Vol 24, No 4 (2022)

One of modern global optimization algorithms is method of Strongin and Piyavskii modified by Sergeev and Kvasov diagonal approach. In recent paper we propose an extension of this approach to continuous multivariable functions defined on the multidimensional parallelepiped. It is known that Sergeev and Kvasov method applies only to a Lipschitz continuous function though it effectively extends one-dimensional algorithm to multidimensional case. So authors modify We modify mentioned method to a continuous functions using introduced by Vanderbei ε-Lipschitz. Because multidimensional parallelepiped is a convex compact set, we demand objective function to be only continuous on a search domain. We describe extended Strongin’s and Piyavskii’s methods in the Sergeev and Kvasov modification and prove the sufficient conditions for the convergence. As an example of proposed method’s application, at the end of this article we show numerical optimization results of different continuous but not Lipschitz functions using three known partition strategies: “partition on 2”, “partition on 2N” and “effective”. For the first two of them we present formulas for computing a new iteration point and for recalculating the ε-Lipschitz constant estimate. We also show algorithm modification that allows to find a new search point on any algorithm’s step.

Currently, additive manufacturing technologies develop actively. This requires creation of computational methods to describe physical processes occurring at the time of manufacturing. One of the methods used for the production of metal powder parts is the method of selective laser melting. This paper presents an SPH-based numerical technique for modeling the process of powder sintering under the influence of a laser beam. The flow of liquid formed as a result of melting is described by the Navier-Stokes equations. Pressure forces, viscous effects and surface forces at the interface are included in the force balance. The thermal state is determined from the energy conservation law, which takes into account thermal processes, volumetric absorption of laser radiation energy, convective heat exchange with the external environment and radiation. Phase transitions between solid and liquid phases are described in the framework of the generalized formulation of the Stefan problem. The calculation method is verified on tests specific to the class of problems under consideration. A comparison is made with the analytical solution, as well as with solutions obtained by other modifications of the SPH method, and with experimental data.

The article investigates the method of peridynamics, which is an alternative approach to solving destruction problems based on integral equations. It is assumed that particles in a continuum interact with each other at a finite distance, as in molecular dynamics. Damage is part of the theory at the level of two-particle interactions, so damage finding and destruction occurs when solving the equation of motion. During this work, bondbased and state-based peridynamics models of destruction used in the Sandia Laboratory were described and implemented within the framework of the MoDyS molecular dynamics software package. In the bond-based model, the defining relationship is the bond stiffness function, which corrects the force of particle-particle interaction and imposes a restriction on the use of the Poisson’s ratio. The state-based model generalizes the bond-based approach and may be applied to materials with any Poisson’s ratio. The relationship of both models is ascertained. Calculation convergence is demonstrated on the example of a one-dimensional elasticity problem. The possibility of using the implemented models for fracture problems is also shown.

Mathematical and computer modeling of daily thermometry allows to study processes of human thermal homeostasis more deeply. In practice, thermometry data is obtained using a digital thermometer, which autonomously reads the temperature of human skin in certain time intervals. The aim of present work is to analyse the methods of modeling and processing of human daily thermometry data. The first method consists in applying linear discrete stochastic models in the state space with Gaussian noises and known vector of input actions, while the estimation of the state vector is performed by discrete covariance Kalman filter. The second method assumes that the vector of input actions is unknown, and the S. Gillijns and B.D. Moor algorithm is used to process daily thermometry data. An alternative option is to use a model with an extended state vector and a Kalman filtering algorithm. The third method takes into account the presence of anomalous measurements (outliers) in the measurement data, and correntropy filter is proposed for their effective filtering. Numerical experiments for modeling and processing of daily thermometry data in MATLAB were carried out in order to compare the quality of discrete filtering algorithms. Modeling of thermometry data was carried out using a three-dimensional model 3dDRCM (3-dimension Discrete-time Real-valued Canonical Model). The results obtained can be used in the study of human daily thermometry processes, for example, to study the reaction of the athlete’s body to the received load.