Least squares support vector machine ( LS-SVM) developed based on support vector machines ( SVM) with better non-linear generalizationability, has higher fitting and prediction precision. Now it is widely used in equipment condition trend prediction. In order to further improve its prediction accuracy, a new trend prediction method combined with full vector spectrum technology based on information fusion homologous with a same source was proposed—full vector LS-SVM. This method was used of full vector spectrum technology to fuse dual-channel information to ensure integrity of LS-SVM prediction data feature extraction compared to the traditional single-channel signal extraction methods, which improved prediction accuracy. The method is applied to predict the vibration data of No. 1 steam turbinein in a power plant, and the experimental results show that full vector LS-SVM has higher prediction accuracy.%在支持向量机(SVM)基础上拓展出的最小二乘支持向量机(LS-SVM)非线性泛化能力更好,具有较高的拟合和预测精度,目前被广泛应用于设备状态趋势预测中。为进一步提高其预测精度,结合基于同源信息融合的全矢谱技术提出一种新的趋势预测方法———全矢LS-SVM。该方法采用全矢谱技术融合双通道信息,相比传统单通道信号提取方法,保障LS-SVM预测数据特征提取的完整性,提高预测精度。将该方法应用于某电厂1号汽轮机振动数据的预测,实验结果表明,全矢LS-SVM方法具有较高的预测精度。
展开▼