Research On Crude Oil Scheduling Problem By Petri Net

With the improvement of management level of the petrochemical industry, planning and scheduling technology is playing a much more important role in the enterprise’s business. By optimizing the rescoure allocation during the daily operation and make the business plan, petrochemical plants and refineries could save costs and increase the marginal profit. In the literature, many works focused on the refinery crude oil scheduling, production planning, products blending. In this paper, we proposed a new heuristic algorithm based on Petri-net model to solve a crude oil scheduling problem. Further more, in order to incorporate the dynamic characteristics in the crude oil scheduling, a modeling method based on color-time Petri-net and event model is developed to construct a dynamic scheduling model in the crude oil processing stage. Finally, an integrated simulation and decision-making software is developed based on the above technology to satisfy the need of engineering and research requirements.The main contents of this dissertation are presented as follows:1) A novel dynamic tank assignment approach is proposed to efficiently solve the scheduling problem in the crude oil tank area. The number of changeovers of the pipeline operation is minimized in order to avoid the high operation cost. With the information obtained from pipeline strategy, the charging tank scheduling problem is modeled by using the color-time Petri-net. Besides, the realizability of a target refining schedule which is affected by the tank assignment is analyzed. Based on the former results, operations that lead to an infeasible state are identified and adjusted. A new heuristic approach is proposed by ignoring the infeasible operations. In the end, an industrial case is presented to illustrate the efficiency of the proposed approach.2) The dynamic scheduling characteristics of crude oil processing are also discussed and studied. When abnormal events or disruption orders happened in-the tank area. Eg. Crude oil delays, unavailability of storage or charging tank often occur in crude oil scheduling process. In order to deal with this problem, a modeling method based on color-time Petri-net and event model is proposed to develop the dynamic scheduling model. The classification, definition and descriptions of the emergent events during crude oil scheduling process are described in detail which are formulated by event model. To explain the process more clearly, we give the configuration process of the event model, and illustrate the modeling process of a crude oil dynamic scheduling model. Finally an industrial case is presented to show the efficiency of the proposed approach.3) Based on the fact that there are no efficient techniques and commercial tools for solving crude oil scheduling problem, an integrated simulation and decision-making software is established. The overall architecture and running process of the software are presented.And from the perspective of development, the design and implementation procedure of four modules are described in detail. The feasibility of the software is proved by three typical cases at last.