基于MTICC-DiPCA的磨煤机多模态运行过程监测方法OA

Under the two-way perturbation of load and fuel,the multimode characteristics of coal mill operating parameters are significant,and it is more difficult to diagnose its faults.To address this problem,this paper proposes a multimodal operation process monitoring method for coal mills based on Mixed Toeplitz Inverse Covariance and Dynamic Inner Principal Compo-nent Analysis(MTICC-DiPCA).The method extracts the spatial and temporal dependencies of the data within a continuous time window using Mixed Toeplitz Inverse Covariance to obtain interpretable mode segmentation results.Dynamic Inner Principal Component Analysis is used to extract the dynamic and static latent variables from the data of each mode,and the combined indexes of Squared Predication Error(SPE)and T2 are utilized to monitor the coal mill opera-tion process in real time.Based on Matlab/Simulink software,a simulation model of MPS-type medium-speed coal mill is constructed,and the simulation data of coal mill operation under fuels with varying moisture content are obtained to analyze the influence of fuel characteristics on the operating modes of the coal mill.The effectiveness of the MTICC-DiPCA method of this paper is verified through the simulation fault data,and the results show that the diagnostic accuracy of this method is significantly better than that of the single model method.