Research On Heat Treatment Furnace Plan Based On Improved Teaching-learning-based Optimization Algorithm For Foundry Enterprises

The heat treatment furnace plan is an incompatible casting scheduling problem under complex constraints,and is also an important part of foundry production plan.Because the heat treatment has the characteristics of long time and high energy consumption,and most foundry enterprises adopt the order-oriented multi-variety and small-batch production model,the heat treatment furnace plan must consider the utilization rate of the heat treatment furnace,the furnace process constraints between different orders,delivery time and other complex factors.However,the existing manual scheduling mode takes much time and effort,causes low equipment utilization,and it is difficult to consider the complex factors to obtain better furnace plan.Thus,this thesis studies modeling and solution method for heat treatment furnace plan of foundry enterprises based on the improved teaching-learning-based optimization algorithm,and the heat treatment furnace plan system with algorithm-aided scheduling mode is developed and applied to casting enterprises.Firstly,the status quo of the heat treatment furnace plan in foundry firms is analyzed,and the model of the heat treatment furnace plan is established.The model includes a process information model and a furnace plan mathematical model.The process information model is to obtain the process information of the heat treatment operation for castings in various orders,and it is the input of the furnace plan mathematical model.The mathematical model of the furnace plan is a 0-1 integer planning model with consideration of the utilization rate of the furnace,furnace combining constrains and delivery date,and it is the foundation for solving the heat treatment furnace plan.Secondly,a step-by-step solution for heat treatment furnace plan based on the improved teaching-learning-based optimization algorithm is proposed.The solution for the original problem consists of two stages.In the first stage,the task candidate sets are generated from casting task set based on the process constraints.In the second stage,the furnace plan is solved by the improved teaching-learning-based optimization algorithm for each candidate set.The best solution among all candidate sets is selected as the final furnace plan.The solution is designed by the algorithm-aided scheduling mode of furnace plan.With the introduction of historical population in teaching-learning-based optimization algorithm,continuous and discrete teaching-learning-based optimization algorithms are designed to solve the furnace plan for task candidate sets.The feasibility of the step-by-step solution for the heat treatment furnace plan and the performance of the improved teaching-learning-based optimization algorithm are verified by the simulation experiments.Finally,a heat treatment furnace plan system with the algorithm-aided scheduling mode is designed and developed,and it is applied in a domestic foundry enterprise.The system consists of heat treatment process library module,heat treatment process specification module and heat treatment furnace scheduling module.The system can provide functionality of heat treatment process management,heat furnace plan and acceptance,and it can be integrated with foundry information management system.The heat treatment furnace plan system was applied to a typical sand casting enterprise in China.The application results show that the algorithm-aided scheduling mode can significantly improve the efficiency of production scheduling,obtain better furnace plan and improve the utilization rate of the furnace.