Engineers often face the question: - Which software package should be chosen to solve a particular problem? The following software systems are used to solve geotechnical problems: 1) Plaxis; 2) Midas GTS NX; 3) Abaqus; 4) Lira - SAPR. Each of the software packages has certain advantages and disadvantages. In this study, the software package "Lira - SAPR" and "Midas GTS NX" are used. With the help of which numerical modeling of the interaction of a group of piles with the base was performed, which is described in the experience of Bartolomey A. A. [1]. The stress-strain state was compared, which was obtained using the following variants of the models of the "base - pile foundation" system: 1) software package "Lira-SAPR": 1.1) piles are modeled by single-node finite elements, which are located with a step specified along the length of the pile and have rigidity in different directions and approximately take into account the surrounding soil around the pile and under its tip (FE-57); 1.2) the soil environment is modeled by non-linearly deformable volumetric finite elements; piles - rod finite elements. 2) "Midas GTS NX": 2.1) the soil environment is modeled by non-linearly deformable volumetric finite elements; piles - rod finite elements that have a "virtual" connection with the surrounding soil; 2.2) soil environment - similarly; piles - volumetric finite elements with the parameters of reinforced concrete. It is shown that the choice of the software package and the method of modeling the base affects the stress-strain state of the "base - pile foundation" system. Modeling the base using the belt stiffness coefficient leads to a quantitative difference between the obtained results and the field study. This modeling method is the least labor-intensive and fast. The disadvantage of this modeling method is that it is necessary to create a separate model for each stage of the load. Modeling the base with volumetric finite elements with a nonlinear deformation law will disturb the identification of the design parameters of the base, which is quite laborious. The disadvantage of this modeling method is that it is necessary to control the correctness of the dimensions of the finite elements and their joint work.

Comparison of the stress-strain state of vertical elements of the frame of a monolithic house (basement, first and fourth floors), depending on the method of modeling the soil environment and piles, is carried out. The use of pile foundations is due to the fact that they provide the transfer of loads to deeper soil layers and, as a rule, a greater bearing capacity compared to shallow foundations. In the design of foundations, engineers face the question of how to model the soil environment and piles? This paper presents the influence of the decision taken (the selected soil model and the method of modeling piles) on the stress-strain state of the vertical load-bearing elements of the house frame. Comparison of the stress-strain state of vertical elements of the frame (basement, first and fourth floors), which were obtained using the following models of the system «base - pile foundation - overhead supporting structures»: 1) the piles are modeled by single-node finite elements, have only vertical stiffness according to the results of testing the piles for vertical static pressing loads, the mutual influence of piles and soil characteristics are not taken into account (FE-56 hereinafter, this is the number of the finite element in the library of elements of the PС «Lira -SAPR») 2) the piles are modeled by single-node finite elements, are located with a given step along the length of the pile and have rigidity in different directions and approximately take into account the surrounding soil around the pile and under its tip (FE-57); 3) the soil environment is modeled by volumetric elastic finite elements; piles - rod finite elements. It is shown that the choice of the foundation model carries stress-strain state not only for the foundation structures, but also for the vertical bearing elements of the house. When using various options for modeling the base: using a single-node finite element that simulates a smoke like elastic ligature (FE-56), using a chain of single-node skinned elements (FE-57), or a volumetric soil massif, it is possible to obtain quantitative differences in stresses from 2 to 20%, and a qualitative change, which is observed in a change in the sign of bending moments.

In the paper, the influence of the selected model of the soil environment on the stress-strain state (SSS) of the pile foundation is studied. The following issue sare considered: 1) analysis of the main models of the soil environment, widely used in modeling the interaction of foundations with soil foundations; 2) Numerical modeling of the stress-strain state of the "base – pile foundation" system was performed using foundation models in the form of: variable stiffness coefficients, volume tricelastic and elastic-plastic elements of the soil mass; 3) a comparison of the SSS of a pile foundation obtained by numerical modelingusing various foundation models and verification of the results by comparing with the data of a field experiment of testing a group of piles is given. This study is based on field experiments on testing full-scale piles, conducted by prof. Bartolomey A.A. and colleagues. In the experiment, a groupof 9 piles with a length of 5 m and a section of 30x30 cm was driven into the ground. The piles were combined with a reinforced concrete grillage. Numerical modeling of the stress-strain state of the system "base - pilefoundation" was carried out using the SP "Lira – SAPR 2019". It was revealed that the calculated values of longitud in alforces in piles modeled by rod elements, and the interaction with the base of the base stiffness factors simulated by variables give good convergence with the data of experimental studies. The error for all experimental fields in a wide range of loads is up to 20%. When determining the value of the variable stiffness coefficients, it is necessary to refine the miteratively more than 3 times. The disadvantage of modeling the foundation with variable stiffness factors is the difficulty in obtaining the correct values of bending moments in piles. When using a soil foundation model in the form of volumetric elastic finite elements, the error in determining the longitudinal forces in piles is up to 45%, and the use of elastic-plastic volumetric soil elements increases the accuracy of the calculation. After comparing the calculated and actual values of piles ettlement, we observe an excellent correlation of the results in the variant of the numerical model using volumetric elastic-plastic finite soil elements with the Mohr-Coulomb strength criterion. The error is within 0.8 ... 2%. The use of the model of volumetric elastic elements of the soil massif leads to an under estimation of settlement in piles within the range of up to 8%. The model using variable foundation stiffness factors also under estimates settlement in piles by up to 15%.

The work compares the stress-deformed states of the pile foundations of the house depending on the method of modeling the joints of the wall panels. The use of wall panels is due to the fact that their installation is a relatively fast technological process, but the disadvantage of such buildings is, among other things, the lack of free spatial planning [1]. During the creation of a numerical model, questions arise: what method (type of connection of panel elements to each other) should be used to model the joints of prefabricated reinforced concrete structures and how does this affect the stress-strain state in above-ground structures and foundations? This paper presents the influence of the adopted decision (chosen method of joint modeling) on the stress-strain state of pile foundations. A comparison was made of the stress-strain state of the pile foundation (piles and grid), which were obtained using the following joint modeling options: 1) reinforced concrete elements: monolithic floor, monolithic staircase-elevator shaft and prefabricated wall panels are rigidly connected to each other. 2) the joints between reinforced concrete elements are made using the principle of "combination of movements", i.e., the nodes of the finite elements of the structures are stitched and interact with each other on the basis of certain parameters: horizontal joints - only vertical movements are taken into account (combination movements in the HSC along the Z axis); vertical – take into account movement only in the horizontal plane (along the X and Y axes, in GCS); 3) joints between reinforced concrete elements are made using the functionality of PC "Sapphire". Horizontal joints take into account filling with solution (the so-called platform joint), the behavior of which is described by the elastic law of deformation. Vertical joints take into account embedded details, with the help of which elements are connected to each other in the corresponding places foreseen by the project. It is shown that the choice of modeling option for the joint of reinforced concrete structures affects the VAT not only of the foundation structures, but also of the vertical load-bearing elements of the building (wall panels and monolithic structures of the stair-elevator shaft). When using various joint modeling options, it is possible to obtain quantitative differences in forces from 2 to 20%, and the type of joint practically does not affect the deformation of foundation structures.

No abstract