Identification and application of data-driven superheated steam temperature system

Superheated steam temperature is one of the important parameters of thermal power plants, and its non-linearity and large inertia pose major challenges for modeling. This article is based on data-driven concepts and focuses on field data. Fading memory recursive least square algorithm was used to establish a desuperheater model and a superheater model based on the input and output data of the superheated steam temperature control system. In addition, three evaluation indexes are introduced to compare performance with traditional identification methods. This paper uses a multi-stage superheater simulation platform and based on the actual operating data of a power plant for verification and investigation. Practice has shown that the model identified in this paper has better accuracy.