Modeling of Superheated Steam Temperature Characteristics Based on Fireworks Algorithm Optimized Extreme Learning Machine

In order to establish a more accurate superheated steam temperature (SST) characteristic model and realize intelligent predictive optimal control of SST, the prediction model of SST was established with external time delay extreme learning machine (TDELM) method using the historical operation data acquired from DCS of a 600MW subcritical coal-fired power unit. The fireworks algorithm (FWA) with adaptive inertia weight was adopted to optimize the ELM neural network parameters. By comparing the model prediction results with the traditional ELM model and FWAELM model, it is shown that the improved fireworks algorithm has faster convergence speed and higher search accuracy, and the SST characteristic model optimized by FWA has better accuracy and generalization ability.