基于孪生模型的堆芯自给能中子探测器信号异常检测

陈静; 卢燕臻; 江灏; 林蔚青; 许勇

doi:10.13832/j.jnpe.2023.03.0210

基于孪生模型的堆芯自给能中子探测器信号异常检测

doi: 10.13832/j.jnpe.2023.03.0210

陈静^{1, 2,},
卢燕臻^{1, 2},
江灏^{1, 2, ,},
林蔚青^{1, 2},
许勇^{3, 4}

1.
福州大学电气工程与自动化学院，福州，350108
2.
福州大学电力系统与装置产业研究院，福州，350108
3.
福建福清核电有限公司，福建福清，350318
4.
上海交通大学自动化系，上海，200240

基金项目: 福建省高校产学合作项目（2022H61010006）

详细信息

作者简介:
陈　静（1988—），女，副教授，现从事智能电器与设备故障检测的研究，E-mail: chenj@fzu.edu.cn

通讯作者:
江　灏，E-mail: jiangh@fzu.edu.cn

中图分类号: TL81;TM623.7
计量
- 文章访问数: 196
- HTML全文浏览量: 74
- PDF下载量: 40
- 被引次数: 0
出版历程
- 收稿日期: 2022-06-23
- 修回日期: 2023-03-23
- 刊出日期: 2023-06-15

Anomaly Detection of Core Self-Powered Neutron Detector Based on Twin Model

Chen Jing^{1, 2
,},
Lu Yanzhen^{1, 2},
Jiang Hao^{1, 2
, ,},
Lin Weiqing^{1, 2},
Xu Yong^{3, 4}

1.
College of Electrical Engineering and Automation, Fuzhou University, Fuzhou, 350108, China
2.
Research Institute of Power System & Power Equipment, Fuzhou University, Fuzhou, 350108, China
3.
Fujian Fuqing Nuclear Power Co.,Ltd., Fuqing, Fujian, 350318, China
4.
Department of Automation, Shanghai Jiao Tong University, Shanghai, 200240, China

摘要

摘要: 堆芯自给能中子探测器（SPND）的健康状态直接影响反应堆安全运行。充分考虑核反应堆内不同位置SPND间的量测关联性，提出一种基于孪生模型的堆芯SPND信号异常检测方法，利用SPND运行状态历史数据，通过随机森林回归算法（RFR）提取邻域SPND的量测信号特征，为SPND构建与其物理实体一致输出的孪生模型。孪生模型与探测器实体共生运行，通过计算探测器观测值与孪生模型估计值之间的残差，作为SPND信号异常检测判据，以此实现对单点和多点异常SPND的辨识和定位。实验表明，本文所提孪生模型的预测误差在1×10⁻¹¹量级，具备极高的输出一致性。针对多种不同SPND信号异常状态，能够获得99%以上的辨识精度，准确地判别单点及多点异常SPND，对于提高堆芯中子注量率测量系统状态监测的可靠性和安全性具有较高的参考价值。
- 自给能中子探测器（SPND） /
- 孪生模型 /
- 随机森林 /
- 异常检测
Abstract: Self-powered neutron detector (SPND) is an important nuclear sensing device in the core, whose health status affects the safe operation of the reactor directly. Considering the measurement correlation between SPNDs at different positions in the reactor, a twin model-based anomaly detection method for SPND signals in the core was proposed in this paper. The characteristics of measurement signals of neighboring SPNDs were extracted by the Random Forest Regression (RFR) algorithm based on the historical operation data of SPND, and a twin model was built for SPNDs which outputs the same as its physical entity. Twin model and entity sensing coexisted. The residual error between the actual observation value of SPND and the twin model estimation value was calculated to serve as the anomaly detection criterion, which realized the identification and location of single-point and multi-point SPND anomalies. The experiments show that the prediction error of the twin model proposed in this paper attains the order of 1×10⁻¹⁰, which has a very high output consistency. The identification accuracy of anomaly detection can reach over 99% under various abnormal states of SPND signals, and the single-point and multi-point abnormal SPND can be accurately identified, which has a high reference value for improving the reliability and safety of the state monitoring of neutron flux measurement system in the core.
- Self-powered neutron detector (SPND) /
- Twin model /
- Random forest /
- Anomaly detection

HTML全文

图 1 SPND组成结构

Figure 1. SPND Structure

下载: 全尺寸图片幻灯片

图 2 SPND 信号异常检测流程图

Figure 2. Flow Chart of Anomaly Detection of SPND Signals

下载: 全尺寸图片幻灯片

图 3 U₁(A)关系网络

Figure 3. Relationship Network of U₁(A)

下载: 全尺寸图片幻灯片

图 4 停机期间SPND电流变化

Figure 4. Change of SPND Current During Shutdown

下载: 全尺寸图片幻灯片

图 5 单点异常检测结果

Figure 5. Single Point Anomaly Detection Results

下载: 全尺寸图片幻灯片

图 6 多点异常检测准确率

Figure 6. Multi-Point Anomaly Detection Accuracy

下载: 全尺寸图片幻灯片

表 1 数据模式划分结果

Table 1. Data Schema Partitioning Results

多维特征变量	标签变量
(s₁, s₂, s₃$, \cdots , $ s_n−2 ,s_n−1)	s_n
(s₁, s₂, s₃$, \cdots , $s_n−2, s_n)	s_n−1
(s₁,s₂,s₃$, \cdots , $s_n₋₁,s_n)	s_n−2
$\cdots $	$\cdots $
(s₂,s₃,s₄$, \cdots , $ s_n−1,s_n)	s₁

下载: 导出CSV

表 2 44 组回归模型 2 项参数最优选值

Table 2. Optimal Values of 2 Parameters in 44 Regression Models

模型号	CART数量	最大特征数	模型号	CART数量	最大特征数
1	14	12	23	11	16
2	20	12	24	17	36
3	10	31	25	10	34
4	16	29	26	20	24
5	19	36	27	17	16
6	12	18	28	10	35
7	10	28	29	12	34
8	10	36	30	19	32
9	17	32	31	10	20
10	12	24	32	14	22
11	11	36	33	15	26
12	10	28	34	17	22
13	11	14	35	15	16
14	11	40	36	10	27
15	11	33	37	15	20
16	11	24	38	14	26
17	12	25	39	14	29
18	13	28	40	12	26
19	12	20	41	15	38
20	12	36	42	14	36
21	15	14	43	13	20
22	11	31	44	13	30

下载: 导出CSV

参考文献(15)

[1]	李树成,胡铸萱. CAP 1400核电厂堆芯钒自给能中子探测器设计与验证[J]. 核电子学与探测技术,2018, 38(5): 699-702.
[2]	黄有骏,李文平,杨戴博,等. 华龙一号堆芯中子注量率测量系统自给能中子探测器关键参数研究[J]. 核动力工程,2020, 41(S2): 45-49. doi: 10.13832/j.jnpe.2020.S2.0045
[3]	YELLAPU V S, TIWARI A P, DEGWEKER S B. Application of data reconciliation for fault detection and isolation of in-core self-powered neutron detectors using iterative principal component test[J]. Progress in Nuclear Energy, 2017, 100: 326-343. doi: 10.1016/j.pnucene.2017.04.017
[4]	GAUTAM S, TAMBOLI P K, ROY K, et al. Sensors incipient fault detection and isolation of nuclear power plant using extended Kalman filter and Kullback–Leibler divergence[J]. ISA Transactions, 2019, 92: 180-190. doi: 10.1016/j.isatra.2019.02.011
[5]	YELLAPU V S, ZHANG W D, VAJPAYEE V, et al. A multiscale data reconciliation approach for sensor fault detection[J]. Progress in Nuclear Energy, 2021, 135: 103707. doi: 10.1016/j.pnucene.2021.103707
[6]	CHATTERJEE S, SURWASE S K, BHUSHAN M, et al. Cluster statistics based normalization for online fault diagnosis of self-powered neutron detectors[C]//2017 International Conference on Information, Communication, Instrumentation and Control (ICICIC). Indore: IEEE, 2017: 1-6.
[7]	ZAREI M. A Kalman filtering of the Self-Powered Neutron Detector current to determine excess reactivity[J]. Radiation Physics and Chemistry, 2021, 188: 109602. doi: 10.1016/j.radphyschem.2021.109602
[8]	VOLODIN V S, TOLOKONSKII A O. Concept of instrumentation of digital twins of nuclear power plants units as observers for digital NPP I&C system[C]//8th International Conference on Mathematical Modeling in Physical Science. Bratislava: IOP, 2019: 012083.
[9]	KOCHUNAS B, HUAN X. Digital twin concepts with uncertainty for nuclear power applications[J]. Energies, 2021, 14(14): 4235. doi: 10.3390/en14144235
[10]	NGUYEN T N, PONCIROLI R, BRUCK P, et al. A digital twin approach to system-level fault detection and diagnosis for improved equipment health monitoring[J]. Annals of Nuclear Energy, 2022, 170: 109002. doi: 10.1016/j.anucene.2022.109002
[11]	胡梦岩,孔繁丽,余大利,等. 数字孪生在先进核能领域中的关键技术与应用前瞻[J]. 电网技术,2021, 45(7): 2514-2522. doi: 10.13335/j.1000-3673.pst.2021.0335
[12]	AKS A,JSK B,LV B. Digital twin in energy industry: Proposed robust digital twin for power plant and other complex capital-intensive large engineering systems[J]. Energy Reports, 2022, 8: 3704-3726. doi: 10.1016/j.egyr.2022.02.305
[13]	LIN L Y, ATHE P, ROUXELIN P, et al. Digital-twin-based improvements to diagnosis, prognosis, strategy assessment, and discrepancy checking in a nearly autonomous management and control system[J]. Annals of Nuclear Energy, 2022, 166: 108715. doi: 10.1016/j.anucene.2021.108715
[14]	刘皓璐,邵建伟,王雪,等. 基于数字孪生的配电自动化终端设备状态评价与故障预判[J]. 电网技术,2022, 46(4): 1605-1613. doi: 10.13335/j.1000-3673.pst.2021.0661
[15]	张西宁,张雯雯,周融通,等. 采用单类随机森林的异常检测方法及应用[J]. 西安交通大学学报,2020, 54(2): 1-8,157.