The Optimization Method for the Rolling Operation of Hot Strip Mills Based on Improved Genetic Algorithms

摘要

During different rolling condition of hot continuous rolling, because of lots of infections and limitations of the load allocation for finishing rolling mill, reasonable load allocation is very difficult. In order to overcome the shortcoming of the standard genetic algorithms, an improved genetic algorithm is proposed. The core of this algorithm is to add adaptive crossover operator and mutation operator during the genetic algorithm operation, so the process of learning speeds up while its astringency is not affected. Based on the traditional empirical press allocation, the IGA optimizes the load allocation of finishing rolling mill, and the algorithms and flow chart of the optimized load allocation are provided. During the typical hot continuous rolling operation, this optimization method also selected suitable parameter and carried on simulation. The simulation results show good performances. It makes good use of finishing rolling mill gaining suffice press capacity, and also meet performance request of shape and thickness of hot strip. The hot continuous rolling system applying this optimization method possesses strong robust and practicality.