UDC 004.8 UDC 721.02

APPLICATION OF GENERATIVE NEURAL NETWORKS IN CONCEPTUAL ARCHITECTURAL AND URBAN DESIGN Application of generative neural networks in conceptual architectural and urban design

Published in Architecture and Modern Information Technologies · Issue 3 (76), 2026 · Pages 220–236 · Rubric: Information Technologies and Architecture
EDN: NYVTCG
Received: 26.06.2026 Accepted: 21.09.2026 Published: 01.10.2026
Authors
This article analyzes modern approaches to the application of generative neural networks in conceptual architectural and urban design. Current AI tools are systematized according to their interface type, and the methodology of working with prompts and references is examined. In the experimental part, the effectiveness of generative AI in solving spatial planning tasks is evaluated using the urban environment of Tomsk as a case study. Based on the research, a classification of generation methods is proposed, taking into account project scale and functional tasks; typical design challenges and their solutions are also identified. In conclusion, the transformation of an architect's professional competencies under the influence of neural network technologies is discussed.
artificial intelligence, generative neural networks, architectural visualization, urban planning, conceptual design, Tomsk, node-based interface, Figma Weave, Higgsfield, Nano Banana Pro
Text References
Text (RU) (PDF)
Read Download

1. Jansson, D. G., & Smith, S. M. Design fixation. Design Studies, 1991, vol. 12(1), pp. 3-11. Available at: https://student.cs.uwaterloo.ca/~cs449/w18/Jansson&Smith1991.pdf DOI:https://doi.org/10.1016/0142-694X(91)90003-F

2. Esaulov G.V., Barchugova E.V., Karelin D.A., Moisseev I.M. Artificial intelligence technologies in architectural and urban planning practice, science, and education. Architecture and Modern Information Technologies, 2026, no. 1(74), pp. 229-247. Available at: https://marhi.ru/AMIT/2026/1kvart26/PDF/15_esaulov.pdf DOI:https://doi.org/10.24412/1998-4839-2026-1-229-247 EDN: https://elibrary.ru/VVKBRT

3. Xu C., Qu C. A Modular AI Workflow for Architectural Facade Style Transfer: A Deep-Style Synergy Approach Based on ComfyUI and Flux Models, Buildings, 2026, vol. 16(3), no. 494. Available at: https://www.researchgate.net/publication/400182786_A_Modular_AI_Workflow_for_Architectural_Facade_Style_Transfer_A_Deep-Style_Synergy_Approach_Based_on_ComfyUI_and_Flux_Models DOI:https://doi.org/10.3390/buildings16030494

4. Plocharski A., Swidzinski J., Musialski P. Pro-DG: Procedural Diffusion Guidance for Architectural Facade Generation. Computer Graphics Forum. 2026. Available at: https://www.researchgate.net/publication/390440276_Pro-DG_Procedural_Diffusion_Guidance_for_Architectural_Facade_Generation DOI:https://doi.org/10.1111/cgf.70487

5. Yao S., Budthimedhee K. From Sketches to Renderings: A Comparison of Rapid Visualization Methods for Sketches Based on ControlNet (CN). 2024 IEEE 7th International Conference on Information and Communication Technologies (ICID 2024), 2024, pp. 181-184. Available at: https://soad.kmutt.ac.th/wp-content/uploads/2024/09/INTC02.pdf DOI:https://doi.org/10.1109/ICID64166.2024.11024685

6. Akshov E.A., Terekhova A.K. Generative neural networks in architectural practice: tools and application scenarios. Architecture and Modern Information Technologies, 2025, no. 2(71), pp. 362-374. Available at: https://marhi.ru/AMIT/2025/2kvart25/PDF/22_akshov.pdf DOI:https://doi.org/10.24412/1998-4839-2025-2-362-374 EDN: https://elibrary.ru/UWNKED

7. Vlasova E.L., Vlasova M.L., Borovikova N.V., Karelin D.V. Artificial intelligence in architectural and urban design. Architecture and Modern Information Technologies, 2023, no. 4(65), pp. 311-324. Available at: https://marhi.ru/AMIT/2023/4kvart23/PDF/20_vlasova.pdf DOI:https://doi.org/10.24412/1998-4839-2023-4-311-324