首页|New Machine Learning Study Findings Have Been Reported by Researchers at Univers ity of Maryland (Interpretable Physics-aware Alkali-silica Reaction Expansion Pr ediction)
New Machine Learning Study Findings Have Been Reported by Researchers at Univers ity of Maryland (Interpretable Physics-aware Alkali-silica Reaction Expansion Pr ediction)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators publish new report on Ma chine Learning. According to news reporting from College Park, Maryland, by News Rx editors, the research stated, "The alkali-silica reaction (ASR) is a major co ntributor to the aging and degradation of infrastructure. Understanding ASR-indu ced expansion in concrete structures and accurately predicting its future progre ssion are critical components of effective risk assessment frameworks." The news correspondents obtained a quote from the research from the University o f Maryland, "This paper presents a study focused developing and interpreting an advanced machine learning model specifically designed to predict expansion. The model strategically integrates two powerful algorithms to achieve this goal. A c omprehensive database comprising 2000 samples of ASR expansion data with various attributes was used to train model. The first algorithm, eXtreme Gradient Boost ing (XGBoost), was employed to establish a predictive model for ASR expansion, a chieving approximately 90% validation accuracy. The second algorit hm, SHapley Additive exPlanations (SHAP), was applied to assess the relative imp ortance of the factors influencing XGBoost model's predictions. This approach pr ovided valuable physical and quantitative insights into input-output relationshi ps, which are often obscured in conventional machine learning methods. The study revealed that higher silica content, elevated alkali levels, and longer reactio n times are strongly correlated increased ASR expansion. In contrast, larger agg regate sizes and higher water-to-cement ratios were associated with reduced expa nsion."
College ParkMarylandUnited StatesN orth and Central AmericaCyborgsEmerging TechnologiesMachine LearningUniv ersity of Maryland
NETL
NSTL
