AlfaBuild

RAR RUS 3401-3401 H-9967-2013 16412815600 0000-0002-8588-3871 Moscow Power Engineering Institute Kirsanov Mikhail Nikolaevich mpei2004@yandex.ru

Moscow, Russian Federation

Formulas for the first two frequencies of natural oscillations of a regular truss The object of research is the frequencies of natural oscillations of a regular spacer flat lattice truss. The mass of the truss is conventionally located at its nodes. Only small vertical oscillations of the masses are considered. The rods of the structure are assumed to be linearly elastic. Method. The modified Dunkerley method is used to derive the formula for the dependence of the first natural frequency of the truss free oscillations. For the second frequency, the form of dependence on the number of panels is taken from the solution of the problem of the first frequency with correction factors that are calculated from the numerical solution by the collocation method at three points. Results. For the first two frequencies of truss oscillations, compact calculation formulas for the dependence on the number of panels are obtained, allowing one to estimate oscillations of a truss with an arbitrary number of panels without loss of accuracy. For an odd number of panels in half a span, the kinematic variability of the structure was discovered and confirmed by the distribution of velocities. © The Author(s), 2025 10.57728/ALF.34.1 69 Lattice truss Natural oscillation frequency Induction Maple Analytical solution Second frequency of oscillations Dunkerley method https://alfabuild.spbstu.ru/article/2025.34.1/ 3401.pdf

REV RUS 3402-3402 0009-0009-9853-3246 Ezra Mesheck 0000-0003-2206-2563 Rynkovskaya Marina Igorevna 0000-0002-9852-3576 Dereje Lami Sileshi 0000-0002-6752-4103 Baza Tewodros Temede Structural dynamics of systems under dynamic loads: A review The object of the research is dynamic load analysis, which plays a fundamental role in structural engineering, ensuring that buildings and infrastructure remain stable under diverse forces such as seismic events, wind induced vibrations, and machinery enervated loads. Method. This study presents a detailed review comparison of widely used software tools, including ANSYS, SAP2000, ABAQUS, and LSDYNA, to assess their capabilities in handling dynamic load calculations. Key factors examined include computational speed, accuracy in predicting structural responses, and adaptability to a range of dynamic scenarios. Results. Notable findings reveal ANSYS excels in transient response computations, ABAQUS demonstrates exceptional reliability in extreme condition simulations, and LSDYNA proves highly effective in modelling impact scenarios. By outlining the specific strengths and limitations of these tools, the study provides engineers with actionable guidance for selecting software aligned with project needs. © The Author(s), 2025 10.57728/ALF.34.2 69 Structural dynamics Dynamic load analysis Seismic analysis Wind induced vibrations Finite element analysis https://alfabuild.spbstu.ru/article/2025.34.2/ 3402.pdf

REV RUS 3403-3403 0000-0002-0035-0903 Trunin Grigorii Aleksandrovich 0000-0003-0541-327X Aksenov Ilya Antonovich 0000-0002-5262-6609 Lisyatnikov Mikhail Sergeevich 0000-0003-4189-877X Prusov Evgeny Sergeevich 0000-0003-0356-1383 Roshchina Svetlana Ivanovna The global lumber market: An overview in the context of food and construction safety The object of research is the international lumber market. The subject of research is the trade and production relations in the global lumber markets. High-grade lumber is used in critical load-bearing structures of buildings and constructions. The quality lumber market and highly skilled labor resources directly impact the construction safety of objects in various countries. The study is based on a comprehensive review of wood production in different countries and the open FAO database. The aim of the study is to identify current trends and the country’s structure of the global lumber market. Methods. The study is based on open FAO data and their statistical analysis using Python analytics tools. Results. Over 60 years, the lumber market has grown by 52.22%, reaching 481.3 million cubic meters in 2022. The market closely follows the dynamics of global crises and shows a steady growth trend. The top ten lumber producers account for 77.2% of the global market, while in 1992, their share was 80.24%. The market structure is highly heterogeneous. The USA, China, and Russia produce 45.7% of the world’s lumber. Almost 30% of all lumber produced globally is exported. In 2022, this figure reached 143.1 million cubic meters and continues to grow. The leading lumber exporters are Canada (18.5%) and Russia (18.2%). Sweden ranks third with a share of 10.4%, followed by Germany (8.6%) and Finland (6.4%). These countries account for 62% of the global lumber export. Lumber imports have also increased, reaching 137.2 million cubic meters, which is 28.5% of global production. The USA and China purchase almost half (45.5%) of all lumber. The UK, Italy, Japan, Germany, and Egypt each occupy less than 5% of the market but more than 3%. © The Author(s), 2025 10.57728/ALF.34.3 69 Lumber in Construction Lumber Market Trends Data Science Statistical Analysis Lumber Production https://alfabuild.spbstu.ru/article/2025.34.3/ 3403.pdf

RAR RUS 3404-3404 0000-0001-9542-2430 Nurmukhametov Renat Rustamovich nrenatkazan@gmail.com

StroyInvestCapital LLC,26, Sinopskaya embankment, Saint Petersburg, Russia, 191167

Multi screw type elements for weak soil reinforcement The object of research is weak soils reinforced with fiber-reinforced plastic pipes with helical screws of increasing diameters from bottom to top, installed in weak soils at defined spacing. The study aims to develop and validate a new predictive settlement calculation model for weak soils reinforced with fiber-reinforced plastic pipes with helical screws of increasing diameters from bottom to top, installed in weak soils at defined spacing. Method. Method is based on a comparative review of closest researches. The reinforced mass is treated as a composite structure in which the vertical elements and surrounding soil undergo joint and simultaneous deformation. An analytical settlement calculation model was formulated based on a newly introduced reinforcement factor, which accounts for the geometry and structural characteristics of the reinforcing elements. Results. To validate the model, five full-scale axial load tests were conducted on soils reinforced with elements of different configurations. At pressure 300 kPa settlement reductions of 63% (samples with two screws) and 72% (three screws) were achieved. Comparative analysis of the calculated settlements and experimental results demonstrated the accuracy and applicability of the proposed method. 10.57728/ALF.34.4 69 Weak soil vertical reinforcement soil reinforcement FRP piles micropiles helical piles screw piles fiber reinforced plastic pipes https://alfabuild.spbstu.ru/article/2025.34.4/ 3404.pdf

RAR RUS 3405-3405 0009-0003-4612-0964 Avdonyushkin Dmitrij Viktorovich 0009-0009-8189-0300 Matveeva Anastasia Igorevna 0009-0004-4214-2418 Kravchinskij Sergej Andreevich 0009-0008-2905-8448 Karchevskaia Anna Stanislavovna 0000-0001-5874-5994 Novokshenov Aleksei Dmitrievich 0000-0003-3177-0959 Borovkov Aleksey Ivanovich borovkov@spbstu.ru 0000-0001-6404-6129 Sherbakov Sergei Sergeevich 0009-0001-7204-1974 Klimkovich Nikita Mikhailovich 0009-0009-8988-6939 Podgayskaya Daria Aleksandrovna Machine learning methods for building reduced-order models The object of research the development of surrogate and reduced-order models for engineering systems based on machine-learning techniques. The study focuses on replacing time-consuming high-fidelity numerical simulations with computationally efficient models that preserve the accuracy of prediction. The approach is verified using a technical system for which reduced models are built from hydrodynamic simulation data. Method. The proposed methodology combines neural networks trained with the Levenberg–Marquardt algorithm and Gaussian process regression with the Matérn kernel. Singular value decomposition is employed to form reduced-order representations of the system. The Levenberg–Marquardt algorithm demonstrated faster convergence and higher stability compared to conventional gradient descent, while Gaussian process regression ensured accurate interpolation of nonlinear dependencies. Results. The integration of singular value decomposition with Gaussian process regression enables rapid reconstruction of the system state vector within seconds while maintaining adequate model fidelity. The developed surrogate models provide reliable approximation of high-fidelity simulation results and significantly reduce computational time. The obtained results confirm the effectiveness of the proposed approach for accelerating engineering analysis and creating digital-twin-based predictive models. 10.57728/ALF.34.5 69 machine learning surrogate model reduced ordered model Gaussian process regression SVD neural network Levenberg-Marquardt algorithm https://alfabuild.spbstu.ru/article/2025.34.5/ 3405.pdf