Optimization Method for Cotton Production Process Based on Hierarchical Clustering

With the increasingly large-scale, complicated and modernized textile, thermoelectric and other process industries, a large number of data are related to materials, energy, process equipment and operation information which have been generated. Dig deep production knowledge, optimal operation conditions and management mode from a large number of production and management data, namely, process industrial data mining. The cotton processing a variety of frequencies in the history of data processing and optimization of removal of impurity of cotton process which are designed a hierarchical clustering method and mathematical modeling for each process the change tendency of the measuring point data. The prediction and simulation of various cotton shaker data at different frequencies are carried out, and the implicit knowledge in the data is obtained, which is used to assist the cotton mill to adjust the parameters such as the speed of the impurity machine. The steps of this method include: key parameters, distribution statistics, correlation mapping, and optimization target determination. In this way, it is optimized to use different frequencies to remove impurities for different kinds of cotton, so as to improve the effect of removing impurities from cotton.