Robust algorithms for multiphase regression models

This paper proposes a robust procedure for solving multiphase regression problems that is efficient enough to deal with data contaminated by atypical observations due to measurement errors or those drawn from heavy-tailed distributions. Incorporating the expectation and maximization algorithm with the M-estimation technique, we simultaneously derive robust estimates of the change-points and regression parameters, yet as the proposed method is still not resistant to high leverage outliers we further suggest a modified version by first moderately trimming those outliers and then implementing the new procedure for the trimmed data. This study sets up two robust algorithms using the Huber loss function and Tukey's biweight function to respectively replace the least squares criterion in the normality-based expectation and maximization algorithm, illustrating the effectiveness and superiority of the proposed algorithms through extensive simulations and sensitivity analyses. Experimental results show the ability of the proposed method to withstand outliers and heavy-tailed distributions. Moreover, as resistance to high leverage outliers is particularly important due to their devastating effect on fitting a regression model to data, various real-world applications show the practicability of this approach.     

Dispersion of elastic waves in a micropolar metamaterial plate with periodical arranged resonators

The wave propagation in a micropolar elastic metamaterial is investigated in this paper. The elastic metamaterial is composed of the micropolar elastic host material and the periodically arranged local resonators. Compared with the classical elastic metamaterial, the micropolar elastic metamaterial has more material parameters that can be elaborately designed to manipulate the elastic wave propagation. By introducing additional displacement fields, a multi-displacement continuum model of the micropolar elastic metamaterial is presented to characterize the resonance behavior of the resonators and the microstructure effects of the unit cell. According to this continuum model, two independent wave systems exist: one is a longitudinal system and the other is a shear and rotation coupled transversal system. The dispersive curves and band gaps of the longitudinal and transversal systems are numerically discussed and the influences of the resonators are mainly considered.     

Assessment of pressure drop in conical spouted beds of biomass by artificial neural networks and comparison with empirical correlations

Pressure drop is an essential parameter in the operation of conical spouted beds (CSB) and depends on its geometric factors and materials used. Irregular materials, like biomass, are complex to treat and, unlike other gas–solid contact methods, CSB turn out to be a suitable technology for their treatment. Artificial neural networks were used in this study for the prediction of operating and peak pressure drops, and their performance has been compared with that of empirical correlations reported in the literature. Accordingly, a multi-layer perceptron network with backward propagation was used due to its ability to model non-linear multivariate systems. The fitting of the experimental data of both operating and peak pressure drop was significantly better than those reported in the literature, specifically in the case of the peak pressure drop, with R² being 0.92. Therefore, artificial neural networks have been proven suitable for the prediction of pressure drop in CSB.     

An adaptive vibrational resonance method based on cascaded varying stable-state nonlinear systems and its application in rotating machine fault detection

A weak character signal with low frequency can be detected based on the mechanism of vibrational resonance (VR). The detection performance of VR is determined by the synergy of a weak low-frequency input signal, an injected high-frequency sinusoidal interference and the nonlinear system(s). In engineering applications, there are many weak fault signals with high character frequencies. These fault signals are usually submerged in strong background noise. To detect these weak signals, an adaptive detection method for a weak high-frequency fault signal is proposed in this paper. This method is based on the mechanics of VR and cascaded varying stable-state nonlinear systems (VSSNSs). Partial background noise with high frequency is regarded as a special type of high-frequency interference and an energy source that protrudes a weak fault signal. In this way, high-frequency background noise is utilized in a VSSNS. To improve the detection ability, manually generated high-frequency interference is injected into another VSSNS. The VSSNS can be transformed into a monostable state, bistable state or tristable state by tuning the system parameters. The proposed method is validated by a simulation signal and industrial applications. The results show the effectiveness of the proposed method to detect a weak high-frequency character signal in engineering problems.

     

Profiling of thermally aged EPDM seals using portable NMR,indenter measurements and IR spectroscopy facilitating separation of different deterioration mechanisms

The changes occurring in EPDM cable transit seals during thermal ageing and the causes of these changes were investigated. Samples were aged at a temperature of 170 °C, and subsequently evaluated with respect to the distance from the surface with modulus profiling, infrared (IR) spectroscopy and nuclear magnetic resonance (NMR) spectroscopy, based on the extractable mass fraction profiles for initial and aged materials. The ageing resulted in an increase in the modulus and in the degree of oxidation and in a decrease in the NMR transverse relaxation time, T₂. The NMR data were obtained in a non-invasive manner by ex situ experiments performed with a portable low-field spectrometer (NMR MOUSE). The results showed deterioration processes that can be attributed to different mechanisms i.e. oxidation, anaerobic crosslinking and migration of oil extender. The unique combination of parameter profiles made it possible to resolve and quantify these three contributing mechanisms. The NMR results highlight the potential of this method for on-site testing.     

Transport properties of the antiferromagnetic systems with fluctuating valence

We present the calculation of the DC resistivity (conductivity) for the antiferromagnetic, two-band, extended s–f model. The influence on the finite bandwidth of the narrow 4f (5f) band on the transport properties of the model is investigated. We notice that the increase of the 4f (5f) bandwidth destroys the antiferromagnetic order and the system becomes paramagnetic in all temperatures. A systematic review of the DC resistivity (conductivity) is presented in the form of the 3D plots including different average occupation numbers of electrons per site (n=0.5, 0.75, 1, 1.25, 1.5, 1.75, 2), different relative positions of the 4f (5f) band and different 4f (5f) bandwidths. The calculated temperature dependence of the DC resistivity (conductivity) shows great similarities to the experimental results for many intermediate-valence rare-earth and actinide-based materials.     

Discrete element method study of parameter optimization and particle mixing behaviour in a soil mixer

Soil mixing is an emerging research in the field of construction resource recovery. In this study, the mixing behaviour of soil particles in a mixer is numerically simulated by the discrete element method (DEM). A four-factor, three-level orthogonal experiment is designed to optimize the mixer design by selecting the fly-cutter speed, spindle speed, number of blades and fly-cutter diameter, using Lacey mixing index and power consumption as evaluation indicators. Then, the impact of soil cohesion and type on the mixing behaviour is investigated. The results show that the optimal parameter combination of this experiment is 280 rpm fly-cutter speed, 40 rpm spindle speed, 4 blades and 250 mm fly-cutter diameter. This optimal combination reaches a comparatively uniform state mix in 5.9 s with an average power consumption of 704.11 W. In addition, the wear and tear of the mixer increases as soil cohesion increases, while the mixing quality of materials declines, resulting in a “shaft hugging” phenomenon. The mixing efficiency varies greatly among different soil types, but the radial and tangential velocities have a similar law. This work can provide some guidance for the optimization design of a mixer and study of soil mixing.     

Particle behavior in a turbulent pipe flow with a flat bed

Particle behavior in a turbulent flow in a circular pipe with a bed height h = 0.5R is studied at Re_b = 40,000 and for two sizes of particles (5 μm and 50 μm) using large eddy simulation, one-way coupled with a Lagrangian particle tracking technique. Turbulent secondary flows are found within the pipe, with the curved upper wall affecting the secondary flow formation giving rise to a pair of large upper vortices above two smaller vortices close to the pipe floor. The behavior of the two sizes of particle is found to be quite different. The 50 μm particles deposit forming irregular elongated particle streaks close to the pipe floor, particularly at the center of the flow and the pipe corners due to the impact of the secondary flows. The deposition and resuspension rate of the 5 μm particles is high near the center of the floor and at the pipe corners, while values for the 50 μm particles are greatest near the corners. Near the curved upper wall of the pipe, the deposition rate of the 5 μm particles increases in moving from the wall center to the corners, and is greater than that for the larger particles due to the effects of the secondary flow. The maximum resuspension rate of the smaller particles occurs above the pipe corners, with the 50 μm particles showing their highest resuspension rate above and at the corners of the pipe.     

Programmed nanoparticle-loaded microparticles for effective antigen/adjuvant delivery

A microscale vaccine containing SiO₂ nanoparticles loaded in CaCO₃ microparticles was constructed using the co-precipitation method. The antigen ovalbumin (OVA) was covalently conjugated with SiO₂ nanoparticles, and these nanoparticles and CpG were co-encapsulated into CaCO₃ microparticles, generating a vaccine with a size of approximately 5.2 μm. Scanning electron microscopy (SEM), energy-dispersive X-ray (EDX), elemental mapping, and Fourier transform infrared (FTIR) analyses confirmed the successful preparation of the microscale vaccine; the vaccine had good storage stability without sustained antigen release, and negligible cytotoxicity to dendritic cells (DCs) and macrophages. Compared to SiO₂ nanoparticles, the microscale vaccine can significantly improve antigen/adjuvant uptake. DCs internalized the entire microscale vaccine into lysosomes via macropinocytosis, and an increase in antigen endo/lysosomal escape was observed by confocal laser scanning microscopy (CLSM). Specifically, DCs pulsed with the vaccine were fully mature, expressing high levels of costimulatory molecules (CD40, CD80, and CD86), MHC II, and MHC I and secreting high levels of proinflammatory cytokines (IL-12, TNF-α, IL-1β, and IL-6). In addition, the vaccine had good in vivo biocompatibility, could protect the antigen from rapid degradation, and increased the retention time in lymph nodes. SiO₂ nanoparticles-in-CaCO₃ microparticles were an excellent carrier for antigen and adjuvant delivery. Hopefully, this study can provide some information on the design of microscale carriers for vaccine delivery systems.