Research on deterioration trend of primary pipeline in nuclear power plant based on data-driven modeling

QIAN Hong; WANG Jianqi; XU Bangzhi; SU Yujun

doi:10.19666/j.rlfd.202110204

English Version Reader LoginJournal Login

您当前的位置：

首页 >

文章列表页 >

Research on deterioration trend of primary pipeline in nuclear power plant based on data-driven modeling

更新时间：2025-09-28

- Research on deterioration trend of primary pipeline in nuclear power plant based on data-driven modeling
- Thermal Power Generation Vol. 51, Issue 6, Pages: 82-88(2022)
- 作者机构：
  
  1. 上海电力大学自动化工程学院
  2. 上海市电站自动化技术重点实验室
  3. 核电安全监控技术与装备国家重点实验室
- 作者简介：
- 基金信息：
- DOI：10.19666/j.rlfd.202110204
  CLC： TM623
- Published：2022
- 稿件说明：
移动端阅览
QIAN Hong, WANG Jianqi, XU Bangzhi, et al. Research on deterioration trend of primary pipeline in nuclear power plant based on data-driven modeling[J]. Thermal power generation, 2022, 51(6): 82-88.
DOI：

QIAN Hong, WANG Jianqi, XU Bangzhi, et al. Research on deterioration trend of primary pipeline in nuclear power plant based on data-driven modeling[J]. Thermal power generation, 2022, 51(6): 82-88. DOI： 10.19666/j.rlfd.202110204.

摘要

提出一种基于管道状态数据的分析方法，可得到管道劣化趋势，有助于预防管道劣化引起的泄漏。具体方法是通过融合主观专家经验层次分析法和客观大数据分析熵权法的优点，定量地构建一回路管道的改进马氏距离健康度模型，进而应用融合卷积神经网络和长短期记忆神经网络的组合预测模型，对一回路管道未来的劣化泄漏情况做出趋势分析和预判。仿真结果表明，构建的健康度模型能够较好的反映出一回路管道的真实健康状态；组合预测模型可以有效地预测一回路管道劣化泄漏的趋势，为保障设备的安全稳定运行提供参考。

Abstract

An analysis method based on pipeline state data is proposed

which can get the pipeline deterioration trend and help to prevent the leakage caused by pipeline deterioration. The specific method is to quantitatively construct the health model of the improved Mahalanobis distance of the primary pipeline by integrating the advantages of the analytic hierarchy process of subjective expert experience and the entropy weight method of objective big data analysis. Then

the combined prediction model of convolution neural network and long-term and short-term memory neural network is applied to analyze and predict the future deterioration and leakage of the primary pipeline. The simulation results show that

the health model constructed in this paper can better reflect the real health state of the primary pipeline. The combined prediction model can effectively predict the trend of deterioration and leakage of primary pipeline

which provides a reference for ensuring the safe and stable operation of equipment.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

WAYS FOR IMPROVING POWER-GENERATING CAPABILITY OF NUCLEAR POWER UNITS

MEASUREMENT AND CALCULATION OF THE WATER-LEVEL IN PRESSURE STABILIZER OF ONE LOOP IN NUCLEAR POWER STATION

PROBLEM AND SOLUTION OF CONDENSED WATER REATING SYSTEM IN A NUCLEAR POWER STATION

DESIGN OF PRESSURE CONTROL STRATEGY FOR DEAERATOR IN NUCLEAR POWER STATION BASED ON DIGITALIZED CONTROL PLATFORM

RIGOROUS METHOD BASED PRIMARY FREQUENCY REGULATION MODEL FOR STEAM TURBINE IN NUCLEAR POWER UNIT

Related Author

LIU Dao-heDayabay Nuclear Power Operation and Management Co Ltd

CAI Wei-dong

ZHANG Da-yong

CHI Zhi-yuan

LI Yuan Dayabay Nuclear Power Operation Management Co Ltd

ZHU Xing- bao The SUE Department of CNPEC

ZHOU Peng

WANG Bo

Related Institution

Daya Bay Nuclear Power Operation Management Co., Ltd

CGN Engineering Co., Ltd

Xuzhou Huarun Pengcheng Power Plant

Department of Automatic Control, Shanghai Electric Power University

School of Power and Mechanical Engineering, Wuhan University

AI问答

⁰