基于字典序优化的核反应堆功率预测控制  

Model Predictive Control of Nuclear Reactor Power Based on Lexicographic Optimization

作　　者：姜頔[1] 刘向杰[1] JIANG Di;LIU Xiang-jie(State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, North China Electric Power University, Beijing 102206, China)

机构地区：[1]华北电力大学新能源电力系统国家重点实验室

出　　处：《控制工程》2018年第4期577-586,共10页Control Engineering of China

基　　金：国家自然科学基金(61673171,61533013)

摘　　要：压水反应堆口WR）功率系统是受约束的本质非线性系统，其负荷跟踪能力是核电机组安全高效运行的重要保障。模型预测控制（MPC）具有内在的约束处理能力，是PWR功率控制的有效方法，采用PWR非线性模型，设计了基于字典序优化的非线性MPC（NMVC）。首先，将PWR功率控制问题转化为具有不同目标优先级的字典序优化问题。其次，考虑控制棒执行机构精度，采用粒子群优化（Ps0）高效求解NMPC。最后，将该控制器应用于PWR的负荷跟踪仿真，结果表明了算法的有效性。The pressurized water reactor （PWR） power system is a constrained essential nonlinear system, and its load tracking ability is an important guarantee for the safe and efficient operation of PWR. Due to the inherent constraint handling ability, model predictive control （MPC） has been a quite effective way of controlling the power level of PWR. Based on the nonlinear model of PWR, a nonlinear MPC （NMPC） is presented using lexicographic optimization method. Firstly, the PWR power control problem is reformulated as the lexicographic one that can deal with the different prioritization of objectives. Then, considering the accuracy of the control rod actuator, particle swarm optimization （PSO） algorithm is used to solve NMPC online efficiently. Finally, this controller is applied to the load following power-level regulation of PWR. The numerical simulation results show the effectiveness of the proposed controller.

关 键 词：压水反应堆 功率控制 模型预测控制 字典序优化 粒子群优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]