Modeling Method for Performance Prediction of Centrifugal Compressor Based on Broad Learning System

摘要

An accurate model is very important for the design and control of centrifugal compressor. This paper proposes a rapid modeling method for centrifugal compressor based on broad learning system. The method can be used for off-line modeling of compressor performance prediction. When new data samples enter into the model, this paper uses incremental learning algorithm to realize fast online modeling of compressor. Furthermore, simulation results show that the proposed method can achieve a good prediction precision and can be used in compressor performance prediction online.