基于小波-PSOSVM的陕甘宁新 农业资源可持续利用评价

王辰璇; 陈莉; 张安安

引用本文:	王辰璇,陈莉,张安安.基于小波-PSOSVM的陕甘宁新农业资源可持续利用评价[J].灌溉排水学报,0,():-.
	wangchenxuan,Chen Li,Zhang Anan.基于小波-PSOSVM的陕甘宁新农业资源可持续利用评价[J].灌溉排水学报,0,():-.

【打印本页】【下载PDF全文】【查看/发表评论】【EndNote】【RefMan】【BibTex】

过刊浏览高级检索

本文已被：浏览 267次下载 0次
分享到：微信更多字体:加大+\|默认\|缩小-
基于小波-PSOSVM的陕甘宁新农业资源可持续利用评价
王辰璇¹, 陈莉², 张安安²
1.厦门理工学院;2.安徽建筑大学

摘要:

【目的】准确评价陕甘宁新农业资源可持续利用水平。【方法】本文选取陕甘宁新地区为研究对象，从经济、科技、社会、自然环境、资源、生态治理6个方面构建了农业资源可持续利用的评价指标体系，并结合2018—2020年的相关数据，运用小波-PSOSVM评价农业资源可持续利用水平。【结果】（1）小波-PSOSVM农业资源可持续评价均方误差MSE为9.411 5×10-5，相关系数为0.968；而PSOSVM在同样的训练集以及同样的测试集下，得到的均方误差MSE、相关系数分别为0.015 3、0.967。小波处理后，PSOSVM预测的精度有所提高，收敛稍加快。（2）小波-SVM农业资源可持续评价均方误差MSE为20.836，相关系数为0.748；而SVM在同样的训练集以及同样的测试集下，均方误差MSE、相关系数分别为30.903、0.634，小波处理后，SVM预测的精度提高，收敛也稍快。【结论】PSO优化后的SVM，预测的精度提高较多，收敛也快很多。陕甘宁新地区农业资源可持续利用评价结果表明，新疆排名第一，甘肃第二，陕西第三，宁夏第四。

关键词: 小波分析；微粒群算法；支持向量机；资源可持续利用评价

DOI：

分类号:S812

基金项目:

Evaluation of sustainable utilization of agricultural resources in Shaanxi-Gansu-Ningxia-Xinjiang based on wavelet-PSOSVM

wangchenxuan¹, Chen Li², Zhang Anan²

1.Xiamen University of Technology;2.Anhui Jianzhu University

Abstract:

【Objective】The quick development of the areas of Shaanxi, Gansu, Ningxia , and Xinjiang has been aided by western development. At the same time, the area's ecological resources look to be vulnerable. Agricultural resources are the foundation of agricultural development, and they must be used in a sustainable manner. 【Methods】The sustainable utilization of agricultural resources in the Shaanxi-Gansu-Ningxia-Xinjiang area was evaluated using wavelet-PSOSVM in this paper. The evaluation index system of sustainable utilization of agricultural resources is constructed from six aspects: economy, science and technology, society, natural environment, resources, and ecological governance, and combined with related data from 2018 to 2020. 【Results】(1) the mean square error of wavelet psosvm agricultural resources sustainable evaluation is MSE = 9.411 5×10-5, and the correlation coefficient is 0.968; psosvm's mean square error MSE and correlation coefficient are 0.015 3 and 0.967, respectively, under the same training and test sets. It shows that following wavelet processing, psosvm's prediction accuracy improves and convergence speeds up marginally. (2) The mean square error of wavelet SVM agricultural resources sustainable evaluation is MSE = 20.836, and the correlation coefficient is 0.748; the mean square error MSE and correlation coefficient of SVM are 30.903 and 0.634, respectively, under the same training set and test set. It also demonstrates that following wavelet processing, SVM prediction accuracy improves and convergence speeds up marginally. 【Conclusion】When we compare wavelet psosvm to wavelet SVM, psosvm to SVM, we get to the same conclusion: the prediction accuracy of PSO optimized SVM is significantly enhanced, and convergence is significantly faster. In short, following wavelet analysis of the index data, the training model's complexity is reduced, the wavelet psosvm's training time is also increased, and the prediction results are satisfactory. More scientific is the wavelet psosvm model. The evaluation results of sustainable utilization of agricultural resources in Shaanxi-Gansu-Ningxia-Xinjiang region show that Xinjiang ranks first, Gansu second, Shaanxi third and Ningxia fourth.

Key words: wavelet analysis; particle swarm optimization; support vector machine; resource sustainability utilization assessment