Sintering end point soft measurement method based on dynamic slow feature regression

The invention discloses a sintering end point soft measurement method based on dynamic slow feature regression, and belongs to the field of industrial process soft measurement modeling. The method comprises the following steps: firstly, reading related parameters generated by a sintering system from a database, and obtaining sintering end point data according to a quadratic curve fitting method; then, a plurality of slow features are extracted from the sintering data through dynamic slow feature analysis, and the extracted slow features are evaluated and quantified to screen the slow features used for establishing a soft measurement model; and finally, establishing a soft measurement model between the slow feature and the sintering end point by adopting a partial least square method. According to the method, the dynamic nature of the sintering process data is mined, the soft measurement precision of the sintering end point is improved, meanwhile, the prediction result has certain smoothness, the characteristi