上海交通大学学报

上海交通大学学报（自然版） ›› 2014, Vol. 48 ›› Issue (02): 210-213.

• 原子能技术 • 上一篇    下一篇

基于多参数耦合的船舶核动力装置参数重构

王晓龙，蔡琦，陈玉清
  

  1. （海军工程大学，动力工程学院， 武汉 430033）
     
  • 收稿日期:2012-10-12 出版日期:2014-02-28 发布日期:2014-02-28

Abnormal Parameter Reconstruction for Marine Nuclear Power Plant Based on MultiSignal Coupling

WANG Xiaolong,CAI Qi,CHEN Yuqing
  

  1. (Naval University of Engineering, Power Engineering College, Wuhan 430033, China)
  • Received:2012-10-12 Online:2014-02-28 Published:2014-02-28

摘要：

针对船用核动力装置在正常运行瞬态、异常瞬态和事故瞬态时,水位类参数易于虚假显示,以及因测量通道、测量仪表故障导致的参数丧失问题，提出了一种基于多信号耦合的重构方法.该方法根据装置参数间的强耦合关系，利用历史运行数据，通过数理统计和人工智能技术，针对装置不同运行工况建立重构数据模型对丧失信号进行重构.试验表明，在装置正常运行瞬态工况时，该方法可以快速、准确地获得重构结果，适于在线使用.
 
 

关键词: 船用核动力装置, 多信号耦合, 支持向量回归, 稳压器

Abstract:

Because of the harsh environment of operating conditions of marine nuclear power plant, some parameters was often lost due to failures of measuring instrument and measurement of transmission channel, which influenced the operator’s judgement about the running states of the system and further jeopardize the operation safety of the system. Therefore, this paper put forward a reconstruction method based on multisignal coupling. This method reconstructed the lost signals with a reconstruction data model designed according to the strong coupling of important parameter and the running historical data of nuclear power plant by the method of artificial intelligence and data statistics. Tests show that this method can reconstruct the lost signal fast and efficiently in normal operation transient conditions of nuclear power plant, and is fit for being used online.
 

Key words: ship nuclear power plant, multi-signal coupling, support vector regression, pressurizer

中图分类号: