Search

Highlights • A novel two-stage modeling framework based on event sequence is developed. • A SIASA algorithm is presented to re-optimize the solution of first-stage model. • A data-driven method is proposed to accomplish the fast optimization of pipeline scheduling. • The experimental results of 50 cases are given to illustrate the superiority of the proposed method. Abstract The detailed scheduling of distributing products through pipelines is one of the most essential tasks in the management of multiproduct pipelines. The current methods of scheduling optimization are still limited by the dilemma between optimality and efficiency, especially when dealing with real-world scheduling issues. This paper introduces a two-stage mathematical model based on the event sequence, whose model scale greatly decreases after stripping the terms of event sequence. Then, a data-driven method is presented to learn a large number of existing scheduling data and accelerate the calculation of first-stage model. The method is decomposed into three parts: (1) take cold start to generate a good deal of basic training data; (2) train the neural network for the fast solution of first-stage model; (3) take real-world cases to improve the self-learning of neural network. A multiproduct pipeline in China is taken as the example to prove the practicability and superiority of the proposed method. The experimental results show that the proposed method could decrease the computation time from about several hours to several minutes. [ABSTRACT FROM AUTHOR]

Highlights • A MINLP model is developed for scheduling of branched multiproduct pipeline networks. • The model rigorously accounts for nonlinear hydraulic calculation associated with flowrate. • A priority algorithm based on continuity, urgency and compatibility is presented. • Three cases are given to validate the high efficiency of the proposed method. Abstract This paper aims at the detailed scheduling of branched multiproduct pipeline networks with multiple sources and an output terminal. Nevertheless, the complexity of the pipeline structure, coupling of hydraulic and scheduling factors, impacts of physical properties of products, and simultaneous injections at different sources make it challenging to draw out the injection operations at each source. Based on a continuous-volume and discrete-time representation, this paper develops a MINLP model to find the optimal injection operations of all sources at minimum total operational cost, rigorously tracking batch migration and power consumption at each pipeline segment. A priority algorithm is presented to obtain a high-quality solution of this large-scale and nonlinear scheduling problem. Finally, experimental results on two virtual pipeline networks and a real-world pipeline network in China are given to validate the proposed approach. [ABSTRACT FROM AUTHOR]

Highlights • A mixed-time MILP model is proposed aiming at optimal scheduling of COPD. • Multiple transportation modes such as pipeline, road and railway are considered. • The detailed operation and transport scheduling of a crude oil port depot can be obtained. • A real case study in China is given to demonstrate this method’s practicality. Abstract Crude oil port depots are key hubs in oil trade and thus play a vital role in the crude oil supply chain. Based on the characteristics of the scheduling system in an oil port depot, the development of a precise schedule is essential for handling asynchronous upstream supply and downstream market demand, thereby reducing operation cost as well as the impact of market fluctuations. Most previous studies focused only on the depot system, and few have taken the export transport scheduling into consideration. In this study, the diversity of crude oil storage modes, complexity of operations, and variety of transportation modes are considered. To minimize the total operation cost, a mixed-time representation based on a mixed-integer linear programming model is established. Finally, three real cases in a crude oil port depot in China are studied, and the monthly schedules of the depot are developed. Compared with previous algorithms, this method yields more accurate results in a considerably shorter time with improved computational stability. [ABSTRACT FROM AUTHOR]