Working condition recognition of grate cooler based on an improved K-means clustering algorithm

The working condition control of grate cooler always affects the efficiency of the whole cement process. In order to solve the working condition of grate cooler too parameter dimension and strong coupling problem, and according to the actual production of collecting the data of grate cooler used principal component analysis combined with an improved k-means clustering algorithm, is used to identify the conditions of grate cooler, working condition of the formation of grate cooler model, through the calculation speed and stability analysis is made of the advantages of the new algorithm, the accuracy as an important parameter of working condition of the grate cooler model verified, it provides the model basis for the subsequent grate cooler automatic control.