An official website of the United States government
This study applies machine learning (ML) models to predict the collapse limit state of steel moment resisting frame (SMRF) buildings, considering uncertainties in system parameters and input ground motion characteristics. Structural global collapse is affected by a large number of linear and nonlinear system parameters. One of the main goals of the study is to find the effectiveness of ML methods to predict collapse, as the number of system’s features is reduced. Because of the lack of sufficient experimental data, an ML approach is followed in which three code-compliant SMRF buildings of varying heights (2, 4 and 8 stories), are evaluated up to the collapse limit state, using nonlinear time history analyses. Variability in system parameters and ground motions, as well as potential correlation among some of the parameters, is considered to generate a database of more than 19,000 realizations of collapsed and non-collapsed systems. The ML models are trained and tested with this database, and the efficiency of the models is categorized using different metrics, such as accuracy, F1-score, precision, and recall. Six different ML classification-based techniques are employed to predict collapse, finding that boosting algorithms (eg, AdaBoost and XGBoost) are the best methods for collapse status classification of the evaluated structural systems. Permutation feature importance is applied to identify the main contributors to collapse. The ML models are then retrained using less features, considering first removal of nonlinear deteriorating parameters, and then removal of the hardening nonlinear parameters.
more »
« less
This content will become publicly available on September 16, 2026
Effect and Implications of Feature Selection on Machine Learning Prediction of Seismic Drift Demands and Collapse of Steel Moment Resisting Frames
Machine learning (ML) techniques were generated for different information levels to identify the minimum set of system parameters required for predicting collapse and maximum interstory drift (SDR_max) of steel moment resisting frame (SMRF) buildings. Five baseline modern SMRFs were evaluated under seismic loading with varying system and ground motion (GM) parameters to generate a database. Classification and regression-based ML models were tested at three system information levels to predict collapse and SDR_max, respectively. The ML predictions were mainly controlled by GM parameters and were relatively insensitive to system parameters defining nonlinear behavior, such as spring backbone curve features.
more »
« less
- Award ID(s):
- 2121169
- PAR ID:
- 10656310
- Publisher / Repository:
- Taylor & Francis
- Date Published:
- Journal Name:
- Journal of Earthquake Engineering
- ISSN:
- 1363-2469
- Page Range / eLocation ID:
- 1 to 26
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Among the sustainable initiatives for renewable energy technologies, anaerobic digestion (AD) is a potential contender to replace fossil fuels. The anaerobic co-digestions of goat manure (GM) with sorghum (SG), cotton gin trash (CGT), and food waste (FW) having different mixing ratios, volumes, temperatures, and additives were optimized in single and two-stage bioreactors. The biochemical methane potential assays (having different mixing ratios of double and triple substrates) were run in 250 mL serum bottles in triplicates. The best-yielding ratio was up-scaled to fabricated 2 L bioreactors. The biodegradability, biomethane recovery, and process efficacy are discussed. The co-digestion of GM with SG in a 70:30 ratio yielded the highest biomethane of 239.3 ± 15.6 mL/gvs, and it was further up-scaled to a two-stage temperature-phased process supplemented with an anaerobic medium and fly ash (FA) in fabricated 2 L bioreactors. This system yielded the highest biomethane of 266.0 mL/gvs, having an anaerobic biodegradability of 67.3% in 70:30 GM:SG co-digestion supplemented with an anaerobic medium. The BMP of the FA-amended treatment may be lower because of its high Ca concentration of 205.74 ± 3.6. The liquid fraction of the effluents can be applied as N and P fertigation. The Ca concentration was found to be 24.3, 25.1, and 6.3 g/kg in GM and GM:SG (TS) and SG solid fractions, respectively, whereas K was found to be 26.6, 10.8, and 7.4 g/kg. The carbon to nitrogen ratio of solid fraction varied between 2.0 and 24.8 for return to the soils to enhance its quality. This study involving feedstock acquisition, characterization, and their anaerobic digestion optimization provides comprehensive information and may assist small farmers operating on-farm anaerobic digesters.more » « less
-
Anaerobic digestion is a proven technology for managing manure while harvesting natural gas and digestate as a biofertilizer. The biochemical methane potential (BMP), biodegradability, and kinetic parameters of goat manure (GM) were investigated at different inoculum to substrate ratios (ISRs). The cumulative biomethane yields at the ISRs of 0.0, 0.3, 0.5, 0.8, 1.1, 1.3, and 2.6 were 191.7, 214.3, 214.9, 225.9, 222.1, 222.8, and 229.9 mL gvs−1, respectively. The biomethane yield at all ISRs was significantly higher than control (0 ISR). Above the ISR of 0.0, the biomethane yield was similar among all ISRs. The biodegradability of GM at the ISRs of 0.3, 0.5, 0.8, 1.1, 1.3, and 2.6 varied between 73.3% and 78.7% and was statistically similar. In total, 90% of the yield was observed in 31 and 32 days in control and all other ISRs, respectively. The modified Gompertz equation fitted very well (R2 = 0.99) to the BMP of GM but predicted the lag phase (λ) of 3.2–5.2 days against observed 8–10 days among control and other ISRs.more » « less
-
Marshall-Colon, Amy (Ed.)Abstract Plant specialized metabolites mediate interactions between plants and the environment and have significant agronomical/pharmaceutical value. Most genes involved in specialized metabolism (SM) are unknown because of the large number of metabolites and the challenge in differentiating SM genes from general metabolism (GM) genes. Plant models like Arabidopsis thaliana have extensive, experimentally derived annotations, whereas many non-model species do not. Here we employed a machine learning strategy, transfer learning, where knowledge from A. thaliana is transferred to predict gene functions in cultivated tomato with fewer experimentally annotated genes. The first tomato SM/GM prediction model using only tomato data performs well (F-measure = 0.74, compared with 0.5 for random and 1.0 for perfect predictions), but from manually curating 88 SM/GM genes, we found many mis-predicted entries were likely mis-annotated. When the SM/GM prediction models built with A. thaliana data were used to filter out genes where the A. thaliana-based model predictions disagreed with tomato annotations, the new tomato model trained with filtered data improved significantly (F-measure = 0.92). Our study demonstrates that SM/GM genes can be better predicted by leveraging cross-species information. Additionally, our findings provide an example for transfer learning in genomics where knowledge can be transferred from an information-rich species to an information-poor one.more » « less
-
null (Ed.)Abstract Anaerobic co-digestion is widely adopted to enhance process efficacy by balancing the C/N ratio of the feedstock while converting organic wastes to biomethane. Goat manure (GM) and cotton gin trash (CGT) were anaerobically co-digested in triplicate batch bioreactors. The process was optimized and evaluated utilizing mathematical equations. The liquid fraction of the digestate was analyzed for nitrate and phosphate. The co-digestions with 10 and 20% CGT having the C/N ratios of 17.7 and 19.8 yielded the highest and statistically similar 261.4 ± 4.8 and 262.6 ± 4.2 mL/g vs biomethane, respectively. The biodegradability (BD) of GM and CGT was 94.5 ± 2.7 and 37.6 ± 0.8%, respectively. The BD decreased proportionally with an increase in CGT percentage. The co-digestion having 10% CGT yielded 80–90% of biomethane in 26–39 d. The modified Gompertz model-predicted and experimental biomethane values were similar. The highest synergistic effect index of 15.6 ± 4.7% was observed in GM/CGT; 30:70 co-digestion. The concentration of nitrate and phosphate was lower in the liquid fraction of digestate than the feedstocks, indicating that these nutrients stay in the solid fraction. The results provide important insights in agro-waste management, further studies determining the effects of effluent application on plants need to be conducted.more » « less
