基于粒子群算法的反应堆功率调节系统优化研究  

Research on Reactor Power Control System Optimization Based on Particle Swarm Optimization

作　　者：黄轲 王琳[1] 肖凯[1] 李羿良 陈冠宇 HUANG Ke;WANG Lin;XIAO Kai;LI Yi-liang;CHEN Guan-yu(Nuclear Power Institute of China,Chengdu Sichuan 610041,China)

机构地区：[1]中国核动力研究设计院核反应堆系统设计技术重点实验室

出　　处：《科技视界》2019年第10期64-66,70,共4页Science & Technology Vision

摘　　要：核动力装置具有时变、非线性等特点,对象特性复杂,控制难度大。反应堆功率调节系统普遍采用前馈-串级PI控制方案,工程上一般采用的整定方法适用范围较窄,且较依赖工程人员经验。粒子群优化算法具有算法简单,收敛速度快的优点,广泛应用于各类非线性最优化问题,但是存在早熟收敛的缺点。本文提出一种自适应惯性权重变化策略,并利用基于Simulink的反应堆及一回路传递函数模型进行功率调节系统控制参数寻优。将寻优结果与通过临界比例度法得到的参数进行控制效果对比。仿真结果显示:采用改进的粒子群算法得到的控制参数能减小核功率超调和反应堆平均温度的稳态误差,提高控制品质。Nuclear power plant has the characteristics of time-varying and non-linearity.Its object characteristics are complex and difficult to control.Feedforward-cascade PI control scheme is widely used in reactor power cobtrol system.The setting method commonly used in engineering has a narrow scope of application and relies on the experience of engineers.Particle swarm optimization( PSO) has the advantages of simple principle and fast search speed which is widely used in various kinds of nonlinear optimization problems.But PSO has the disadvantage of premature convergence.In this paper,an adaptive inertia weight change strategy is proposed.The control parameters of the power control system are optimized by using the Simulink-based model of the reactor and the primary loop transfer function.The optimization results are compared with the parameters obtained by the critical scale method.The simulation results show that the control parameters obtained by the improved particle swarm optimization algorithm can reduce the nuclear power overshoot and the steady-state errors of average temperature of reactor,and improve the control quality.

关 键 词：功率调节系统 粒子群算法 传递函数模型 SIMULINK 

分 类 号：F275.5[经济管理—企业管理]