摘要
熔盐堆具有固有安全性和部署灵活性,能够广泛应用于多种场景。功率控制系统是确保熔盐堆安全稳定运行的关键系统之一,其在控制堆芯燃料盐温度的同时,还需要对负载变化进行快速响应,以满足基于熔盐堆的多样化复合能源系统中的潜在需求。控制目标之间存在相互影响,且执行器的数量和可调范围往往有限,因此给控制器参数的选择带来了巨大挑战。本文使用反应堆功率控制系统上普遍采用的比例-积分-微分(Proportional-Integral-Derivative,PID)控制器对堆芯温度和功率进行控制。在MATLAB/Simulink软件中建立熔盐增殖堆(Molten Salt Breeder Reactor,MSBR)一回路非线性模型。基于该模型采用多目标粒子群(Multi-Objective Particle Swarm Optimization,MOPSO)算法对控制器参数进行了优化,并根据灵敏度比偏差筛选每个目标都可接受的优解,实现了自动化选取控制器参数的最佳方案。通过仿真不同工况对控制器的性能进行了验证,结果表明,多目标粒子群算法优化后的控制器具有良好的负荷跟踪能力,能够实现功率及温度双目标的平衡,在负载阶跃变化时能抑制超调量在2%范围内,响应速度快,抗干扰能力强。
[Background]Molten Salt Reactor(MSR),distinguished by its inherent safety and flexible deployment,has a wide range of applications in various scenarios.Power control system(PCS)plays a critical role in ensuring the safe and stable operation of MSRs.[Purpose]This study aims to develop an advanced controller gains tuning method for PCS to simultaneously regulate reactor fuel temperature and achieve rapid load-following performance,addressing the demands of future composite energy systems.[Methods]The Molten Salt Breeder Reactor(MSBR)was taken as research object,and the Proportional-Integral-Derivative(PID)controller was adopted to regulate the temperature and power of MSBR.Then,a nonlinear model of the MSBR primary loop of was developed using MATLAB/Simulink software,and Multi-Objective Particle Swarm Optimization(MOPSO)algorithm was applied to tuning the controller gains based on the nonlinear model.Finally,the sensitivity ratio deviation was introduced to automatically select Pareto-optimal solutions that balance competing control objectives,and the tuned PID controllers were evaluatively tested under different working conditions.[Results]The results demonstrate that the tuned PID controllers with the MOPSO exhibit better load-following capability and achieve a balance between the power and temperature objectives.The system effectively suppresses overshoot to within 2%under step-change conditions,while maintaining rapid response speed and strong anti-interference ability.[Conclusions]The proposed MOPSObased tuning method has been proved to be a viable and efficient approach for optimizing PID controllers in MSR power control systems.
作者
邱芷艺
王鹤翔
李明海
朱贵凤
郭建
刘亚芬
严睿
邹杨
QIU Zhiyi;WANG Hexiang;LI Minghai;ZHU Guifeng;GUO Jian;LIU Yafen;YAN Rui;ZOU Yang(Shanghai Institute of Applied Physics,Chinese Academy of Sciences,Shanghai 201800,China;University of Chinese Academy of Sciences,Beijing 100049,China)
出处
《核技术》
北大核心
2025年第8期171-180,共10页
Nuclear Techniques
基金
中国科学院青年创新促进会(No.2020261)
中国科学院战略性先导科技专项(No.XDA02010000)
中国科学院上海分院上海基础研究特区计划(No.JCYJ-SHFY-2021-003)资助。
关键词
熔盐堆
多目标粒子群算法
功率控制
系统仿真
Molten salt reactor
Multi-Objective particle swarm optimization
Power control
System simulation
