Traceability Analysis of Heavy Metals in Soils Based on Data Analysis and Self-Organization Map Method

In order to investigate the sources of soil heavy metals in the mines near Hebei Region A, the 2019 soil heavy metal data from Hebei Region A were used as a benchmark for ecological hazard coefficient assessment and classification traceability of eight heavy metal elements using statistical analysis, correlation analysis, and principal component analysis. The improved Principal Component Analysis plus Self-Organizing Map (PCA-SOM) model algorithm is used to verify and analyze the classification and traceability results of the above heavy metals. The study shows that the ecological risk level around the three mining sites in the study area is at high to very high risk; The source of heavy metals is classified by the Analysis, Cr, Cu, Ni three heavy metals as natural sources, As, Pb, Zn three heavy metal pollution from mines. The heavy metal Cd was derived from agricultural pollution and the heavy metal Hg was derived from mercury-related smelting, both from operations and human activities. Finally, the improved PCA-SOM model algorithm verified that the heavy metals were classified correctly. The improved PCA-SOM model algorithm improved the classification speed by nearly 10% compared to the traditional SOM algorithm, which showed that the Accuracy and validity of the model.