摘要
为克服单一模型预测精度较低这一缺陷,提出一种基于灰色模型(grey model,GM)和最小二乘支持向量机(least squares support vector machine,LSSVM)的组合预测方法。通过灰色累加对原始数据序列进行处理,建立灰色预测模型,利用灰色预测模型的预测结果作为输入,原始数据作为输出,训练构建LSSVM预测模型进行预测。选取航空发动机主燃油泵作为具体研究对象,采集排气温度作为其状态预测参数进行状态预测。研究结果表明,相比单一预测模型,灰色最小二乘支持向量机预测精度更高,为航空发动机状态预测提供了一种有效的解决途径。
A condition combination forecasting method based on grey model（GM）and least squares support vector machine（LSSVM）was proposed for overcoming the shortcomings of low precision of the forecasting model based on simple forecasting method.The origin data were processed with grey accumulation to build grey forecasting method,and the forecasting results of grey forecasting model were used as input,the original data sequence was used as the output,the least squares support vector machine forecasting model were trained and established to forecast.The aeroengine main fuel pump was selected as the specific research object,and exhaust temperature was collected as aeroengine condition forecasting parameters for condition prediction.The research results show that compared with the simple forecasting method,the proposed model has higher prediction accuracy,which provides an effective way for the problem of aeroengine condition prediction.
出处
《计算机工程与设计》
北大核心
2017年第10期2809-2813,共5页
Computer Engineering and Design
基金
国防基础科研基金项目(Z052012B002)
辽宁省自然科学基金项目(2014024003)
航空科学基金项目(20153354005)
