基于KFCM-RBF优化算法的肥力评价与玉米产量预测
针对传统FCM模糊聚类不适宜处理具有多维关系的土壤肥力评价问题，影响玉米产量的预测精度。因而，本文采用带有基于高斯核函数的模糊聚类算法(KFCM)，优化传统模糊聚类算法，提出了一个基于模糊聚类的土壤肥力评价模型，以提高肥力聚类的准确性、高效性；在此基础上，融合RBF神经网络，创建一个基于RBF神经网络模型，提出了KFCM-RBF优化算法的玉米产量预测模型。并将上述模型运用于吉林省农安县的土壤肥力评价与玉米产量预测仿真实验，结果表明该模型预测精度高，可用于玉米产量的预测，并可为精准施肥提供决策依据。在此基础上，研究表明：该模型具有结构稳定、训练速度快、适应性强、鲁棒性好、预测精度高的特点。
英文摘要:
It is not suitable for traditional FCM fuzzy clustering to deal with soil fertility evaluation with multi - dimensional relationship, and to predict the prediction accuracy of maize yield. In this paper, a fuzzy clustering algorithm based on Gaussian kernel function (KFCM) is used to optimize the traditional fuzzy clustering algorithm, and a soil fertility evaluation model based on fuzzy clustering is proposed to improve the accuracy and efficiency of soil fertility clustering Based on the RBF neural network, a RBF neural network model was established, and the above model was applied to the soil fertility evaluation and maize yield prediction experiment of Nong\&\#39\;an County, Jilin Province. The results show that the model has high prediction accuracy and can be used for Maize yield prediction, and can provide decision-making basis for precision fertilization. On this basis, this paper also proposes a KFCM-RBF optimization algorithm for maize yield prediction model. The results show that the model has the characteristics of stable structure, fast training speed, strong adaptability, good robustness and high prediction accuracy.
查看全文
;
