Research On The Ladle Intelligent Scheduling Method And Application Of Steelmaking-Refining-continuous Casting Process

Multiple converters,refining furnaces,continuous casting machines,and ladles filled with molten steel,and cranes for transporting ladles form the large modern steelmaking-refining-continuous casting production process.Molten steel in converter will be injected into ladle and then be sent to refining positions for refining and continuous casting machines for casting consecutively.Steelmaking-refining-continuous casting production scheduling includes heat(molten steel in a converter)scheduling and ladle scheduling.The heat scheduling is to determine the starting time of the processing equipment.The scheduling should ensure that the furnace in steelmaking and refining process does not conflict,and the continuous casting process operates on time and the casting process is not interrupted.Ladle scheduling should meet the equipment assignment of the start and the end time.The scheduling considers the heat plan as the basis,selects the ladle bearing the heat to determine the crane that transport ladle,ladle transport path and operation start/end time and to ensure that the heat reach the designated equipment and start processing on time.Ladle scheduling includes selection,ladle path planning and crane scheduling.The choice of ladle is based on the production process.Select the decarbonizing ladle for the furnace or select the dephosphorizing ladle and then select the decarburizing ladle.The ladle path preparation determines the path from the slag station to the refining furnace,the caster,and the slag dumping station for the selected ladle after crane transportation.The crane dispatcher determines the start and the end time of the operation of the crane and crane that carry the ladle according to the path planning plan and the heat schedule of the ladle.Because the ladle scheduling must satisfy multiple conflicting objectives and conflicting constraints,it is difficult to use the existing optimal scheduling methods.Therefore,it is in the hands of an empirically-developed scheduling plan,resulting in the time spent in the preparation of the scheduling plan,and the number of ladle used online is large,and The number of strikes started at planned times is low.In view of the problems mentioned above,this paper focuses on the research on the ladle scheduling system software of steelmaking-refining-continuous casting process and the main work is as follows:1.The mathematical model of ladle scheduling for the process of steelmaking-refining-continuous casting is established and that includes ladle optimization matching model,ladle optimization path preparation model and crane optimization scheduling model.The optimal decision-making problem of ladle optimization scheduling for multi-conflict targets and multi-conflict constraints is analyzed.(1)Ladle optimization matching model includes the dephosphorization package matching model and the decarbonization package matching model.The selection of dephosphorization ladle takes the performance index of temperature,life,and the remaining online ladle life.Take the ladle temperature in the specified range,the ladle life within the specified range and the ladle maintenance end time as early as the furnace processing time as the constraints to determine the dephosphorization.The selection of decarburization ladle takes the performance index of maximum temperature,highest life,and the lowest grade of ladle material and the smallest number of ladle.Take the ladle temperature within the scope of the provisions.The ladle life within the scope of the provisions.Ladle material in the furnace within the prescribed scope.Ladle maintenance end time early in the furnace processing start time.Ladle nozzle frequency of use in furnace regulation range and continuous casting of the first furnace time requirements of steel ladle baking time more than half an hour as the constraint,decision the decarburization ladle.(2)The ladles path selection is to determine the path of ladle transportation based on the related performance indexes which are to minimize the ladle transportation path,minimize the number of the lifting down time and minimize the transportation time.Take the path of the crane load greater than full of molten steel ladle weight,the path that crane can be transit time in the furnace plan within the prescribed scope,maximum crane speed and furnace plan determined available path length,the path in transportation of molten steel temperature drop does not exceed furnace process temperature range as the constraint conditions.The path of ladle transportation is the decision variable.(3)The crane scheduling is to decide the number of the crane based on the performance indexes which are the minimum the number of crane mutual avoidance and the maximum of crane operation efficiency;the constraints which are the crane truck constraints,the crane transportation available time constraints,the crane space constraints and steel temperature drop constraints.After analyzed the difficulties for solving the optimal problem,this paper proposed the crane scheduling heuristic algorithm based on confliction resolution policy.This algorithm is to determine the crane that transport heats first,and then use conflicts optimization strategy to resolve transportation conflict.Finally,decide the crane that transport the ladle and transport start/end time.Through the above scheduling model,ladle optimization scheduling is a multi-conflict goals and multi-conflict optimization problem.2.Using intelligent approaches based on rule-based reasoning,heuristics,and Gantt-based human-computer interaction,and combining with the characteristics of the ladle scheduling process,an intelligent dispatch method for ladle is proposed.It includes the ladle matching method based on the least general generalization rule inference,the heuristic ladle path preparation method based on multi-priority,the heuristic crane scheduling method based on the conflict resolution strategy and the Gantt diagram editing man-machine interaction adjustment.The ladle matching method uses the least general generalized intelligence method to establish the ladle matching rules.The priority of ladle path optimization preparation is determined according to the importance of performance indicators.Sort the available paths and optimize the ladle transport path.Crane scheduling is a problem of conflicting in the scheduling.It will combine the Gantt editing interactive adjustment and heuristic method.The proposed method significantly improves the hit rate of heats scheduled start time.3.The proposed ladle intelligent scheduling algorithm is used to develop a ladle scheduling software system for the steelmaking-refining-continuous casting process.It has successfully applied the steelmaking-refining-continuous casting production process with a large domestic steel company.It is developed using object-oriented idea and modular reuse technology.The software system includes graphical configuration of scheduling algorithms,algorithm management,visual simulation,result display and analysis function modules.The graphical configuration of the scheduling algorithm uses graphical configuration techniques to configure algorithm rules and generate ladle scheduling methods.The algorithm management is responsible for the registration,maintenance,and group management of the ladle scheduling algorithm.The verification of visual simulation adopts computer animation technology to simulate the ladle and furnace scheduling plan.It displays the operating parameters of the ladle scheduling process in real time and performs data statistics and uses a graphical form to display the ladle scheduling results.The dispatcher can easily adjust the ladle scheduling through Gantt.The developed ladle scheduling software system is successfully applied to the steelmaking refining continuous casting production line in a large-scale iron and steel enterprise.The application results showed that:the average calculation time for the ladle scheduling by using the proposed algorithm is 3.4 second which is much less than the calculation time 30 second compared with manual scheduling;the number of online ladles reduced from 23 reduced to 19;the ladle maintenance frequency decreased from 17 to 12.Compared with artificial scheduling,the time hit ratio of heats increased from 61%to 65%(the ratio that the heats start on time,according to the plan).Therefore,the ladle scheduling software system enables the production work well and improve the economic efficiency of enterprises.