首页|Study Results from New Mexico State University in the Area of Machine Learning R eported (Comparison of Machine Learning and Electrical Resistivity Arrays To Inv erse Modeling for Locating and Characterizing Subsurface Targets)
Study Results from New Mexico State University in the Area of Machine Learning R eported (Comparison of Machine Learning and Electrical Resistivity Arrays To Inv erse Modeling for Locating and Characterizing Subsurface Targets)
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
Investigators publish new report on Ma chine Learning. According to news reporting originating in Las Cruces, New Mexic o, by NewsRx journalists, research stated, "This study evaluates the performance of multiple machine learning (ML) algorithms and electrical resistivity (ER) ar rays for inversion with comparison to a conventional Gauss-Newton numerical inve rsion method. Four different ML models and four arrays were used for the estimat ion of only six variables for locating and characterizing hypothetical subsurfac e targets." Financial support for this research came from United States Department of Energy (DOE). The news reporters obtained a quote from the research from New Mexico State Univ ersity, "The combination of dipole-dipole with Multilayer Perceptron Neural Netw ork (MLP-NN) had the highest accuracy. Evaluation showed that both MLP-NN and Ga uss-Newton methods performed well for estimating the matrix resistivity while ta rget resistivity accuracy was lower, and MLP-NN produced sharper contrast at tar get boundaries for the field and hypothetical data. Both methods exhibited compa rable target characterization performance, whereas MLP-NN had increased accuracy compared to GaussNewton in prediction of target width and height, which was att ributed to numerical smoothing present in the Gauss- Newton approach."
Las CrucesNew MexicoUnited StatesN orth and Central AmericaCyborgsEmerging TechnologiesMachine LearningNew Mexico State University
NETL
NSTL
