase transforms gradually to seven-layered modulated 14M martensite with an increased density of stacking faults. Finally, the non-modulated tetragonal NM phase appears within the 14M phase by detwinning of nanotwins forming the modulated phases. High resolution TEM further confirms that nanotwinning and stacking faults are inherent structure features tightly connected with the lattice modulation and intermartensite transformations. Overall, the evolution of average lattice shows the same general trend, an increasing simple shear with (11¯0) shuffling plane, across the whole phase sequence. Additionally, we found large local variation of lattice parameters in all phases, which is is ascribed to strong lattice softening in the vicinity of the martensitic transformation and high density of stacking faults in 14M martensite lattice.

Arterial input functions in dynamic susceptibility contrast MRI (DSC-MRI) in longitudinal evaluation of brain metastases.

Dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) could be helpful to separate true disease progression from pseudo-progression in brain metastases when assessing the need for retreatment. However, the selection of arterial input functions (AIFs) is not standardized for analysis, limiting its use for this application. To compare population-based AIFs, AIFs specific to each patient, and AIFs specific to every visit in the longitudinal follow-up of brain metastases. Longitudinal data were collected from eight patients before treatment (6 of 8 patients) and after treatment (6-17 visits). Imaging was performed using a 1.5-T MRI system. Lesions were segmented by subtracting precontrast images from postcontrast images. Cerebral blood volume (rCBV) and cerebral blood flow (rCBF) were computed, and Pearson's product moment correlation coefficients were calculated to evaluate similarity of DSC parameters dependent on various AIF choices across time. AIF shape characteristics were compared. Parameter differences between white matter (WM) and gray matter (GM) were obtained to determine which AIF choice maximizes tissue differentiation. Although DSC parameters follow similar patterns in time, the various AIF selections cause large parameter variations with relative standard deviations of up to ±60%. AIFs sampled in one patient across sessions more similar in shape than AIFs sampled across patients. Estimates of rCBV based on scan-specific AIFs differentiated better between perfusion in WM and GM than patient-specific or population-based AIFs (P ≤ 0.02). Results indicate that scan-specific AIFs are the best choice for DSC-MRI parameter estimations in the longitudinal follow-up of brain metastases.

From microwave gas sensor conditioning to ammonia concentration prediction by machine learning

This paper proposes an innovative approach to understand the conditioning process of a microwave gas sensor operating at room temperature based on the combination of its response and the interpretation of the mass spectrometer data. A large variation of the dielectric parameters and thus of the microwave response is due to water departure from the sensor surface. Consequently, the first step of the conditioning process is a carrier gas sweep of the sensor surface. The second step consists in pre-saturating the microwave gas sensor surface with a high concentration of the polluting gas which will be detected (here ammonia). This process results in a very good quality microwave response on the qualitative and quantitative aspects for the detection of ammonia in the air and is helping the work carried out in this article in artificial intelligence on microwave responses. A regressor machine learning model is applied on time samples of this sensor response to predict the ammonia concentration. Several machine learning algorithms are tested and compared. Principal Component Analysis is also tested to reduce the input data dimension, but results are not conclusive. The concentration profile is revised to reduce the bias induce by the presence of too much measurement data when no pollutant is present in the air. And the Mutlti-Layer Perceptron regressor give the best results with a mean absolute error of 32.13 ppm (8 %) over a range of 0–400 ppm and R-squared score of 0.87.

Evolution of transformation behavior and tensile functional properties with process parameters for electron beam wire-feed additive manufactured NiTi shape memory alloys

Electron beam wire-feed additive manufacturing (EBAM) has been shown to be an ideal method to additively manufacture NiTi alloys with good tensile functional properties. However, the influences of process parameters on the microstructure and thermomechanical response of EBAM processed NiTi alloys have not been revealed adequately up to now. In this study, the evolution of formation quality, microstructure, phase transformation behavior, and functional properties with the variation of process parameters were comprehensively studied for the NiTi alloys fabricated by EBAM. The results show that the process parameters may significantly affect the surface formation quality of the as-built NiTi alloy parts. All samples exhibit continuous and coarse columnar grains in the building direction in spite of the large variation of EBAM process parameters. One-stage A↔M transformation behavior was found for all as-built samples, while the phase transformation temperatures vary for different process parameters. The change in phase transformation temperature mainly results from different amount of Ni-evaporation, on which the beam power has the most significant effect. Most of the samples, which have low transformation temperatures, exhibit good tensile superelasticity after post heat treatment, while the sample with high transformation temperature shows good shape memory effect at as-built state. This study shows that by regulating process parameters, the EBAM can be used to fabricate NiTi alloys with different thermomechanical responses according to the demands of practical applications, while no need to change the raw material.

Study and Application of the Advanced Frequency Control Techniques H∞ in the Voltage Automatic Regulator of Powerful Synchronous Generators (Application under Gui/Matlab)

This article presents a study of of the advanced frequency control techniques based on loop-shaping H∞ optimization method applied on automatic excitation control of powerful synchronous generators (AVR and PSS), to improve transient stability and its robustness of a single machine- infinite bus system (SMIB).The computer simulation results (static and dynamic stability), with test of robustness against machine parameters uncertainty (electric and mechanic), have proved that good dynamic performances, showing a stable system responses almost insensitive to large parameters variations, and more robustness using the robust H-infinity controller, in comparison with the classical Russian PID power system stabilizer . Our present study was performed using a GUI realized under MATLAB in our work.

Robust active disturbance rejection control for systems with internal uncertainties: Multirotor UAV application

AbstractActive Disturbance Rejection Control (ADRC) has recently stood out as a viable alternative to the proportional–integral–derivative controllers. An interesting field of application of this approach is the control of multirotor unmanned aerial vehicles (UAVs) which are inevitably subject to various force and torque disturbances. What makes ADRC attractive is the enhanced trajectory tracking and disturbance rejection capabilities that it allows while requiring minimal knowledge about the system. Although in theory, large uncertainties on the few required system parameters do not affect the stability of the ADRC's closed‐loop, they cause performance deterioration and can lead to dangerous oscillations that result in instability in practice. In this study, we design a robust ADRC for multirotor UAVs that allows maintaining the performance despite large parameter variations, without changing the initial control gains tuning. Simulation and experimental results support the theoretical findings.

Multimodal transistors as ReLU activation functions in physical neural network classifiers

Artificial neural networks (ANNs) providing sophisticated, power-efficient classification are finding their way into thin-film electronics. Thin-film technologies require robust, layout-efficient devices with facile manufacturability. Here, we show how the multimodal transistor’s (MMT’s) transfer characteristic, with linear dependence in saturation, replicates the rectified linear unit (ReLU) activation function of convolutional ANNs (CNNs). Using MATLAB, we evaluate CNN performance using systematically distorted ReLU functions, then substitute measured and simulated MMT transfer characteristics as proxies for ReLU. High classification accuracy is maintained, despite large variations in geometrical and electrical parameters, as CNNs use the same activation functions for training and classification.