Ławryńczuk Maciej
Warsaw University of Technology, Institute of Control and Computation Engineering, Faculty of Electronics and Information Technology, ul. Nowowiejska 15/19, 00-665 Warsaw, Poland.
ISA Trans. 2020 Apr;99:270-289. doi: 10.1016/j.isatra.2019.10.009. Epub 2019 Oct 26.
Control of Solid Oxide Fuel Cells (SOFCs) is a challenging task since they are nonlinear dynamic systems and it is essential to precisely satisfy the existing technological constraints which must be imposed on the manipulated variable (the fuel flow) and on fuel utilisation. This paper details a constrained computationally efficient nonlinear Model Predictive Control (MPC) algorithm for the SOFC process. The predicted voltage and fuel utilisation trajectories are successively linearised on-line which leads to a simple to solve quadratic optimisation MPC problem. The emphasis is put on three aspects: (a) selection of tuning parameters, (b) feasibility of the constrained MPC optimisation problem, (c) good control quality and low computational burden. At first, tuning is thoroughly described. It is demonstrated that soft fuel utilisation constraints lead to feasible MPC optimisation. It is shown that control accuracy and constraints' satisfaction ability of the algorithm are very similar to those of the "ideal" MPC strategy with nonlinear on-line optimisation, but its computational burden is much lower. Finally, it is shown that the algorithm is much more precise than the simple MPC algorithm with successive on-line model linearisation and the classical MPC algorithm based on a parameter-constant linear model.
固体氧化物燃料电池（SOFC）的控制是一项具有挑战性的任务，因为它们是非线性动态系统，精确满足现有技术约束至关重要，这些约束必须施加于操纵变量（燃料流量）和燃料利用率上。本文详细介绍了一种用于SOFC过程的带约束的计算高效非线性模型预测控制（MPC）算法。预测的电压和燃料利用率轨迹在线依次线性化，这导致了一个易于求解的二次优化MPC问题。重点放在三个方面：（a）调谐参数的选择，（b）约束MPC优化问题的可行性，（c）良好的控制质量和低计算负担。首先，对调谐进行了全面描述。结果表明，软燃料利用率约束导致可行的MPC优化。结果表明，该算法的控制精度和约束满足能力与具有非线性在线优化的“理想”MPC策略非常相似，但其计算负担要低得多。最后，结果表明该算法比具有连续在线模型线性化的简单MPC算法和基于参数恒定线性模型的经典MPC算法精确得多。
