It is challenging and crucial to achieve unbiased tracking control for parabolic trough collector field as it is vulnerable to various types of disturbances or uncertainties such as unmeasured external disturbances, parameter perturbation and model mismatch. To solve this issue, an optimal model predictive rejection control strategy is put forward in a composite designed manner, in which all disturbances/uncertainties are dealt with as lumped disturbances. A generalized extended state observer is firstly employed to estimate the lumped disturbances, and then a feedback controller is devised based on optimal model predictive control to compensate the influences of the lumped disturbances on output. Stability analysis of the closed-loop system has been presented. It shows that the proposed composite controller can track given references without offset in the presence of lumped disturbances while not sacrificing its nominal performance in the absence of disturbances. Simulations conducted on a numerical example and a practical application for parabolic trough collector validate our conclusions.

对于抛物线槽收集器场实现无偏跟踪控制是极具挑战性和至关重要的，因为它容易受到各种类型的干扰或不确定因素的影响，例如无法测量的外部干扰，参数扰动和模型失配。为解决这一问题，提出了一种综合设计的最优模型预测拒绝控制策略，将所有干扰/不确定性作为集中干扰进行处理。首先采用广义扩展状态观测器估计集总扰动，然后基于最优模型预测控制设计反馈控制器，以补偿集总扰动对输出的影响。提出了闭环系统的稳定性分析。它表明，所提出的复合控制器可以在存在集中干扰的情况下跟踪给定的参考而不会发生偏移，同时在不存在干扰的情况下也不会牺牲其标称性能。在数值示例上进行的仿真和抛物线槽收集器的实际应用验证了我们的结论。