Failure pattern recognition in uniaxial unidirectional loading of EMB nodes via RBF-Hermite-TWSVM classification model

Structures, Год журнала: 2025, Номер 74, С. 108654 - 108654

Опубликована: Март 13, 2025

Язык: Английский

---

Prediction Model of Farmland Water Conservancy Project Cost Index Based on PCA–DBO–SVR

Sustainability, Год журнала: 2025, Номер 17(6), С. 2702 - 2702

Опубликована: Март 18, 2025

With the gradual cessation of budget quota standards and emphasis on market-based pricing, accurately predicting project investments has become a critical issue in construction management. This study focuses cost indicator prediction for irrigation drainage projects to address absence farmland water conservancy achieve accurate efficient investment prediction. Engineering characteristics affecting indicators were comprehensively analyzed, principal component analysis (PCA) was employed identify key influencing factors. A model proposed based support vector regression (SVR) optimized using dung beetle optimizer (DBO) algorithm. The DBO algorithm SVR hyperparameters, resolving issues poor generalization long times. Validation 2024 data from Liaoning Province showed that PCA–DBO–SVR achieved superior performance. For electromechanical well projects, root mean square error (RMSE) 1.116 million CNY, absolute (MAE) 0.910 percentage (MAPE) 3.261%, R2 reached 0.962. ditch 0.500 MAE 0.281 MAPE 3.732%, 0.923. outperformed BP, SVR, PCA–SVR models all evaluations, demonstrating higher accuracy better capability. provides theoretical developing offers valuable insights dynamically adjusting national improving fund

Язык: Английский

---

Hybrid Intelligent Model for Predicting Corrosion Rate of Carbon Steel in CO₂ Environments

Materials and Corrosion, Год журнала: 2025, Номер unknown

Опубликована: Март 23, 2025

ABSTRACT This study aims to construct a prediction model for the internal corrosion rate of offshore pipelines in CO 2 environments, with intention providing effective and protection strategies oil gas industry. By conducting investigative analysis integrating experimental data, principal component (PCA) was employed extract primary influencing factors, which were used as input variables support vector regression (SVR) output variable. The particle swarm optimization (PSO) algorithm utilized optimize hyperparameters model, enhancing accuracy. results indicate that first eight components account 95.9% cumulative contribution, optimized SVR achieved correlation coefficient (R ) exceeding 0.90. Compared other models methods, PCA PSO effectively predicts pipelines, offering theoretical protection.

Язык: Английский

---

A review of crack research in concrete structures based on data-driven and intelligent algorithms

Structures, Год журнала: 2025, Номер 75, С. 108800 - 108800

Опубликована: Апрель 8, 2025

Язык: Английский

---

Machine Learning Algorithm-Based Prediction Model and Software Implementation for Strength Efficiency of Cemented Tailings Fills

Minerals, Год журнала: 2025, Номер 15(4), С. 405 - 405

Опубликована: Апрель 11, 2025

A novel artificial intelligence (AI) application was proposed in the current study to predict CTF’s compressive strength (CS). The database contained six input parameters: age of curing for specimens (AS), cement–sand ratio (C/S), maintenance temperature (T), additives (EA), additive type (AT), concentration (AC), and one output parameter: CS. Then, adaptive boosting (AdaBoost) applied existing AI soft computing techniques, using AdaBoost, random forest (RF), SVM, ANN. Data were arbitrarily separated into training (70%) test (30%) sets. Results confirm that AdaBoost RF have best prediction accuracy, with a correlation coefficient (R2) 0.957 between these sets AdaBoost. Using Python 3.9 (64-bit), IDLE (Python 64-bit), PyQt5 achieve machine learning model computation software function interface development, this can quickly property CTF specimens, which saves time costs efficiently backfill researchers developing new eco-efficient components.

Язык: Английский

---

Corrosion fatigue life prediction method of aluminum alloys based on back-propagation neural network optimized by Improved Grey Wolf optimization algorithm

Journal of Materials Science, Год журнала: 2024, Номер 59(23), С. 10309 - 10323

Опубликована: Июнь 1, 2024

Язык: Английский

---

Predicting bond strength of corroded reinforced concrete after high-temperature exposure: A stacking model and feature selection

Construction and Building Materials, Год журнала: 2024, Номер 456, С. 139290 - 139290

Опубликована: Ноя. 26, 2024

Язык: Английский

---

Bond strength and failure mode prediction model for recycled aggregate concrete based on intelligent algorithm optimized support vector machine

Structures, Год журнала: 2024, Номер 71, С. 107999 - 107999

Опубликована: Дек. 20, 2024

Язык: Английский

---

Study of factors affecting the magnetic sensing capability of shape memory alloys for non-destructive evaluation of cracks in concrete: using response surface methodology (RSM) and artificial neural network (ANN) approaches

Heliyon, Год журнала: 2024, Номер 10(15), С. e35772 - e35772

Опубликована: Авг. 1, 2024

Currently, the field of structural health monitoring (SHM) is focused on investigating non-destructive evaluation techniques for identification damages in concrete structures. Magnetic sensing has particularly gained attention among innovative techniques. Recently, embedded magnetic shape memory alloy (MSMA) wire been introduced cracks components through while providing reinforcement as well. However, available research this regard very scarce. This study analyses parameters affecting capability MSMA crack detection beams. The response surface methodology (RSM) and artificial neural network (ANN) models have used to analyse first time. were trained using experimental data obtained literature. aimed predict alteration flux created by a beam that 1 mm wide after experiencing fracture or crack. results showed change was affected position with respect magnet beam. RSM optimisation maximum when placed at depth 17.5 from top beam, present an axial distance 8.50 permanent magnet. 9.50 % considering aforementioned parameters. ANN prediction optimal 10 1.1 mm, respectively. suggested larger requires diameter multiple sensors magnets

Язык: Английский

---

Study on bonding properties and constitutive model of steel bar and nano-SiO2 reinforced recycled aggregate concrete

Construction and Building Materials, Год журнала: 2024, Номер 449, С. 138411 - 138411

Опубликована: Сен. 25, 2024

Язык: Английский

---