نوع مقاله : مقاله پژوهشی
نویسندگان
1 دانشجوی کارشناسی مدیریت ساخت، گروه مهندسی عمران ، واحد اهواز، دانشگاه آزاد اسلامی، اهواز، ایران.
2 2- دانشیار گروه علوم و مهندسی آب، واحد اهواز، دانشگاه آزاد اسلامی، اهواز، ایران
3 استادیار، گروه علوم و مهندسی آب، واحد اهواز، دانشگاه آزاد اسلامی، اهواز، ایران.
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Morning glory spillways play a critical role in water flow management in dams and reservoirs, influenced significantly by the discharge coefficient. This coefficient determines the efficiency and risk of spillway performance under flood conditions. In this study, using 80 experimental datasets collected from two morning glory spillway inlet sections with square and circular zigzag shapes (featuring 4, 8, and 12 zigzags), two machine learning models—Support Vector Machine (SVM) and Gene Expression Programming (GEP)—were applied to simulate the discharge coefficient. Independent variables included the number of zigzags (n), Froude number (Fr), relative water head (H/P), and spillway shape index (R/D). Performance metrics (RMSE, MAE, R²) were employed to evaluate the accuracy of the models. Among various SVM models, the RBF kernel with γ = 0.1 yielded the most optimal results. The training and testing phases for the circular spillway showed (RMSE, MAE, R²) values of (0.9262, 0.0696, 0.0848) and (0.9820, 0.0346, 0.0398), respectively, while for the square spillway, these values were (0.9707, 0.073, 0.0904) and (0.9334, 0.0676, 0.0787). The GEP model demonstrated superior performance, particularly for the circular spillway with three genes, a head size of 9, and 45 chromosomes, yielding (RMSE, MAE, R²) values of (0.9778, 0.0375, 0.0451) and (0.9811, 0.0315, 0.0396) in the training and testing phases, respectively. For the square section, the GEP model with 55 chromosomes achieved (RMSE, MAE, R²) values of (0.9741, 0.0494, 0.0597) and (0.9591, 0.0503, 0.0594) for training and testing, respectively.
کلیدواژهها [English]
Project risk encompasses unforeseen events affecting time, cost, and quality. Effective management involves analysis and mitigation, crucial in engineering, particularly dam design. Karamoz et al. (2015) examined water diversion system design challenges. Maghrebi et al. (2016) assessed spillway risks for Chandir Dam. Bahadori and Karimai-Tabarestani (2019) analyzed reservoir dam height and crossing risks. Faizi et al. (2019) used FEMA and RAMCAP methods for Liro dam risk evaluation. Iqbalizadeh et al. (2023) advocated multi-level risk analysis for spillway redesign, while Rezapour and Hashempour (2017) proposed hybrid models for optimizing spillway dimensions. Lakos et al. (2020) and Frizel et al. (2020) highlighted the need for safe, economical spillways due to large dam construction and higher safety standards.
The literature review confirms extensive research in risk assessment, focusing on hydraulic and hydrological scenarios. However, the forthcoming study uniquely applies MLMs, including SVM and GEP to evaluate the Cd of the glory-labyrinth spillway. This is achieved by introducing labyrinth configurations at the inlet inlet—a novel approach not previously explored in existing studies.
In this study, a comprehensive approach was employed to simulate the Cd of the glory-labyrinth spillway, utilizing 80 laboratory data sets collected from two distinct inlet sections featuring square and circular labyrinth configurations. These configurations varied in the number of labyrinths, specifically four, eight, and twelve. To accurately model the Cd, two advanced MLMs were implemented: SVM and GEP. The independent variables considered in the simulations included the number of labyrinth (n), Froude number (Fr), relative water load (H/P), and the weir shape index (R/D). These variables were chosen due to their critical influence on the hydraulic behavior of the weir. To assess the accuracy and reliability of the models, performance evaluation indices, namely Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and the coefficient of determination (R²), were employed. These indices provided a quantitative measure of the models’ predictive capabilities, ensuring that the simulated results closely align with the observed data.
In the evaluation of various SVM models, the RBF kernel function with γ set to 0.1 yielded the most optimal results. The model’s performance metrics (RMSE, MAE, R²) during the training and testing phases were (0.9262, 0.0696, 0.0848) and (0.9820, 0.0346, 0.0398) for the circular spillway, and (0.9707, 0.073, 0.0904) and (0.9334, 0.0676, 0.0787) for the square section. Superior results were obtained using the GEP model, particularly with three genes, a head size of 9, and 45 chromosomes. For the circular spillway, the GEP model achieved indices of (0.9778, 0.0375, 0.0451) and (0.9811, 0.0315, 0.0396) during training and testing, respectively. In the square section, the model with 55 chromosomes showed performance values of (0.9741, 0.0494, 0.0597) and (0.9591, 0.0503, 0.0594) in the training and testing phases, respectively.
The evaluation of various SVM models identified the RBF kernel function with a specific γ value as yielding the most optimal results. The model's performance was assessed for both circular and square spillways, showing strong metrics in both training and testing phases. Additionally, the GEP model, particularly with specific genetic configurations, demonstrated superior performance across different spillway geometries, in
All authors contributed equally to the conceptualization of the article and writing of the original and subsequent drafts.
Data available on request from the authors.
The authors would like to thank all participants of the present study.
The authors avoided data fabrication, falsification, plagiarism, and misconduct.
The author declares no conflict of interest.
)