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.     

Densification mechanism during reactive hot pressing of B4C-ZrO2 mixtures

The densification behaviors of pure B₄C and B₄C-ZrO₂ mixtures were compared during hot pressing. The results showed that in-situ formed ZrB₂ effectively enhanced the densification process of B₄C-ZrO₂ mixtures, more significantly during the intermediate stage. Within the relative density ranging from 0.75 to 0.90, the B₄C-15?wt%ZrO₂ mixture (B15Z) achieved the maximum densification rate as twice much as that of pure B₄C. The stress exponent n>3 indicated plastic deformation was the dominant densification mechanism of B15Z. The viscosities of plastic flow were evaluated using Murray-Rodger-William equation and the viscosity of B15Z was only a quarter of that in pure B₄C. The sintering activation energy was calculated to be 305.9?kJ/mol for pure B₄C and 197?kJ/mol for B15Z, respectively. It was proposed that the lower viscosity of plastic flow and activation energy accelerated the sliding and propagating motions of plastic flow, by which underlain the enhanced densification behaviors of B₄C-ZrO₂ mixtures.     

Back-analysis of geotechnical parameters on PVD-improved ground in the Mekong Delta

This study presents a back-analysis of geotechnical parameters on prefabricated vertical drain improved ground at a site in the Mekong Delta. Various time?settlement behaviors that reflected different clay thicknesses and loading patterns were observed. The total surface settlement behavior at several monitoring locations was simulated using an updated exponential method that considered staged construction. The analyzed results were validated by substituting the values into a theoretical solution for radial consolidation. The estimated theoretical behaviors were comparable with the monitored behaviors. The geotechnical parameters were back-analyzed by applying the previously analyzed results to various theoretical and empirical formulas. However, the use of extensometer data that were installed at large intervals produced different values of the geotechnical properties. Furthermore, finite element analysis supported the back-analyzed total settlement behaviors and nearly disregarded the application of the geotechnical properties that were obtained using either surface or subsurface settlement data. However, settlements and excess pore pressures in the sublayers were not successfully predicted even when the geotechnical properties were adjusted. Thus, subsurface instruments that can be installed closely in thick clay deposits are required to reliably reevaluate the variations in geotechnical properties along a certain depth.     

智能化救援监控系统在煤矿安全生产中的应用

针对煤矿发生事故后传统救援监控系统无法实时对井下人员进行动态定位，导致矿井救援盲目性大、救援效率差、救援难度大等技术难题，为了进一步提高煤矿救援效率，通过技术研究，设计了一套以通信基站为核心的智能化救援监控系统，分析了该系统结构组成、工作原理，通过在担水沟煤矿井下实际应用效果来看，智能化救援监控系统对人员定位精准度达95%，实现人员动态位置三维成像，救援效率提高至80%以上，有效缩短了煤矿事故救援时间，取得了显著应用成效。     

Classification and regression models of audio and vibration signals for machine state monitoring in precision machining systems

We present a data-driven method for monitoring machine status in manufacturing processes. Audio and vibration data from precision machining are used for inference in two operating scenarios: (a) variable machine health states (anomaly detection); and (b) settings of machine operation (state estimation). Audio and vibration signals are first processed through Fast Fourier Transform and Principal Component Analysis to extract transformed and informative features. These features are then used in the training of classification and regression models for machine state monitoring. Specifically, three classifiers (K-nearest neighbors, convolutional neural networks and support vector machines) and two regressors (support vector regression and neural network regression) were explored, in terms of their accuracy in machine state prediction. It is shown that the audio and vibration signals are sufficiently rich in information about the machine that 100% state classification accuracy could be accomplished. Data fusion was also explored, showing overall superior accuracy of data-driven regression models.     

Waves in generalized thermo-viscoelastic infinite continuum with cylindrical cavity due to three-phase-lag time-nonlocal heat transfer

Abstract

Generalized thermoelastic interactions due to three-phase-lag time-nonlocal heat transfer in a Kelvin-Voigt type infinitely extended visco-thermoelastic continuum with cylindrical cavity has been investigated. The two-temperature generalized thermoelasticity theory has also been taken into account. The problem has been solved in the domain of Laplace on the assumption that the surface of the cavity is free from traction and is subjected to a smooth and time-dependent-heating effect. Laplace inversion of the transformed solutions has been carried out numerically. The obtained numerical data for different considerations are plotted in graphs to study the effects of time-nonlocal parameter, two-temperature parameter and visco-thermoelastic relaxation parameter on different thermoelastic quantities of physical interest.     

基于云概率密度分布估计的大坝监测数据分析

大坝运行监测易受自然环境和监测条件影响，存在时间和空间上的变异性，监测数据具有不确定性。以云理论的随机性和不确定性分析方法为基础，并与空间数据辐射思想相结合，建立了云滴概率密度分布估计模型，然后导出云概率密度分布函数，依据样本监测数据推求母体空间数据的分布特征，并设计了基于逆向云算法云变换的计算程序。分析陆浑水库1979~1999年测压管监测数据和位移变形数据的云概率密度分布特征和云数字特征，得出了20 a来大坝的数据分布特征和运行状态。监测数据分析结果表明，云概率密度分布估计不仅能有效合理地分析大坝的运行状态，而且能够依据云数字特征来判断监测状态和监测环境的异常变化。      

Effect of grinding-induced cyclic heating on the hardened layer generation in the plunge grinding of a cylindrical component

This paper discusses the effects of the grinding-induced cyclic heating on the properties of the hardened layer in a plunge cylindrical grinding process on the high strength steel EN26. It was found that a multi-pass grinding brings about a uniform and continuous hardened layer along the circumference of the cylindrical workpiece. An increase of the number of grinding passes, leads to a thicker layer of hardening, a larger compressive residual stress and a deeper plastic deformation zone. Within the plastic deformation zone, the martensitic grains are refined by the thermo-mechanical loading, giving rise to a hardness of 12.5% higher than that from a conventional martensitic transformation. The coupled effects of heat accumulation and wheel wear in the multi-pass grinding are the main causes for the thickening of the hardened layer. A too small infeed per workpiece revolution would result in insufficient grinding heat, and in turn, bring about an undesirable tempered hardened layer and a reduction of its hardness.     

Flame Front Deformation Instabilities of Filtration Combustion for Initial Thermal Perturbation

The flame front deformation instability of low-velocity filtration combustion within an inert packed bed is studied based on the initial preheating non-uniformity. Based on the experimental phenomena, an initial thermal perturbation model is numerically proposed so as to predict the deformation behaviors of the flame front instabilities. The numerical prediction indicates that the assumption of an initial thermal perturbation is a feasible explanation as the cause of the flame front inclination instability. As the initial thermal perturbation increases, the phenomena of the flame front break and shrinking instabilities could easily occur at high filtration velocity or low equivalence ratio. Moreover, the evolutions of the flame front break rate and the shrinking rate are quantitatively analyzed.