Study On Simulation Method Of Crane Scheduling In Workshop Of Steel-making Plant Based On Cellular Automata

Production scheduling in steel-making plant is a link which connects steel-making planning with process controlling in Computer Integrated Manufacturing System of steel-making plant, and it plays an important role in utilizing resource, decreasing production cost and improving production efficiency. Crane scheduling in workshop of steel-making plant, as a key part of production scheduling, is the pivot of materials matching, joining and adjusting among operation sites in steel-making plant. Reasonable and effective crane scheduling will help us to improve production efficiency of every operation site, even the whole enterprise, and lay a foundation for the implementation of production scheduling. Simulation is a useful approach to reveal the nature of crane scheduling, and to avoid the difficulty of analyzing scheduling that is a NP hard problem in theory.Most of previous research about production scheduling neglected the influence of crane scheduling, or supposed that crane movement did not restrict production scheduling, and they never concretely considered the crane scheduling problem. This dissertation aims at cracking the hard nut to make crane scheduling in workshop of steel-making plant. It is our task to explore an adaptable and flexible simulation method that not only describe the complex characteristics of crane scheduling in workshop of steel-making plant, but also optimize crane scheduling, which is very meaningful in theory and practice.Cellular automata model is a kind of computational model, which can describe macroscopic complex behavior and evolution in a system according to simple microscopic local interaction and self-organization evolutionary rules of components in the system. After analyzing the characteristics of crane scheduling in workshop of steel-making plant, the dissertation describes a simulation model of crane scheduling in workshop of steel-making plant based on Cellular Auotmata. The crane scheduling system is composed of operation sites, cranes, materials and crane scheduling rules. In the simulation model, operation sites are considered as fixed grid nodes, cranes as movable nodes, the materials transported as mobile particles, and crane scheduling rules as evolution rules of the model. The properties of operation sites, cranes, materials are set especially through the transfer of mobile particles transported by movable nodes among fixed nodes. The process of crane scheduling is implemented by parallel self-organization evolution mechanism with equal time step. When the model is evolving, Genetic Algorithm is utilized to optimize crane scheduling through choosing optimal strategies of selecting crane and route, whose objective is to minimize throughput time of all materials.A series of simulation cases were conducted according to crane scheduling system of steel-making materials preparation in Vanadium-extracting and Steel-making Plant of Panzhihua Iron and Steel Corporation. The comparison between simulation results and actually measured data shows that the simulation model of crane scheduling in workshop of steel-making plant based on cellular automata is valid. Then in the same simulation conditions, simulation results gotten from the model combined with genetic algorithm are compared with those obtained from the model based on fixed rules, which shows that the role of genetic algorithm in optimizing crane scheduling is effective.Based on above conclusions, the dissertation analyzes the effect of entrance conditions of main materials and equipment status to crane scheduling and production. Simulation results show that compared with normal entrance condition of main materials (1 ladle/14min) gotten from actually measured data, excessively frequent entrance of main materials (1 ladle/10min) makes crane scheduling and production abnormal. This entrance condition should be avoided in practice. However, when the entrance of main materials is at the speed of 1 ladle/18min, the entrance condition has little effect to crane scheduling.At the same time, it is also shown that in the steel-making materials session with 7 cranes and 3 converters, when a large crane fails to function, production can proceed normally and the utilization ratio of all cranes is not beyond the specific limit (70%) for keeping production normal. However, when a converter fails to function, production is abnormal and crane scheduling is affected, thus three converters should be kept in good condition.The exploratory research about simulation model of crane scheduling in workshop of steel-making plant based on cellular automata reveals that the emphases of following study should be further perfect the model and embed it into simulation model of production scheduling.