首页|New Machine Learning Study Findings Have Been Reported by Researchers at Univers ity of Ottawa (Development of a Machine Learning Model for On-site Evaluation of Concrete Compressive Strength By Sonreb)
New Machine Learning Study Findings Have Been Reported by Researchers at Univers ity of Ottawa (Development of a Machine Learning Model for On-site Evaluation of Concrete Compressive Strength By Sonreb)
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-Researchers detail new data in Machine Learning. According to news reporting originating in Ottawa, Canada, by NewsRx journalists, research stated, "Coring or destructive testing is typically the de fault choice for the evaluation of concrete compressive strength in reinforced c oncrete (RC) structures. However, it can be impractical and/ or not representati ve of all areas of interest in a structure." Funders for this research include FPrimeC Solutions Inc., Mitacs. The news reporters obtained a quote from the research from the University of Ott awa, "While various non-destructive test (NDT) methods can be correlated to conc rete strength, the accuracy of any single NDT used for this purpose is generally low. The SonReb method, which combines ultrasonic pulse velocity readings and r ebound number, has been shown to have improved accuracy over single test methods . However, empirical SonReb equations, calibrated to specific datasets using reg ression analysis, cannot necessarily be applied to concrete from other sources w ithout introducing significant errors. This study presents a practical machine l earning (ML) model for on-site concrete strength prediction. A large database wa s created from available literature along with new experimental test data. Three different ML models based on an adaptive neuro-fuzzy inference system (ANFIS) w ere developed along with a graphical user interface application to facilitate it s use in the field. In addition to the ML models, linear and non-linear regressi on analyses were also conducted and compared with existing equations in the lite rature. The accuracy of each model was subsequently validated against core sampl es extracted from a reinforced concrete slab. The results show that the proposed ML model and non-linear regression provided the most reliable predictions of co ncrete strength of the validation specimen with a mean absolute error of less th an 10 % compared with twelve core samples."
OttawaCanadaNorth and Central Americ aCyborgsEmerging TechnologiesMachine LearningUniversity of Ottawa
NETL
NSTL
