The Optimization Of The Power Supply Curve For Electric Arc Furnace Based On Genetic Algorithm

In steel-making process with Electric Arc Furnace (EAF), effective power supply strategy will not only ensure smooth operation, but also help reduce power consumption, electrode loss, furnace wall corrosion and shorten tap-to-tap cycle, which will bring in great profits.On the basis of consulting large amount of literatures on relative subjects at home and abroad, with the electrical characteristics of modern EAF, its non-linear reactance model is regressed and analyzed according to field data. And by in-depth research on electrical and thermal characteristics and energy balance of the smelting process of EAF, a power supply model based on economic indices of EAF is established, whose target is to minimize power consumption, tap-to-tap cycle and electrode loss, achieving power supply strategy optimization as a whole.The power supply model based on economic indices of EAF is a complicated multi-objective mixed integer nonlinear problem with constraints. In solving this kind of problems, strengths and weaknesses of traditional methods and genetic algorithms are compared. Then aiming at constraint handling Constraint Handling Genetic Algorithm is brought forward. And for dealing with mixed integer variables and multi-objective a Mixed Coding NSGA-II is formed. Then they are both implemented by C++.Taking the No.3 EAF of some steelmaking plant in Shanghai for example, Constraint Handling Genetic Algorithm and Mixed Coding NSGA-II are applied to solve the power supply model of EAF and the simulation results are analyzed and reasonable power supply curve is established. In addition, in order to apply theory to practice, power supply curve optimizing software for EAF is designed according to the actual architecture of control system.