Classifying the quality of vermicompost using PCA-HCA combined with PLS regression model
The vermicompost performs a series of different physiochemical properties.An issue is urgent to be addressed before the vermicompost fertilizing,which is:how to high-efficiently and rapidly evaluate the comprehensive quality of vermicompost from parts of simple and key indexes rather than the entire physiochemical properties.In this study,the vermicompost with different transformation time and different types of materials were produced.A total of 23 main vermicompost physiochemical indexes were tested and analyzed using statistics and chemometrics methods.Among them,13 specific indexes were selected for further evaluation.Based on 13 indexes,firstly,principal component analysis(PCA)and hierarchical cluster analysis(HCA)were employed to categorize vermicompost quality.Secondly,partial least squares regression(PLS)-discriminant analysis(DA)was utilized to assess the effectiveness of the classification results.Finally,an overall evaluation method based on PLS model was constructed and verified.The results indicated that the quality of vermicompost could be classified into three grade levels by the coupling method of PCA and HCA.And this classification was validated to be effective and significant according to the verification model results of PLS-DA.The vermicompost evaluation PLS classification model could be as follows:Vermicompost quality grade Y=3.0796+0.0026×TOC-0.1381×HS-0.1446×HA-0.1378×TN-0.1355×TP-0.1494×AK-0.1324×AN-0.1402×AP+0.0004×EOC+0.03985×ROC+0.07685×C/N-0.0049×Kos-0.1481×HI(TOC,HS,HA,TN,TP,AK,AN,AP,EOC,ROC,C/N,Kos,HI represent total organic carbon,humic substance,humic acid,total nitrogen,total phosphorus,available potassium,alkaline hydrolyzable nitrogen,available phosphorus,easily oxidizable organic carbon,difficult to oxidize organic carbon,carbon nitrogen ratio,oxidation stability coefficient,humification index,respectively).The calculation result of this model showed that the standard of vermicompost quality was:if Y was between 0.45-1.56,the vermicompost quality could be classified as best quality(I);if Y was between 1.63-2.20,the vermicompost quality could be classified as medium quality(II);and if Y was between 2.28-3.72,the vermicompost quality could be classified as normal quality(III).The variable important value in projection indicated that the top five key indexes affecting the quality of vermicompost was HI>TN>HS>HA>AN.This study successfully provided effective coupling technology of"PCA+HCA+PLS"in comprehensively evaluating the vermicompost quality,which will be helpful and significant for the vermicompost fertilizing application and promoting the vermicomposting industry development.
vermicompostgrade evaluationprincipal component analysishierarchical cluster analysispartial least squares regression analysis
NETL
NSTL
万方数据
