Adam Kulawik

Toward Parallel Modeling of Solidification Based on the Generalized Finite Difference Method Using Intel Xeon Phi

Modern heterogeneous computing platforms have become powerful HPC solutions, which could be applied for a wide range of applications. In particular, the hybrid platforms equipped with Intel Xeon Phi coprocessors offers performance advantages over conventional homogeneous solutions based on CPUs, while supporting practically the same parallel programming model. However, there is still an open issue how scientific applications can utilize efficiently the hybrid platforms equipped with Intel coprocessors.

Porting and optimization of solidification application for CPU–MIC hybrid platforms:

Modern heterogeneous computing platforms have become powerful HPC solutions, which could be applied to a wide range of real-life applications. In particular, the hybrid platforms equipped with Intel Xeon Phi coprocessors offer the advantages of massively parallel computing, while supporting practically the same parallel programming model as conventional homogeneous solutions. However, there is still an open issue as to how scientific applications can efficiently utilize hybrid platforms with Intel MIC coprocessors. In this article, we propose an approach for porting a real-life scientific application to such hybrid platforms, assuming no significant modifications of the application code. It allows us to take advantage of all the computing components, including two CPUs and two coprocessors, for the parallel execution of computational workloads. In this study, we focus on the parallel implementation of a numerical model of the dendritic solidification process in isothermal conditions. We develop a sequence of steps that are necessary for the porting and optimization of the solidification application to hybrid platforms with Intel coprocessors. The main challenges include not only overlapping data movements with computations, but also ensuring adequate utilization of cores/threads and vector units of processors, as well as coprocessors. To reach this aim, we propose an efficient and flexible method for the workload distribution between heterogeneous computing components. For implementing the potential benefits of the proposed approach, we choose a heterogeneous programming model based on a combination of the offload mode for Intel MIC and OpenMP programming standard. The developed approach allows us to execute the whole application up to 9.33× faster than the original parallel version that uses two CPUs. Furthermore, the CPU–MIC hybrid platforms enable achieving the speedup of about 1.9× that of the CPU platform with 24 cores based on the Ivy Bridge architecture, and about 1.5× that of the Haswell-based CPU platform with 36 cores.

Using hStreams Programming Library for Accelerating a Real-Life Application on Intel MIC

The main goal of this paper is the suitability assessment of the hStreams programming library for porting a real-life scientific application to heterogeneous platforms with Intel Xeon Phi coprocessors. This emerging library offers a higher level of abstraction to provide effective concurrency among tasks, and control over the overall performance. In our study, we focus on applying the FIFO streaming model for a parallel application which implements the numerical model of alloy solidification. In the paper, we show how scientific applications can benefit from multiple streams. To take full advantages of hStreams, we propose a decomposition of the studied application that allows us to distribute tasks belonging to the computational core of the application among two logical streams within two logical/physical domains. Effective overlapping computations with data transfers is another goal achieved in this way. The proposed approach allows us to execute the whole application 3.5 times faster than the original parallel version running on two CPUs.

Analysis of the influence of material parameters of the numerical model on the obtained shape of the heat affected zone for the TIG heating process

In this paper the mathematical and numerical models of the heating process of the medium-carbon (AISI 1045) steel element are presented. The heat equation including the convection term is solved using the Petrov-Galerkin formulation. The temperature fields are determined using the copyright software based on the finite element method (three-dimensional). Influence of assumed material properties on the shape of the heat affected and fusion zones is analyzed. The results obtained from the numerical model are compared to experimental ones.

Calculations of the heat source parameters on the basis of temperature fields with the use of ANN

In the paper, a method of automatic choice of the boundary conditions is presented. This method allows the initial estimation the value of the boundary conditions during the heating operation for the welding process. The solutions described in this paper can successfully support the work of the specialists performing numerical simulations. They also significantly reduce the time required to estimate the boundary conditions parameters consistent with the actual experiment. The accuracy of the solution in the presented method is not dependent on the empirical models of boundary conditions for the heating process. It is also much more universal. Artificial neural networks after the learning process can be used to perform a reverse analysis for the heat treatment process, for example, to determine the parameters of welding process in the production process of steel elements.

The impact of the bead width on the properties of the anti-abrasion surfacing weld

This work presents the results of research on the anti-abrasion surfacing welds designated to operate under wear conditions. The main purpose of the work was to produce single-layer surface welds by means of semi-automatic hard-facing/surface welding with the use of filler material containing carbide precipitate and with the use of 10mm- and 20mm- wide beads. The samples were subject to visual and penetrant testing and to destructive testing in the form of macro and micro metallographic testing, hardness testing and bend testing with a view to determine the effect which the beads of various widths have on the analysed factors.

Optimization of the transition path of the head hardening with using the genetic algorithms

An automated method of choice of the transition path of the head hardening in heat treatment process for the plane steel element is proposed in this communication. This method determines the points on the path of moving heat source using the genetic algorithms. The fitness function of the used algorithm is determined on the basis of effective stresses and yield point depending on the phase composition. The path of the hardening tool and also the area of the heat affected zone is determined on the basis of obtained points. A numerical model of thermal phenomena, phase transformations in the solid state and mechanical phenomena for the hardening process is implemented in order to verify the presented method. A finite element method (FEM) was used for solving the heat transfer equation and getting required temperature fields. The moving heat source is modeled with a Gaussian distribution and the water cooling is also included. The macroscopic model based on the analysis of the CCT and CHT diagrams of the med...

Using the artificial neural networks in the modelling of the induction heating

In this paper the method of identification of the parameters of the volumetric heat source simulating the induction heating is presented. The presented method allows one to estimate the power and radius of the heat source for the superficial hardening process with the moving heat source. The solution described in this paper can successfully support the work of the specialist on improvement of numerical simulation and also significantly reduce the time required to estimate the parameters of the boundary conditions. The accuracy of the solution of the presented method is independent from the empirical models of boundary conditions during heating process. It is also much more universal. In the paper the analysis of learning process for artificial neural network (ANN) is described. The error of estimated values by ANN is presented. The precision of presented model is tested for a different tasks.

Computation of phase transformations, strains and stresses fields during multipass superficial hardening process with tempering

In the paper mathematical and numerical models of strains calculation during the heating and cooling process of the medium-carbon steel element on the basis of the finite element method are presented. In the model the melting phenomenon of the superficial layer is taken into account. In the modelling of strains, except thermal strains, structural strains occurring during the phase transformation in the solid state are also included. The phenomena of partly overlapping heat affected zones for a successive path of the moving heat source are taken into account. To determine the phase transformations in the solid state the macroscopic model based on the analysis of the CCT and CHT diagrams is used. In the model the influence of the heating rate on the temperature of sorbite transformation during tempering of the C45 steel element is taken into account. It has a particular importance in the process when heat treatment is performed by the moving heat source. The plastic and transformation strains distributions ...