Research On Continuous Casting Secondary Cooling Optimization Method Based On Improved PSO

Since the steel industry has been developing fast recently, the constant production of casting slab with zero defects gradually becomes the main trends in the field of casting technology. The research on the secondary cooling plays an important role on solidification of slabs and ensuring slabs quality, for the quality of slabs is highly related to the process of the solidification and solidification is finished in secondary cooling. All these years, the study on the slab quality mostly relied on different kinds of testing methods and numerical simulation techniques. With the development of computer science, numerical simulation techniques, which could provide convenience for the making and optimizing of the institution of secondary cooling, has become a major method of research on the secondary cooling.In this paper, we first build a micro-unit model to simulate the process of heat transmission of the slabs. Then we use numerical difference method to analyze the model and the method of equivalent specific heat to analyze the latent heat of solidification, after which comes the temperature field under particular circumstances. Through the simulation experiment, the accuracy and efficiency of this model we build have been tested and verified, which provide a solid basis for the optimizing of secondary cooling.To deal with the premature of PSO, we propose an improved version, chaotic PSO, after efforts of intensive study. In this algorithm, whether the optimizing would be implemented depends on the states of every single particle. When the particle is in motion state, the algorithm would do nothing to accelerate convergence; when the particle is in stable state, chaotic optimizing would help it to avoid a local optimal solution. Based on that modification on PSO, we also adopt a self-adaption method for the choice of inertia coefficient. Therefore, when the particle is close to the optimal value, the algorithm would choose a small inertia coefficient to enhance search accuracy; when the particle is not, a big inertia coefficient would be chosen to make particle approach optimal value faster. Through the comparison of simulation experiment results of different optimization algorithms, we conclude that ACPSO proposed by this paper has strong stability, high computational accuracy, fast convergence rate and high convergence accuracy.In this paper, we build an optimization model for secondary cooling, which is based on the metallurgical criteria of secondary cooling function and the model of slabs’ heat transmission. It is also optimized by the ACPSO. After simulation, it has been verified that the improved ACPSO has an excellent effect solving problems of secondary cooling optimization.