Prediction of concrete and FRC properties at high temperature using machine and deep learning: A review of recent advances and future perspectives

Journal of Building Engineering, Год журнала: 2023, Номер 83, С. 108369 - 108369

Опубликована: Дек. 29, 2023

Язык: Английский

---

An explainable machine learning approach to predict the compressive strength of graphene oxide-based concrete

Construction and Building Materials, Год журнала: 2024, Номер 449, С. 138346 - 138346

Опубликована: Сен. 17, 2024

Язык: Английский

---

A 2020 forest age map for China with 30 m resolution

Earth system science data, Год журнала: 2024, Номер 16(2), С. 803 - 819

Опубликована: Фев. 7, 2024

Abstract. A high-resolution, spatially explicit forest age map is essential for quantifying carbon stocks and sequestration potential. Prior attempts to estimate on a national scale in China have been limited by sparse resolution incomplete coverage of ecosystems, attributed complex species composition, extensive areas, insufficient field measurements, inadequate methods. To address these challenges, we developed framework that combines machine learning algorithms (MLAs) remote sensing time series analysis estimating the China's forests. Initially, identify develop optimal MLAs estimation across various vegetation divisions based height, climate, terrain, soil, forest-age utilizing ascertain information. Subsequently, apply LandTrendr detect disturbances from 1985 2020, with since last disturbance serving as proxy age. Ultimately, data derived are integrated result produce 2020 China. Validation against independent plots yielded an R2 ranging 0.51 0.63. On scale, average 56.1 years (standard deviation 32.7 years). The Qinghai–Tibet Plateau alpine zone possesses oldest 138.0 years, whereas warm temperate deciduous-broadleaf averages only 28.5 years. This 30 m-resolution offers crucial insights comprehensively understanding ecological benefits forests sustainably manage resources. available at https://doi.org/10.5281/zenodo.8354262 (Cheng et al., 2023a).

Язык: Английский

---

Identifying interactive effects of spatial drivers in soil heavy metal pollutants using interpretable machine learning models

The Science of The Total Environment, Год журнала: 2024, Номер 934, С. 173284 - 173284

Опубликована: Май 18, 2024

Язык: Английский

---

Predicting the Compressive Strength of Ultra-High-Performance Concrete Based on Machine Learning Optimized by Meta-Heuristic Algorithm

Buildings, Год журнала: 2024, Номер 14(5), С. 1209 - 1209

Опубликована: Апрель 24, 2024

Ultra-high-performance concrete (UHPC) is a recently developed material which has attracted considerable attention in the field of civil engineering because its outstanding characteristics. One key factors design compressive strength (CS) UHPC. As one most potent tools artificial intelligence (AI), machine learning (ML) can accurately predict concrete’s mechanical properties. Hyperparameter tuning crucial ensuring prediction model’s reliability. However, it complex work. The purpose this study to optimize CS method for Three ML methods, random forest (RF), support vector (SVM), and k-nearest neighbor (KNN), are selected Among them, RF model demonstrates superior predictive accuracy, with testing dataset R2 0.8506. In addition, three meta-heuristic optimization algorithms, particle swarm (PSO), beetle antenna search (BAS), snake (SO), utilized hyperparameters. values SO-RF, PSO-RF, BAS-RF 0.9147, 0.8529, 0.8607, respectively. results indicate that SO-RF exhibits highest performance. Furthermore, importance input parameters evaluated, findings prove feasibility model. This research enriches

Язык: Английский

---

Comprehensive Assessment of E. coli Dynamics in River Water Using Advanced Machine Learning and Explainable AI

Process Safety and Environmental Protection, Год журнала: 2025, Номер unknown, С. 106816 - 106816

Опубликована: Янв. 1, 2025

Язык: Английский

---

Explainable AutoML models for predicting the strength of high-performance concrete using Optuna, SHAP and ensemble learning

Frontiers in Materials, Год журнала: 2025, Номер 12

Опубликована: Янв. 21, 2025

Accurately predicting key engineering properties, such as compressive and tensile strength, remains a significant challenge in high-performance concrete (HPC) due to its complex heterogeneous composition. Early selection of optimal components the development reliable machine learning (ML) models can significantly reduce time cost associated with extensive experimentation. This study introduces four explainable Automated Machine Learning (AutoML) that integrate Optuna for hyperparameter optimization, SHapley Additive exPlanations (SHAP) interpretability, ensemble algorithms Random Forest (RF), Extreme Gradient Boosting (XGB), Light (LGB), Categorical (CB). The resulting interpretable AutoML O-RF, O-XGB, O-LGB, O-CB are applied predict strengths HPC. Compared baseline model from literature, O-LGB achieved improvements predictive performance. For it reduced Mean Absolute Error (MAE) by 87.69% Root Squared (RMSE) 71.93%. 99.41% improvement MAE 96.67% reduction RMSE, along increases R 2 . Furthermore, SHAP analysis identified critical factors influencing cement content, water, age curing age, water-binder ratio, water-cement ratio strength. approach provides civil engineers robust tool optimizing HPC reducing experimentation costs, supporting enhanced decision-making structural design, risk assessment, other applications.

Язык: Английский

---

Benchmarking AutoML solutions for concrete strength prediction: Reliability, uncertainty, and dilemma

Construction and Building Materials, Год журнала: 2024, Номер 423, С. 135782 - 135782

Опубликована: Март 19, 2024

Язык: Английский

---

Machine Learning in Solid‐State Hydrogen Storage Materials: Challenges and Perspectives

Advanced Materials, Год журнала: 2024, Номер unknown

Опубликована: Дек. 20, 2024

Abstract Machine learning (ML) has emerged as a pioneering tool in advancing the research application of high‐performance solid‐state hydrogen storage materials (HSMs). This review summarizes state‐of‐the‐art ML resolving crucial issues such low capacity and unfavorable de‐/hydrogenation cycling conditions. First, datasets, feature descriptors, prevalent models tailored for HSMs are described. Specific examples include successful titanium‐based, rare‐earth‐based, solid solution, magnesium‐based, complex HSMs, showcasing its role exploiting composition–structure–property relationships designing novel specific applications. One representative works is single‐phase Ti‐based HSM with superior cost‐effective comprehensive properties, to fuel cell feeding system at ambient temperature pressure through high‐throughput composition‐performance scanning. More importantly, this also identifies critically analyzes key challenges faced by domain, including poor data quality availability, balance between model interpretability accuracy, together feasible countermeasures suggested ameliorate these problems. In summary, work outlines roadmap enhancing ML's utilization research, promoting more efficient sustainable energy solutions.

Язык: Английский

---

Using voice recognition and machine learning techniques for detecting patient‐reported outcomes from conversational voice in palliative care patients

Japan Journal of Nursing Science, Год журнала: 2025, Номер 22(1)

Опубликована: Янв. 1, 2025

Abstract Aim Patient‐reported outcome measures (PROMs) are increasingly used in palliative care to evaluate patients' symptoms and conditions. Healthcare providers often collect PROMs through conversations. However, the manual entry of these data into electronic medical records can be burdensome for healthcare providers. Voice recognition technology has been explored as a potential solution alleviating this burden. research on voice is lacking. This study aimed verify use machine learning automatically using clinical conversation data. Methods We recruited 100 home‐based patients from February May 2023, conducted interviews Integrated Palliative Care Outcome Scale (IPOS), transcribed their an existing tool. calculated rate developed model symptom detection. Model performance was primarily evaluated F1 score, harmonic mean model's positive predictive value, recall. Results The age 80.6 years (SD, 10.8 years), 34.0% were men. Thirteen had cancer, 87 did not. patient 55.6% 12.1%) significantly lower than overall 76.1% 6.4%). scores five total ranged 0.31 0.46. Conclusion Although further improvements necessary enhance our performance, provides valuable insights settings. expect findings will reduce burden recording providers, increasing wider PROMs.

Язык: Английский

---