Neural network model predictive control of core power of Qinshan nuclear power plant based on reinforcement learning

被引：2

作者：

Lv, Wei ^{[1
,2
]}

Chen, Jie ^{[3
]}

Tong, Li ^{[1
,2
]}

Liu, Yongchao ^{[1
,2
]}

Tan, Sichao ^{[1
,2
]}

Wang, Bo ^{[1
,2
]}

He, Zhengxi ^{[3
]}

Tian, Ruifeng ^{[1
,2
]}

Shen, Jihong ^{[4
]}

机构：

[1] Harbin Engn Univ, Coll Nucl Sci & Technol, Heilongjiang Prov Key Lab Nucl Power Syst & Equipm, Harbin 150001, Peoples R China

[2] Harbin Engn Univ, Key Lab Nucl Safety & Adv Nucl Energy Technol, Sch Nucl Sci & Technol, Harbin 150001, Peoples R China

[3] Nucl Power Inst China, Chengdu 610213, Peoples R China

[4] Harbin Engn Univ, Sch Math Sci, Harbin 150001, Peoples R China

来源：

ANNALS OF NUCLEAR ENERGY | 2024年 / 207卷

关键词：

Power control; Model predictive control; Reinforcement learning; Neural network; INTELLIGENT CONTROL; CONTROL-SYSTEM; OPERATION; OPTIMIZATION; ALGORITHM; DESIGN;

D O I：

10.1016/j.anucene.2024.110702

中图分类号：

TL [原子能技术]; O571 [原子核物理学];

学科分类号：

0827 ; 082701 ;

摘要：

The power change of Pressurized water reactors (PWRs) is a highly complex nonlinear coupling process, which requires the power controller to have a high nonlinear control capability. Since the traditional model predictive control (MPC) algorithm has the problems of insufficient robustness and adaptivity, this study proposes a neural network-based neural network model predictive control method based on model predictive control and optimizes its control parameters in real time by using the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm. The designed control method is validated on the simulation model of the Qinshan nuclear power plant (NPP). Simulation results show that the proposed control method can realize effective control of power under a variety of power changing conditions and dynamic adjustment of control parameters.

引用

页数：15

共 50 条

[1] Deep-Neural-Network-Based Economic Model Predictive Control for Ultrasupercritical Power Plant
Cui, Jinghan
Chai, Tianyou
Liu, Xiangjie
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2020, 16 (09) : 5905 - 5913
[2] Multi-model Predictive Control based on Neural Network and Its Application in Power Plant
Hou, Guolian
Bai, Xu
Zhang, Jinfang
Zhao, Zhilong
2014 11TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION (WCICA), 2014, : 4379 - 4383
[3] Study on reactor power change and ambiguous control of third Qinshan nuclear power plant
Wang, Gong-Zhan
Gu, Jun
Hedongli Gongcheng/Nuclear Power Engineering, 2006, 27 (02): : 9 - 13
[4] Adaptive neural model-based predictive control of a solar power plant
Gil, P
Henriques, J
Carvalho, P
Duarte-Ramos, H
Dourado, A
PROCEEDING OF THE 2002 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-3, 2002, : 2098 - 2103
[5] Multivariable Integrated Model Predictive Control of Nuclear Power Plant
Xia, Guoqing
Su, Jie
Zhang, Wei
2008 SECOND INTERNATIONAL CONFERENCE ON FUTURE GENERATION COMMUNICATION AND NETWORKING SYMPOSIA, VOLS 1-5, PROCEEDINGS, 2008, : 408 - 411
[6] Following calculations for core operation of Qinshan 300MW nuclear power plant
Research Inst of Nuclear Power, Operation, Wuhan, China
Hedongli Gongcheng, 3 (205-210, 225):
[7] DESIGN AND CONSTRUCTION OF THE QINSHAN NUCLEAR-POWER-PLANT
YU, OY
TRANSACTIONS OF THE AMERICAN NUCLEAR SOCIETY, 1985, 49 : 294 - 303
[8] Commissioning test programme of Qinshan Nuclear Power Plant
Sun, Guangdi
Fan, Yongfu
Chen, Shenglin
Zhu, Kezhang
Hedongli Gongcheng/Nuclear Power Engineering, 1993, 14 (01): : 11 - 18
[9] Diagnosis of Nuclear Power Plant Based on Probabilistic Neural Network
Xie, Chun-ling
APPLIED SCIENCE, MATERIALS SCIENCE AND INFORMATION TECHNOLOGIES IN INDUSTRY, 2014, 513-517 : 3992 - 3995
[10] Review of nuclear power plant control research: Neural network-based methods
Zhou, Gang
Tan, Da
ANNALS OF NUCLEAR ENERGY, 2023, 181

← 1 2 3 4 5 →