Data-driven Coordinated Control for a 600 MW Coal-Ammonia Co-firing Unit

To address the issues of severe fluctuations in main steam pressure and temperature caused by increased model uncertainties of the controlled object in the ammonia co-firing coordinated control system, a data-driven adaptive integral sliding mode control (DAISMC) method was proposed. First, based on the particle swarm optimization (PSO) algorithm, a detailed model of the controlled object in the coordinated control system was established to explicitly characterize the ammonia co-firing characteristics. Second, discrete-time prediction technology was utilized to implicitly express the time-delay terms in the original system model. Simultaneously, a data-driven controller considering disturbance terms was designed by combining model-free adaptive control (MFAC) with discrete sliding mode control (DSMC). Finally, using a transfer function model and a neural network model constructed from actual field data. Results show that the proposed control method has minimal model dependence, along with high simulation accuracy and strong robustness.