本文对具有大纯时延、大惯性、不确定性的链条锅炉燃烧被控对象,提出一种基于自校正多步预报器的燃烧智能控制系统。它不仅具有比Smith预估补偿更好的性能,而且具有较强的适应性和鲁棒性,有较高的控制精度。大量仿真实验证明了它的正确性和有效性。 In this paper, a controlled boiler is burned by a chain boiler with large pure time delay, large inertia and uncertainty, and a combustion intelligent control system based on self-tuning multi-step predictor is proposed. It not only has better performance than Smith estimated compensation, but also has strong adaptability and robustness and high control precision. A large number of simulation experiments prove its correctness and effectiveness.