Application Research Of Improved Grey Wolf Optimization Algorithm In Optimal Scheduling Of Steelmaking And Continuous Casting

The steel industry is a pillar industry of the national economy,which plays an irreplaceable role in the national economy,social development,infrastructure construction,and science and technology for national defense.The production process of a modern largescale steel manufacturing enterprise usually consists of three major processes: ironmaking at the front,steelmaking-continuous casting in the middle,and rolling at the back.From the perspective of the production process,steelmaking-continuous casting is an intermediate process in the whole process of steel production,which acts as a bridge and link to the entire production process.From the perspective of integrated management,production scheduling is the core of production management and plays a vital role in product quality and efficiency.Therefore,steelmaking-continuous casting production scheduling is the key to ensuring efficient,high-quality and intelligent production of steel manufacturing.Aiming at the complex production process of steelmaking-continuous casting production process with multiple processes,parallel machines,and multiple varieties,this paper deeply studies the steelmaking-continuous casting production scheduling problem.The main research work is as follows:(1)Describe the background of the steelmaking-continuous casting production scheduling problem,and summarize the current research status of steelmaking-continuous casting production scheduling.It mainly includes existing research results and research methods.First of all,the importance of optimizing steelmaking-continuous casting production scheduling problem is emphasized.Secondly,the application of the gray wolf optimization algorithm and deep deterministic policy gradient algorithm are described,respectively,to provide a theoretical basis for the following.(2)In-depth analysis and research on steelmaking-continuous casting production scheduling issues.According to actual production requirements,the optimization objectives and constraints for solving the steelmaking-continuous casting production scheduling problem are abstracted to establish a scheduling mathematical model with the minimum equipment operation conflict time,the shortest heat waiting time,the minimum deviation between the pouring start time and the given pouring start time,and the shortest pouring cut time at the same time as the optimization objective.(3)Aiming at the complex steelmaking-continuous casting production scheduling problem and the shortcomings of the gray wolf optimization algorithm,such as easy to fall into the local optimum,poor global search ability in the later of the iteration,etc.An improved gray wolf optimization algorithm based on deep deterministic strategy gradient algorithm is proposed,which combining deep reinforcement learning with gray wolf optimization algorithm.The deep deterministic policy gradient algorithm is used to train the agent to improve the data utilization rate,break the local optimum,and improve the global search ability in the later of the iteration.(4)The algorithm proposed in this paper is applied to solve the steelmakingcontinuous casting production scheduling problem.The actual production data of a large steel mill in China is used for simulation experiments to obtain an optimized scheduling plan,which shows the feasibility of the algorithm.In the same heat scale and different heat scales,the algorithm proposed in this paper is compared with the standard gray wolf optimization algorithm and the cuckoo algorithm with genetic operators retaining elite strategy.The experimental results show that the proposed algorithm has better solution accuracy and faster solution speed under the premise of ensuring that a feasible plan is obtained.