Coordinate Optimization Method Research Of Internal Quality For Billet Crack And Shrinkage

Internal quality of billet is an important factor affecting the quality of the final product which affect the processing and performance of the product. Internal quality defects of billet, including crack, segregation, shrinkage and other, which crack is the most common quality defects. Internal quality defects of billet often have characteristics associated with each other, the routine quality control often determine solve a variety of quality defects, comprehensive quality is difficult to balanced. For this problem, in this thesis, with the cracking and shrinkage as the research object, and coordinating crack and shrinkage quality as final target, put forward the solution of coordinating shrinkage and crack base on the quality of model. The main contents are as follows:Firstly, this thesis analyses the process and factors influencing the formation of crack and shrinkage, according to the principle of the solidification front of the critical stress, criteria for the solidification front of the equivalent stress and critical stress ratio is greater than one, as resulting billet cracks; According to the end of solidification of liquid steel solid fraction is greater than the critical flow solid fraction, then feeding liquid steel will lose the ability to produce shrinkage which result amount of shrinkage, with this part of the amount of shrinkage as the criterion of shrinkage quality defect. Building quality model for shrinkage and crack which combining these criterions and the stress and solidification models.Secondly, building crack and shrinkage coordination optimization model based on the quality of the model. Determining coordinating optimization objective function of shrinkage and crack, control variables, optimization methods. This thesis applies collaborative optimization method, for the problem of system-level consistency constraints of collaborative optimization method causing computational difficulties, this thesis put forward an improved method of punishment based on dynamic factors. With Particle Swarm Optimization(PSO) as the optimizer, but PSO is easy to fall into local optimum and time-consuming, this thesis put forward adaptive PSO updates inertia weight based on stable second-order system, which improved PSO efficiency.Finally, for there is middle crack and center shrinkage issue during production of Q235 steel in a steel company, by using the corrected quality model and coordination optimize system to improve the quality of its crack and shrinkage. Adjust the speed and optimized separately, using linear regression after get two cold amount water to build water model, and comparing with the water which metallurgists adjusted in site, the amount of water for first two segments is basically consistent with the actual adjustment after the model optimization, the amount of water for three segment is large than actual the maximum difference of 6.2% of the actual amount of water, come out of the water model optimization in line with reality.