Research On Dynamic Variable Specifications Of Hot Strip Mill Based On Improving Particle Swarm Optimization Algorithm

Hot strip mill has high production efficiency, good economic returns and the development in steel rolling production very quickly. Hot rolling strip steel industry as one of the most important steel products, which has a large potential market in China and the world. In the traditional rolling process,if the change in rolling mill will slow down the speed regulation, and then adjust the roll gap and rolling speed in order to adapt to the new rolling conditions. Although the traditional rolling method can achieve the purpose of rolling, but time-consuming and unable to meet a large number of industrial demand for steel products.In view of the above problems, FGC rolling will be applied in hot strip mill, which ensures that the hot strip production, and also meet the requirements of full continuous rolling. Therefore, this method has the economic effect, and the advantages of high efficiency. The model is established according to the FGC control, and the simulation research on it.The traditional particle swarm algorithm is very easy to fall into local minima in the optimization process, resulting in the optimization result deviation. In order to decrease the probability of it fall into local optimum, through research and comparison, subject to improvement on the basic particle swarm algorithm, using the method of nonlinear dynamic inertia weight to optimize the parameters of particle swarm. The improved algorithm greatly strengthen the searching ability, not only improve the optimization accuracy, but also enhance the speed of convergence.In addition, the looper system of rolling mill is a double input and double output multivariable coupling system, with the rolling processing, mill roll gap and rolling speed will produce deviation.Therefore, in order to reduce the coupling degree of the looper height and tension, need the decoupling control of the looper system, the study by using the improved particle swarm optimization algorithm for the decoupling control of looper height and tension, then model and simulate the system after decoupling.