Research on Evaluation Methods for Soil Fertility
|School||Northwest University of Science and Technology|
|Course||Applied Computer Technology|
|Keywords||Soil fertility Combined Evaluation Fuzzy Comprehensive Evaluation Principal Component Analysis Rough Set|
Soil fertility are direct providers of agricultural growth, soil foundation for sustainable development, it has a direct impact on the yield and quality of agricultural products. Soil fertility evaluation can provide important theoretical basis for guiding agricultural production, more and more of the evaluation methods applied to soil fertility. At present, the evaluation methods used principal component analysis, gray correlation method, expert analysis, fuzzy comprehensive evaluation, information entropy method, but the fertility evaluation studies mainly focused on using a single method to evaluate the soil fertility, lack of comparison and analysis. In this paper, the three evaluation methods in soil fertility, and comparative analysis of the reasons for the inconsistent results of the evaluation. On this basis, based on the combination of rough set evaluation method applied to soil fertility evaluation, a combination of the results of the three single evaluation method, to obtain a more accurate evaluation results. The main contents of this paper are as follows: (1) soil fertility evaluation based on a single evaluation method, principal component analysis (PCA), fuzzy comprehensive evaluation method and gray relational analysis is applied to soil fertility evaluation, experimental data sets used for long-term experiment measured, a total of 211 groups. Sample data were analyzed using principal component analysis to identify the most \Process (AHP) to determine the weight, use the S-type membership function to establish fuzzy relationship matrix samples obtained weights and fuzzy relationship matrix derived evaluation value; gray relational analysis to select the optimal value of the evaluation factors as a reference series, calculate the degree of correlation between the evaluation object reference series, weighted Method associate degree, the final classification results. The weight expression formula (2) differences in the results of the comparative analysis of three single evaluation method analyze the three reasons for the inconsistency of the results of the evaluation, the realization of the principle of the PCA simplification reasoning, drawn PCA. Seen compared with the other two methods, gray correlation analysis and fuzzy comprehensive evaluation, the same weight, but its normalized approach, while the PCA main difference with their weight, leading to the inconsistency of the results of the evaluation of the three methods. (3) based on the combination of rough set evaluation method of fertility evaluation in order to resolve the inconsistency of the results of the three single evaluation method to study a combination of evaluation methods in fertility evaluation. The papers will be based on the combination of rough set evaluation method applied to the soil fertility evaluation, a combination of the three single evaluation results, appraisals and expert evaluation, test results show that the combination of the evaluation of the results of the correct rate of 91.47%, the average results than a single evaluation method 7% higher, in the evaluation of soil fertility, combination evaluation method is better than a single evaluation method.