Research On Optimal Control Of Sintering End Point Of Sintering Machine Based On Big Data Technology

The sintering end point is an important process parameter in the sintering production process.Whether its position is reasonable and stable is the key to ensuring high-quality sintered ore in the blast furnace.However,due to the dynamic characteristics of the sintering process and the inability to directly detect the sintering end point,the sintering end point control has been a weak link in the sintering process control for a long time.It is extremely important for improving the quality of sintered ore.By studying the influencing factors of the sintering end point and the sintering process,the main control parameters of the sintering end point were determined,and the sintering end point control plan was designed;13360 sets of parameter data related to the subject of the Chenggang No.2 sintering machine were collected,and big data was used.The technology has completed all data preprocessing operations.The combination of process theory and big data mining technology is used to intelligently select all the characteristic variables related to the permeability of the material layer;based on the deep neural network,random forest and support vector machine,the permeability prediction model of the material layer is constructed.The test results prove that the deep neural network has a good predictive effect;in addition,the ideal control range of the air permeability of the material layer and its corresponding core operating parameters are given.A soft-sensing model of sintering end point was established by mathematical methods to realize online judgment of sintering end point;a sintering end point controller was designed using fuzzy control theory,and by introducing a weighting factor,the online self-adjustment of fuzzy control rules was realized and the fuzzy The dynamic adaptive performance of the controller.Figure 21;Table 9;Reference 65...