SVMR在部分稳定氧化锆稳定率预测中的应用

Partially Stabilized Zirconia Stability Prediction Based on SVMR

为了预测部分稳定氧化锆稳定率,讨论了国内外数据处理的相关研究方法,采用回归型支持向量机(Support Vector Machine for Regression,SVMR)方法应用到预测中去。首先,通过工程实例建立预测模型;其次,使用Matlab仿真了SVMR模型和BP(Back Propagation)神经网络模型;最后,完成了两种模型下的部分稳定氧化锆稳定率预测。预测结果表明,SVMR方法的性能优于BP神经网络方法,回归型支持向量机是预测部分稳定氧化锆稳定率的有效方法。

In order to predict the stability of partially stabilized zirconia, the related domestic and foreign research methods for data processing are discussed; partially stabilized zirconia stability prediction method based on support vector machine for regression has been put forward. First of all, relevant data for test are selected from the project example; secondly, support vector machine for regression model and BP neural network model are created by using Matlab; finally, experiment simulation has been finished. The predicted results show that, support vector machine for regression is an effective method for partially stabilized zirconia stability prediction, and the performance of support vector regression is better than BP neural network model.