Font Size: a A A

Study On The Optimization For Molten Iron Allocation And TPC Scheduling Integrated Problem In Steel Industry

Posted on:2020-08-29Degree:MasterType:Thesis
Country:ChinaCandidate:J LiuFull Text:PDF
GTID:2481306044958969Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
The molten iron logistics is a high-temperature hot chain logistics process which links ironmaking production and steelmaking production,whose mission is to ensure that the molten iron produced by the blast furnace can be delivered to the steelmaking workshop on time and on quantity with high quality.The hot metal logistics process not only needs to consider the rail transportation mode,the connection between ironmaking and steelmaking,the recycling of torpedo tanks,etc.,but also needs to consider the operating characteristics such as the energy loss cost brought by the high temperature hot chain,so that the molten iron logistics scheduling is very complicated.Reasonable allocation of molten iron and TPCs scheduling can reduce the temperature drop of molten iron and the cost of steel production,increase the turnover rate of TPCs.Meanwhile,it can also reduce the operating cost of torpedo cars and improve the service life of TPCs.Therefore,the study on the optimization for integrated molten iron allocation and TPC scheduling problem in steel industry is meaningful not only in academic research but also in practical application.In this thesis,we emphasis on two key problem in molten iron logistics,the molten iron allocation problem and TPC scheduling problem,and integrate this two problems into one whole scheduling problem.Under the promise of satisfying production safety,according to the blast furnace tapping information,the iron demand information of the converter,as well as the status information of slag,dephosphorization,desulfurization and other pretreatment station,we determines the flow direction of the molten iron and scheme the scheduling of the TPC,thereby achieving the purpose of reducing the transportation time of the molten iron and reduce the temperature drop loss of molten iron during transportation We establish the mathematical model based on the problem characteristics and then design the optimization algorithm.Finally,the decision support system is developed.The main research contents of this thesis are as follows:1.In the process of molten iron transportation,the distribution of molten iron and the scheduling of TPCs seriously restrict the stable and timely supply of molten iron,the turnover rate and the temperature drop loss of molten iron during transportation.In this thesis,we study on the integration of molten iron distribution problem and the TPCs scheduling problem.A mixed integer programming model is established for the weighted sum of tank transport time and empty tank waiting time,and the actual process constraints are considered such as the molten iron transportation time window.2.Develop branch and price algorithm to solve the integrated molten iron allocation and TPC scheduling problem in steel industry.The Danzig-Wolfe decomposition method is used to transform the MIP model into a set-partitioning model which includes a master problem and many independent sub problems.A column generation algorithm is designed to solve the linear relaxation of the set partitioning master problem During the process of column generation algorithm,a heuristic algorithm based on greedy strategy is proposed to obtain the initial feasible solution of master problem.As a result,the restricted master problem is obtained.The price sub problem is an elementary shortest path problem with resource constraints that has been proven to be an NP-Hard problem.In order to solve the sub problem,a pulse algorithm with three pruning strategies to speed up the algorithm is proposed.3.During the branching process of branch-and-price algorithm,engineering.optimization is added to speed up the algorithm.Select appropriate features and labels,and simulate the strong branching strategy to-replace the most infeasible branching strategy by SVM sorting algorithm.The experimental results show that the SVM sorting algorithm simulating the strong branching strategy has fewer branch nodes then the most infeasible branching strategy and meanwhile it is faster.4.A molten iron logistic scheduling system is designed and developed to provide supports for steel industry with scientific decisions.
Keywords/Search Tags:molten iron logistic, mixed integer programming, ESPPRC, column generation, branch and price algorithm
PDF Full Text Request
Related items