Real Time Monitoring Method for Combustion Process of Coal-fired Power Plant Boilers Based on Sparse Representation and Nearest Neighbor Embedding
Conventional real-time monitoring of the combustion process of coal-fired power plant boilers mainly uses wave value signal analysis to achieve real-time monitoring,ignoring the impact of signal processing on real-time monitoring,resulting in low accuracy of monitoring re-sults.Therefore,a real-time monitoring method for the combustion process of coal-fired power plant boilers based on sparse representation and nearest neighbor embedding is proposed.To add Gaussian white noise to the obtained boiler combustion process signal,sparse representation is introduced to express the added signal,obtaining a small number of special features of the signal,reducing signal processing workload,and converting it into image information.Analyze the characteristics of boiler combustion flame images,use nearest neighbor embedding to recon-struct the combustion process image,and obtain real-time monitoring information.The experimental results show that the fluctuation points of the monitoring results obtained after the application of the proposed method are highly consistent with the actual results,with high accuracy,and meet the practical needs of combustion control work in coal-fired power plant boilers.
coal-fired power plantsboiler combustionmonitoring of combustion processsparse representationnearest neighbor embed-dingreal time monitoring methods
万方数据
维普
