Operational time-series data modeling via LSTM network integrating principal component analysis based on human experience

Today’s information technologies involve increasingly intelligent systems, which come at the cost of increasingly complex equipment. Modern monitoring systems collect multi-measuring-point and long-term data which make equipment health prediction a “big data” problem. It is difficult to extract information from such condition monitoring data to accurately estimate or predict health statuses. Deep learning is a powerful tool for big data processing that is widely utilized in image and speech recognition applications, and can also provide effective predictions in industrial processes. This paper proposes the Long Short-term Memory Integrating Principal Component Analysis based on Human Experience (HEPCA-LSTM), which uses operational time-series data for equipment health prognostics. Principal component analysis based on human experience is first conducted to extract condition parameters from the condition monitoring system. The long short-term memory (LSTM) framework is then constructed to predict the target status. Finally, a dynamic update of the prediction model with incoming data is performed at a certain interval to prevent any model misalignment caused by the drifting of relevant variables. The proposed model is validated on a practical case and found to outperform other prediction methods. It utilizes a powerful deep learning analysis method, the LSTM, to fully process big condition monitoring series data; it effectively extracts the features involved with human experience and takes dynamic updates into consideration.     

Influence of Bridgman solidification on microstructures and magnetic behaviors of a non-equiatomic FeCoNiAlSi high-entropy alloy

The non-equiatomic FeCoNiAlSi alloy is prepared by the Bridgman solidification (BS) technique at different withdrawal velocities (V = 30, 100, and 200 μm/s). Various characterization techniques have been used to study the microstructure and crystal orientation. The morphological evolutions accompanying the crystal growth of the alloy prepared at different withdrawal velocities are nearly the same, from equiaxed grains to columnar crystals. The transition of coercivity is closely related to the local microstructure, while the saturation magnetization changes little at different sites. The coercivity can be significantly reduced from the equiaxed grain area to the columnar crystal area when the applied magnetic field direction is parallel to the crystal growth direction, no matter what is the withdrawal velocity. In addition, the alloy possesses magnetic anisotropy when the applied magnetic field is in different directions.     

Formation and wear mechanism of nickel titanium intermetallics during heat treatment of nickel coating on Ti-6Al-4V substrate

In the present work, inter-diffusion of nickel and titanium and formation of Ni-Ti intermetallic compounds on Ti-6Al-4V substrate have been studied. Initially, nickel was electrodeposited on the alloy using a modified Watts bath solution at a current density of 2 A dm^?2 for 1?h. The coated specimens were then heat treated for different durations at 750, 800 and 850 °C under argon atmosphere. The effects of temperature and time on the characteristics, hardness and wear resistance of intermetallic phases were investigated. The results showed that a multilayer structure was formed after heat treatment, an outer layer of residual nickel, an area of intermetallic layers with different compositions followed by a solid solution of Ni-Ti. It was also observed that an increase in time or temperature at first led to the formation of thicker intermetallic layers; however, after passing a critical point, the intermetallic layers seem to dissolve into the substrate. Furthermore, the wear rates of the diffusion treated samples were four times lower compared to Ti-6Al-4V alloy when sliding against AISI 52100 hardened steel.     

Assessing the expectations associated with pharmaceutical pill colour and shape

Two experiments are reported in which people’s sensory, hedonic, and efficacy expectations associated with pharmaceutical pills of differing colour and shape were assessed. In Experiment 1, 101 participants from the USA viewed online photos of tablets having one of 7 colours and 3 shapes. The participants had to arrange the 21 tablets based on the expectations generated solely by the tablets’ visual properties. The results revealed that the colour of the tablets influenced expected bitterness, expected alertness, and expected efficiency in combating headaches, whereas the shape of the tablets influenced the expected difficulty of swallowing. In Experiment 2, the major findings of Experiment 1 were replicated while using a greater variety of colours, in 358 participants from China, Colombia, and the USA. Importantly, the results revealed some shared expectations across cultures, such as the high expected efficacy of white tablets in combating headaches, or the high expected difficulty of swallowing the diamond-shaped tablets. The results also revealed some differences among the three groups, such as that the colour of the pills influenced how difficult the Chinese participants (but not the other two groups) expected the pills would be to swallow. These findings clearly demonstrate that the differing colours and shapes of pharmaceutical pills set-up significantly different expectations which likely play an important role in terms of people’s subsequent experience.     

Low cycle fatigue behavior of a eutectic 80Au/20Sn solder alloy

The eutectic 80Au/20Sn solder alloy is widely used in high power electronics and optoelectronics packaging. In this study, low cycle fatigue behavior of a eutectic 80Au/20Sn solder alloy is reported. The 80Au/20Sn solder shows a quasi-static fracture characteristic at high strain rates, and then gradually transforms from a transgranular fracture (dominated by fatigue damage) to intergranular fracture (dominated by creep damage) at low strain rates with increasing temperature. Coffin-Manson and Morrow models are proposed to evaluate the low cycle fatigue behavior of the 80Au/20Sn solder. Besides, the 80Au/20Sn solder has enhanced fatigue resistance compared to the 63Sn/37Pb solder.     

高强WSTi544221合金组织与性能研究

研究了轧制变形量对WSTi544221合金棒材显微组织和力学性能的影响,并对Φ10 mm规格的棒材进行不同制度的固溶+时效处理,对比了不同热处理状态下棒材的组织和力学性能。结果表明,随着轧制变形量的增大,WSTi544221合金棒材的晶粒细化程度增大,强度逐渐提高,但塑性变化不大。经870℃×1 h/WC+520℃×6 h/AC固溶+时效处理后,强度与塑性可以获得良好匹配,当抗拉强度达到1 610 MPa、屈服强度达到1 531 MPa时,延伸率和断面收缩率可分别保持在12%和43%。     

Extending SSD Lifespan with Comprehensive Non-Volatile Memory-Based Write Buffers

Journal of Computer Science and Technology - New non-volatile memory (NVM) technologies are expected to replace main memory DRAM (dynamic random access memory) in the near future. NAND flash...     

Study on the hydrogen storage properties of a TiZrNbTa high entropy alloy

High-entropy alloys (HEAs), as a new class of metallic materials, have received more and more attention due to its excellent mechanical properties. In this study, the hydrogen absorption properties, such as hydrogen absorption capacity, thermodynamics, kinetics and cyclic properties, as well as the hydride structure of a newly designed TiZrNbTa HEA were investigated. The results showed that multiple hydrides including ε-ZrH₂, ε-TiH₂ and β-(Nb,Ta)H were found in the TiZrNbTa HEA after hydrogenation. With the increase of temperature from 293 K to 493 K, the maximum hydrogen absorption capacity decreased from 1.67 wt% to 1.25 wt% and the plateau pressure related with β-(Nb,Ta)H hydrides increased from 1.6 kPa to 14.8 kPa. The formation enthalpy of β-(Nb,Ta)H hydride was determined to be −6.4 kJ/mol, which was less stable than that of NbH and TaH hydrides. The results also showed that the TiZrNbTa HEA exhibited a rapid hydrogen absorption kinetic even at the room temperature with a short incubation time, and the hydrogen absorption mechanism was determined to be the nucleation and growth mechanism. Moreover, the hydrogen absorption capacity at 293 K decreased slowly with the cycle numbers, and remained 86% capacity after 10 cycles. Cracking occurred after hydrogen absorption and became worse with cycles.