CIVILICA We Respect the Science
(ناشر تخصصی کنفرانسهای کشور / شماره مجوز انتشارات از وزارت فرهنگ و ارشاد اسلامی: ۸۹۷۱)

Providing the bivariate anomaly map of Pb and Zn using combination of efficient separating methods in Parkam district, Iran

عنوان مقاله: Providing the bivariate anomaly map of Pb and Zn using combination of efficient separating methods in Parkam district, Iran
شناسه ملی مقاله: RSTCONF03_143
منتشر شده در سومین کنفرانس بین المللی پژوهش در مهندسی، علوم و تکنولوژی در سال 1395
مشخصات نویسندگان مقاله:

Seyyed Saeed Ghannadpour - PhD Candidate, Amirkabir University of Technology, Tehran
Ardeshir Hezarkhani - Full Professor, Amirkabir University of Technology, Tehran
Taraneh Roodpeyma - MSc Student, Amirkabir University of Technology, Tehran
Abbas Golmohammadi - PhD, Ferdowsi University of Mashhad, Mashhad

خلاصه مقاله:
The U-statistic method is one of the most important univariate structuralmethods which consider spatial situation of samples. This method is akind of weighted mean that surrounding points of samples areconsidered in U value determination and by increasing samples number,mentioned error is decreased. However, it separates anomaly based onone variable. In present study, the U-statistic applied on Mahalanobisdistance values of samples because Mahalanobis distance is calculatedbased on several variables. This method is a combination of efficient Ustatistic and Mahalanobis distance. Combination results for Pb and Znelements in Parkam district, Kerman, Iran, led to better performance ofthese two methods. Results show that samples indicated by thecombination of these methods as anomalous are more regular; lessdispersed and are more accurate than using just one of them.

کلمات کلیدی:
Parkam, U-statistic, Mahalanobis distance, Anomaly separating

صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/557474/