【作者】 周曙光； 孟庆虎； 孟庆丰；

【Author】 Zhou Shuguang Meng Qinghu Meng Qingfeng Mechanical Engineering School,Xi’an Jiao tong University,Xi’an Shaanxi 710049

【机构】 西安交通大学机械工程学院；

【摘要】 核岛主泵作为核电站的心脏,对其可靠性和安全性要求特别高。但在对其进行故障诊断和状态预测时,却常常缺乏大量的故障样本做参照,制约了相关技术的发展。基于统计学习理论的支持向量机方法正好克服了这方面的不足。本论文构建了LSSVM和时间序列模型相结合的支持向量机预测模型,并分别用仿真数据和现场数据对其进行了验证。然后将该支持向量回归机应用于核岛主泵振动信号的预测,在对现场数据的预测中取得了较好的效果,表明该算法对核岛主泵的运行状态趋势具有较好的预测能力。更多还原

【Abstract】 As the heart of a nuclear power plant,there are high requirements in reliability and security of the reactor coolant pump.But developments of the relating technologies are restricted as there are not large numbers of fault samples as references when the faults are diagnosed and the future states are forecasted.The Support Vector Machine which is based on Statistical Learning Theory could solve this problem.In this paper a forecast model which combines LSSVM and Time series model is constructed.Meanwhile,the performance of LSSVM is verified by simulation data and field data.Then LSSVM is used to predict vibration signals of reactor coolant pump and it has achieved good results in forecasting field data.Therefore,LSSVM has a good capacity to forecast running trends of the reactor coolant pump.更多还原