首页|The machine learning method for overlapping peak decompositions in differential scanning calorimetry
The machine learning method for overlapping peak decompositions in differential scanning calorimetry
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
NSTL
Elsevier
An overlapping peak resolving (OPR) method based on the machine learning theory is developed to identify the independent peaks constituting the overlapping peaks in differential scanning calorimetry. The content prediction model (CPM) is presented to analyze the contents of the chemical constituents by decomposing the overlapping peaks and establishing the relationship between the contents and peak areas of the substances. The model is applied respectively to some theoretically constructed and experimentally measured overlapping peaks to validate the reliability and feasibility of the proposed method. The results show that the contents of the chemical constituents can be accurately calculated by the CPM. The validations indicate that the analysis errors of the CPM show a relatively stable trend with the changes of the separation degrees and peak-height ratios. Furthermore, the computational burden of the proposed method is also less than that of the existing OPR techniques.
NSTL
Elsevier
