Search

This work introduces a new permutation-symmetry-adapted machine learning diabatization procedure, termed the diabatization by equivariant neural network (DENN). In this approach, the permutation symmetric and anti-symmetric elements in diabatic potential energy metrics (DPEMs) were simultaneously simulated by the equivariant neural network. The diabatization by deep neural network scheme was adopted for machine learning diabatization, and non-zero diabatic coupling was included to increase accuracy in the near degenerate region. Based on DENN, the global DPEMs for 11A′ and 21A′ states of MgH2 have been constructed. To the best of our knowledge, these are the first global DPEMs for the MgH2 system. The root-mean-square-errors (RMSEs) for diagonal elements (H11 and H22) and the off-diagonal element (H12) around the conical intersection region were 5.824, 5.307, and 5.796 meV, respectively. The RMSEs of global adiabatic energies for two adiabatic states were 4.613 and 12.755 meV, respectively. The spectroscopic calculations of the 11A′ state in the linear HMgH region are in good agreement with the experiment and previous theoretical results. The differences between calculated frequencies and corresponding experiment values are 1.38 and 1.08 cm−1 for anti-symmetric stretching fundamental vibrational frequency and first overtone. The global DPEMs obtained in this work should be useful for further quantum mechanics dynamic simulations on the MgH2 system. [ABSTRACT FROM AUTHOR]

In this study, the principle of minimum potential energy has been used to deduce the mapping relationship between beam deformation and rail deformation with consideration of the constraint effects of a subgrade system. Based on the above mapping relationship, a dynamic simulation analysis on the train-track-bridge coupled system has been carried out. Taking the index of riding comfort as the investigation object, in combination with the specified limits in the relevant specification, an analysis has been carried out to determine the control threshold of beam deformation of high-speed railways. According to the simulation results, the dynamic response of different types of high-speed trains shows significantly different sensitivities to the pier settlement and the beam faulting. The threshold value in settlement control in the case of simultaneous settlement of two adjacent piers is smaller than that in the case of the single-pier settlement. Under the operating condition of beam faulting, the wheel unloading rate is more sensitive to the faulting of the beams than the vertical acceleration and Sperling's ride index. To avoid the vertical acceleration specified limit of car body, it is recommended to maintain the pier settlement to less than 9.7 mm, and the beam faulting to less than 5.3 mm. [ABSTRACT FROM AUTHOR]

Effectively homogenizing microstructure heterogeneity within the coarse-graining volume is a long-lasting challenge in crystal plasticity theories. In this paper, we propose a data-based homogenization method that utilizes discrete dislocation dynamic simulations to derive the nearfield correction stress (back stress) for continuum models. This stress accounts for the effective stress field induced by microstructure heterogeneity under the length scale of a coarse-graining volume, providing a physically based homogenization approach. To bridge the gap between discrete and continuous regimes, we introduce two versions of the nearfield correction stress, as well as a criterion based on microstructure and numerical parameters to determine the discrete and continuous transition. Moreover, by analyzing the mathematical connections with the work-conjugated gradient plasticity theory, we further provide a physical explanation for the observed material length scale in the thermodynamically consistent back stress term. This work presents a novel methodology for effectively addressing microstructure heterogeneity and advancing the understanding and modeling of material behavior bridging different length scales. • Data-based homogenization approach addressing the internal stress field within coarse-graining volumes at high strain gradient. • A criterion for the transition between discrete and continuous modelings based on microstructure and numerical length scales. • Physical explanation of the material length scale linked to defect energy potential as the average spacing that hinders heterogeneity within coarse-graining volumes. [ABSTRACT FROM AUTHOR]

[Display omitted] • A GCN-based ML model is developed to predict physical properties of MD simulation. • The ML model is employed for prediction of potential energy of solid–liquid system of metals. • The learned ML model gives a good prediction on the property of training data. • The ML model has generalization performance to predict from unknown graph structures. • The graph representation can be a good notation to predict properties from MD simulations. A graph convolutional networks (GCN)-based machine learning (ML) model is constructed to predict physical properties of metallic materials from graph representation of atomic configuration of molecular dynamics (MD) simulation. The developed ML model is employed for the prediction of time variation of the potential energy of a solid–liquid biphasic system of nickel. The learned ML model gives a good prediction on the property of training data. Moreover, it is confirmed that the ML model has generalization performance sufficient to make adequate predictions on unknown graph structures despite the lack of information on interatomic distances in the graph representation. It is significant in this study to show that the graph representation can be a good notation for the prediction of various properties from MD simulations since there is no established notation for atomic configuration of MD simulation especially for large-scale system of metallic materials. [ABSTRACT FROM AUTHOR]

• Replacement of thermal generator is a simple and efficient building retrofit action. • Three generators are compared using environmental, energy and economic metrics. • A building stock energy model for Italy has been defined (4 buildings, 3 climates). • Heat pumps are preferrable for emission reduction, but lead to high operative costs. • The flexibility of hybrid heat pump allows for high savings on all the objectives. In many countries, old inefficient generators are used for heating purposes. A way to improve energy efficiency in those countries can be the choice of the most suitable heating generators to replace the current ones, without the inconveniences caused by other retrofit measures. In this context, we present a methodology to choose the most appropriate generator through dynamic simulations applied to building stock energy models. Economic, energetic, and environmental parameters guide the choice. The methodology is applied in the Italian residential building scenario, where the most widespread heating device is traditional boiler. Three options are considered: condensing boiler, heat pump, and hybrid heat pump (an alternative-working system with a condensing boiler and heat pump). For a detailed analysis, a white-box bottom-up building stock energy model is defined, through a statistical analysis, consisting of four building typologies in three different external climates. An hourly dynamic building-system simulation has been performed to obtain energy performance, emissions, and operational costs. The results have demonstrated that condensing boiler allows limited economic, environmental, and energy savings (8 - 14 %), compared to traditional boiler. The highest economic savings are obtained with the hybrid heat pump (20 %), while the heat pump alone leads to higher costs, but the highest savings in emissions and non-renewable primary energy (25 - 50 %). The flexibility offered by the hybrid heat pump allows to obtain high savings in different cost scenarios, but also to further reduce CO 2 emissions by implementing an environmental-based control strategy. [ABSTRACT FROM AUTHOR]

[Display omitted] DNAs' photostability is significant to the normal function of organisms. P-Z is a hydrogen bonded artificial DNA base pair, where P and Z represent 2-amino-imidazo[1,2-a]-1,3,5-triazin-4(8H)one and 6-amino-5nitro-2(1H)-pyridone, respectively. The excited-state relaxation mechanism of P-Z pair is investigated using static TDDFT calculations combined with the non-adiabatic dynamic simulations at TDDFT level. The roles of nitro rotation, nitro out-of-plane deformation, and single proton transfer (SPT) along hydrogen bond are revealed. The results of potential energy profile calculations demonstrate that the SPT processes along the hydrogen bonds are unfavorable to occur statically, which is in great contrast to the natural base pair. The non-adiabatic dynamic simulations show that the excited-state nitro rotation and nitro out-of-plane deformation are the two important relaxation channels which lead to the fast internal conversion to S 0 state. The SPT from Z to P is also observed, followed by distortion on P, inducing the fast internal conversion to S 0 state. However, this channel (decay via SPT process) plays minor roles on the excited-state relaxation mechanism statistically. This work shows the great differences of the excited-state relaxation mechanism between the natural base pairs and artificial base pair, also sheds new light into the role of hydrogen bond and nitro group in P-Z base pair. [ABSTRACT FROM AUTHOR]

Gear dynamics analysis based on time-varying meshing stiffness (TMS) is an important means to understand the gear fault mechanism. Based on Jones bearing theory, a bearing statics model was established and introduced into a gear system. The lateral–torsion coupling vibration model of the gear shaft was built by using a Timoshenko beam element. The lumped parameter method was used to build the dynamic model of a gear pair. The dynamic model of a spur gear system was formed by integrating the component model mentioned above. The influence of rectangular and elliptical spalling on TMS was analyzed by the potential energy method (PEM). The fault feature of tooth spalling was studied by dynamic simulation and verified by experiments. It is found that the gear system will produce a periodic shock response owing to the periodic change of the number of meshing gear teeth. Due to the contact loss and the decrease of TMS, a stronger shock response will be generated when the spalling area is engaged. In the spectrum, some sidebands will appear in the resonance region. The results can provide a theoretical guide for the health monitoring and diagnosis of gear systems. [ABSTRACT FROM AUTHOR]