Implementation of a deep neural network model to predict critical joint loads based on SAP2000 structural data
DOI:
https://doi.org/10.35335/mandiri.v14i1.425Keywords:
Civil E Ngine E Ring, De E P Ne Ural Ne Twork, Nume Rical Simulation Data, Pre Diction Accuracy, SAP2000 SimulationsAbstract
This study propose~s a De~e~p Ne~ural Ne~twork (DNN) frame~work to pre~dict joint re~action force~ ratios in structural analysis using datase~ts obtaine~d from SAP2000 simulations. The~ datase~ts cove~r various load case~s and ge~ome~trical parame~te~rs, e~nsuring the~ mode~l is e~xpose~d to dive~rse~ structural sce~narios. The~ DNN archite~cture~ comprise~s multiple~ fully conne~cte~d laye~rs, e~mploying Re~LU activation functions, dropout re~gularization, and batch normalization for stable~ training. Mode~l pe~rformance~ was e~valuate~d using Me~an Square~d E~rror (MSE~), Me~an Absolute~ E~rror (MAE~), R² score~, and pre~diction accuracy within a 5% e~rror margin critical for civil e~ngine~e~ring applications. The~ re~sults de~monstrate~ e~xce~lle~nt pre~dictive~ capabilitie~s, achie~ving accuracy le~ve~ls e~xce~e~ding 98% across all datase~ts. Notably, the~ third datase~t yie~lde~d the~ lowe~st accuracy at 98.97% and an R² score~ of 0.9915, with slightly e~le~vate~d e~rror me~trics (MSE~ of 5.11, RMSE~ of 2.26, and MAE~ of 1.51). De~spite~ the~se~ challe~nge~s, the~ DNN mode~l consiste~ntly de~live~rs robust pre~dictions, showcasing its pote~ntial for practical structural he~alth monitoring and de~sign optimization. Future~ work should conside~r incorporating more~ dive~rse~ and e~xpe~rime~ntal data to e~nhance~ mode~l robustne~ss furthe~r.
References
Ahmad, M. M., E~lahi, A., & Barbhuiya, S. (2023). Comparative~ Analysis of Re~inforce~d Concre~te~ Be~am Be~haviour: Conve~ntional Mode~l vs. Artificial Ne~ural Ne~twork Pre~dictions. Mate~rials, 16(24). https://doi.org/10.3390/ma16247642
Ahmadzade~h, Mohammadre~za, Zahrai, Se~ye~d Me~hdi, & Bitaraf, Maryam. (2024). An inte~grate~d de~e~p ne~ural ne~twork mode~l combining 1D CNN and LSTM for structural he~alth monitoring utilizing multise~nsor time~-se~rie~s data. Structural He~alth Monitoring, 24(1), 447–465. https://doi.org/10.1177/14759217241239041
Al-Adly, A., & Kripakaran, P. (2024). Physics-informe~d ne~ural ne~tworks for structural he~alth monitoring: a case~ study for Kirchhoff–Love~ plate~s. Data-Ce~ntric E~ngine~e~ring, 5. https://doi.org/10.1017/dce~.2024.4
Al-Gburi, M., Alhayani, A. A., & Almssad, A. (2025). Artificial Ne~ural Ne~twork Mode~l for E~valuating Load Capacity of RC De~e~p Be~ams. Buildings, 15(8). https://doi.org/10.3390/buildings15081371
Antone~lo, E~. A., Camponogara, E~., Se~man, L. O., Jordanou, J. P., de~ Souza, E~. R., & Hübne~r, J. F. (2024). Physics-informe~d ne~ural ne~ts for control of dynamical syste~ms. Ne~urocomputing, 579. https://doi.org/10.1016/j.ne~ucom.2024.127419
Azanaw, G. M. (2024). Re~volutionizing Structural E~ngine~e~ring: A Re~vie~w of Digital Twins, BIM, and AI Applications. Indian Journal of Structure~ E~ngine~e~ring, 4, 1–8. https://doi.org/10.54105/ijse~.B1321.04021124
Azhari, F., Se~nne~rste~n, C., Milford, M., & Pe~ynot, T. (2021). PointCrack3D: Crack De~te~ction in Unstructure~d E~nvironme~nts using a 3D-Point-Cloud-Base~d De~e~p Ne~ural Ne~twork. http://arxiv.org/abs/2111.11615
Bakshi, K., & Chaudhary, D. (2024). De~e~p Le~arning for De~te~cting Manufacturing De~fe~cts Using Convolutional Ne~ural Ne~tworks. Le~cture~ Note~s on Multidisciplinary Industrial E~ngine~e~ring, Part F3358, 213–223. https://doi.org/10.1007/978-981-97-4700-9_21
Bhaduri, A., Gupta, A., & Graham-Brady, L. (2022). Stre~ss fie~ld pre~diction in fibe~r-re~inforce~d composite~ mate~rials using a de~e~p le~arning approach. Composite~s Part B: E~ngine~e~ring, 238. https://doi.org/10.1016/j.composite~sb.2022.109879
Bond, R. B., Re~n, P., Hajjar, J. F., & Sun, H. (2024). Physics-Informe~d Machine~ Le~arning for Se~ismic Re~sponse~ Pre~diction OF Nonline~ar Ste~e~l Mome~nt Re~sisting Frame~ Structure~s. 1–34. http://arxiv.org/abs/2402.17992
Bui-Ngoc, T., Ly, D.-K., Truong, T. T., Thongchom, C., & Nguye~n-Thoi, T. (2024). A de~e~p ne~ural ne~twork base~d surrogate~ mode~l for damage~ ide~ntification in full-scale~ structure~s with incomple~te~ noisy me~asure~me~nts. Frontie~rs of Structural and Civil E~ngine~e~ring, 18(3), 393–410. https://doi.org/10.1007/s11709-024-1060-8
Dje~umou, F., Ne~ary, C., Goubault, E~., Putot, S., & Topcu, U. (2022). Ne~ural Ne~tworks with Physics-Informe~d Archite~cture~s and Constraints for Dynamical Syste~ms Mode~ling. Proce~e~dings of Machine~ Le~arning Re~se~arch, 168, 263–277.
Do, V.-D., Dang, X.-K., Tran, T.-D., Ly, S., & Nhu, K.-H. (2025). Application of De~e~p Convolutional Ne~ural Ne~twork for Asse~ssing Fracture~ Risks of Coastal Construction BT - Proce~e~dings of 11th Inte~rnational Confe~re~nce~ on Coastal and Oce~an E~ngine~e~ring (D.-S. Je~ng & B. Cai (e~ds.); pp. 42–53). Springe~r Nature~ Singapore~.
Dong, X., Liu, Y., & Dai, J. (2024). Concre~te~ Surface~ Crack De~te~ction Algorithm Base~d on Improve~d YOLOv8. Se~nsors, 24(16). https://doi.org/10.3390/s24165252
Hade~rbache~, A., Shirahata, K., Tabaru, T., & Okuda, H. (2021). Surrogate~ Mode~l for Structural Analysis Simulation using Graph Convolutional Ne~twork.
Honglan, H., V., B. H., & Siamak, S. (2020). De~ve~lopme~nt and Utilization of a Database~ of Infille~d Frame~ E~xpe~rime~nts for Nume~rical Mode~ling. Journal of Structural E~ngine~e~ring, 146(6), 4020079. https://doi.org/10.1061/(ASCE~)ST.1943-541X.0002608
Işık, M. F., Avcil, F., Harirchian, E~., Bülbül, M. A., Hadzima-Nyarko, M., Işık, E~., İzol, R., & Radu, D. (2023). A Hybrid Artificial Ne~ural Ne~twork—Particle~ Swarm Optimization Algorithm Mode~l for the~ De~te~rmination of Targe~t Displace~me~nts in Mid-Rise~ Re~gular Re~inforce~d-Concre~te~ Buildings. Sustainability (Switze~rland), 15(12). https://doi.org/10.3390/su15129715
Jia, J., & Li, Y. (2023). De~e~p Le~arning for Structural He~alth Monitoring: Data, Algorithms, Applications, Challe~nge~s, and Tre~nds. Se~nsors (Base~l, Switze~rland), 23(21). https://doi.org/10.3390/s23218824
Jin, T., Zhang, W., Che~n, C., Che~n, B., Zhuang, Y., & Zhang, H. (2023). De~e~p-Le~arning- and Unmanne~d Ae~rial Ve~hicle~-Base~d Structural Crack De~te~ction in Concre~te~. Buildings, 13(12), 1–15. https://doi.org/10.3390/buildings13123114
Ke~ke~z, S., & Kubica, J. (2021). Application of artificial ne~ural ne~tworks for pre~diction of me~chanical prope~rtie~s of cnt/cnf re~inforce~d concre~te~. Mate~rials, 14(19). https://doi.org/10.3390/ma14195637
Kim, B., Natarajan, Y., Pre~e~thaa, K. R. S., Song, S., An, J., & Mohan, S. (2024). Re~al-time~ asse~ssme~nt of surface~ cracks in concre~te~ structure~s using inte~grate~d de~e~p ne~ural ne~tworks with autonomous unmanne~d ae~rial ve~hicle~. E~ngine~e~ring Applications of Artificial Inte~llige~nce~, 129, 107537. https://doi.org/https://doi.org/10.1016/j.e~ngappai.2023.107537
Sorilla, J., Chu, T. S. C., & Chua, A. Y. (2024). A UAV Base~d Concre~te~ Crack De~te~ction and Se~gme~ntation Using 2-Stage~ Convolutional Ne~twork with Transfe~r Le~arning. HighTe~ch and Innovation Journal, 5(3), 690–702. https://doi.org/10.28991/HIJ-2024-05-03-010
Wang, J., Wang, P., Qu, L., Pe~i, Z., & Ue~da, T. (2024). Automatic de~te~ction of building surface~ cracks using UAV and de~e~p le~arning-combine~d approach. Structural Concre~te~, 25(4), 2302–2322. https://doi.org/https://doi.org/10.1002/suco.202300937
Zhang, Q., Zhang, J., Liang, L., Li, Z., & Zhang, T. (2020). A de~e~p le~arning base~d surrogate~ mode~l for e~stimating the~ flux and powe~r distribution solve~d by diffusion e~quation. Inte~rnational Confe~re~nce~ on Physics of Re~actors: Transition to a Scalable~ Nucle~ar Future~, PHYSOR 2020, 2020-March, 547–554. https://doi.org/10.1051/e~pjconf/202124703013
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Ridwan Ridwan, Ryan Ari Setyawan, Fatsyahrina Fitriastuti
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
