Application Of Improved Gravitational Search Algorithm In Steelmaking-Continuous Casting Scheduling

The development level of iron and steel industry is an important symbol to measure the strength of a country’s comprehensive national strength.whether Chinese iron and steel enterprises and competitive in the international market mainly depends on the green process of production and the intelligentization of management.The steelmaking-continuous casting process is a crucial section of the steel manufacturing process and is also a key process for energy-saving,emission reduction,and green production.The main of production management is production scheduling.The quality of scheduling directly affect the energy consumption and carbon emission of steel manufacturing process,as well as the quality and efficiency of final products.Due to high temperature and energy consumption during the steelmaking-conti nuous casting process,the complex physical and chemical changes involved in the process can be optimized with a reasonable production scheduling plan that reduces waiting time between processes,saves energy and raw material consumption,and ultimately lowers production costs,improves product quality and competitiveness.Therefore,it is of great significance to research intelligent scheduling methods for steelmaking and continuous casting production processes.Based on the production process of steelmaking and continuous casting,this article puts forward some measures to solve the shortcomings of the traditional gravitation algorithm,and applies the improved gravitation algorithm to the static and dynamic scheduling problems of steelmaking and continuous casting.The main work is as follows:1)Due to the drawbacks of the traditional gravitational search algorithm,such as the imbalance of global and local searches and the tendency to fall into local optima,an improved algorithm called JGSA,which is prone to escaping local optima,has been proposed.Firstly,the group sharing strategy of particle swarm optimization(PSO)is introduced to improve GSA,thus solving the problem of the imbalance of global and local searches in GSA.Next,the boundary mutation strategy and the random velocity update strategy are introduced to solve the problem of GSA falling into local optima.To verify the effectiveness of the improvement,six test functions(two high-dimensional single-peak test functions,two high-dimensional multi-peak test functions,and two multi-peak low-dimensional test functions)are used to test the improved JGSA algorithm.The test results are compared with those of PSO and GSA,respectively,by conducting 50 independent tests on maximum value,minimum value,median,mean,and standard deviation.The results show the effectiveness of the JGSA algorithm proposed in this paper.2)The JGSA algorithm is applied to the static scheduling of steelmakingcontinuous casting,with the optimization objectives of minimizing the deviation of casting start time and minimizing the redundant waiting time of heats on equipment.Simulation experiments are conducted to compare JGSA with other algorithms such as gray wolf optimization and cuckoo search algorithm to simulate steelmaking-continuous casting static scheduling using actual production data.In order to verifies the stability of JGSA,antitruth of different populations were carried out,and compared with cuckoo search algorithm(CS)and grey wolf optimization(GWO),to prove that JGSA in this article has stability in solving the static scheduling problem of steelmaking and continuous casting.3)Steelmaking-continuous casting is dynamically scheduled using the JGSA algorithm.To deal with disturbance phenomena such as converter and refining furnace failures,emergency task insertions,and order cancellations that occur on-site,which,if continued to be implemented according to the original schedule,would lead to casting conflicts or interruptions,the optimization objectives of minimizing the time of casting interruptions and casting conflicts are adopted.Dynamic disturbance events are handled using JGSA in a timely manner to obtain a new rescheduling plan.Genetic algorithm(GA)and genetic scheduling algorithm(GSA)rescheduling results are compared.It has been shown that JGSA is not only able to generate dynamic rescheduling schemes for steelmaking-continuous casting,but is also highly stable.