Refined oil distribution refers to the process of transporting oil products from the upstream refinery to the downstream sales depot according to the transportation plan. In addition to guaranteeing the economic benefits of the refined oil distribution system, the high utilization efficiency of the refined oil distribution facilities should be maintained as far as possible to improve the overall operating efficiency of the refined oil distribution system. At present, for the optimization of refined oil distribution system, the existing domestic and foreign refined oil distribution plans usually take the lowest distribution cost as the goal, while ignoring the utilization efficiency of transit oil depot. In addition, as an important node in the refined oil distribution system, the inventory management of the transit oil depot is extremely critical. If the inventory management of the oil depot is chaotic, ineffective or inefficient, it will not only harm the efficient operation of the refined oil distribution system, but even affect the normal supply of refined oil. In this paper, based on the existing research, four evaluation indexes of the utilization rate of oil depot were introduced, including turnover times, per capita turnover, running cost per ton of oil and working time per ton of oil, and based on these, whether the transit oil depot put into use is in a state of efficient operation. After that. The operation cost of oil product transportation and transit oil depot as well as utilization efficiency of transit oil depot were comprehensively considered, and constraints such as supply capacity of refinery, demand of selling oil depot, transportation capacity and storage of oil depot were taken into account. The multi-objective mixed integer linear programming model of the refined oil distribution system was constructed with the minimum cost of the system and the maximum turnover of the transfer terminal as the objective function, and the model was solved by the augmented ε-constraint method. The research results have been successfully applied to a refined oil distribution system in northwest China. Through the verification of field real data on the spot, it can be concluded that the turnover times and per capita turnover of the transit oil depot can be increased by 48.1% and 54.3%, respectively, and the operation cost and operation time of ton oil can be decreased by 52.8% and 58.3%, respectively. The results of this study have certain guiding significance for the formulation of regional refined oil supply and transportation plan and transit oil depot operation plan.

The operation conditions of multi-product pipeline changes frequently and it is difficult to judge the operation state accurately. Therefore, the recognition and monitoring by on-site personnel is easy to cause misjudgment. In order to realize the accurate recognition of pipeline operation conditions, considering the physical spatial characteristics of the pipeline, the operation parameters (pressure, flow rate and density) of each station are sorted out. Considering the time series characteristics of pipeline operation, operating data matrix is formed to overcome the transient disturbance at a single moment based on the SCADA data. Aiming at the high-dimensional and non-linear characteristics of pipeline operating data, the powerful feature compression and reconstruction capabilities of the convolutional autoencoder (CAE) are used to reduce the noise of pipeline data. T-distributed stochastic neighbor embedding algorithm (T-SNE) is used to perform dimensionality reduction and clustering processing on pipeline data, and finally the model based on CAE-TSNE for pipeline operation condition recognition is established. Taking two real multi-product pipeline as example, the mainstream machine learning nonlinear classification models (ANN, DT and RF)were compared with the proposed method. The results show that the operating condition identification model based on CAETSNE has the highest accuracy, and the recognition rate of clustering identification of operating data after noise reduction can reach 99%, which can guide the operation and management of on-site pipelines.

The pipeline serves as a vital link connecting the upstream and downstream segments of the oil and gas industry chain, playing a crucial role in modern energy systems and comprehensive transportation systems. Amidst the deepening institutional reforms within the oil and gas sector and the strategic goal of "carbon peak and carbon neutrality", oil and gas pipelines are evolving towards large scales, networking and diversification, which brings both new opportunities and challenges to the operation management of extensive and intricate pipeline networks. Leveraging insights from the current development status of China's oil and gas pipeline networks, this paper analyzed the trends and difficulties in the research of oil and gas pipeline network operation across four pivotal domains: system analysis, simulation and optimization, operation monitoring, and new pipeline transportation. Relevant research included reliability and resilience evaluation technology for pipeline network system, operation simulation and transportation optimization technology for pipeline networks, pipeline body monitoring and repair technology, as well as pipeline transportation technology for hydrogen, methanol, liquid ammonia, and LNG. In light of the new era of "one network nationwide" and the emerging trend of "multi-energy complementarity", forward-looking technological research directions have been proposed, such as digital twin, intelligent scheduling and control, intelligent early warning, and multi-network integration, aiming to drive the safe, efficient, and green development transformation of China's oil and gas pipeline networks. Finally, this paper put forward the prospect: at this stage, it is necessary to capitalize on China's energy and resource endowments, accelerate the adaptation to the operation mode of pipeline networks amidst the new landscape, and play the role of oil and gas pipeline networks as an energy artery in the new stage of "one network nationwide". In the future, it is necessary to progressively advance the flexible transportation capabilities of pipeline networks across multiple media, and tap the development potential of oil and gas pipeline networks within comprehensive transportation systems and integrated energy systems.

Cooperation among enterprises can bring overall and individual performance improvement, and a smooth coordination method is indispensable. However, due to the lack of customized coordination methods, cooperation in the downstream oil supply chain cannot be carried out smoothly. This paper intends to propose a multi-party coordination method to promote cooperation between oil shippers and pipeline operator by optimizing oil transportation, oil substitution and pipeline pricing schemes. An integrated game-theoretic modeling and analysis approach is developed to characterize the operation behaviors of all stakeholders in the downstream oil supply chain. The proposed mixed integer nonlinear programming model constrains supply and demand capacity, transportation routes, oil substitution rules and pipeline freight levels. Logarithm transformation and price discretization are introduced for model linear approximation. Simulation experiments are carried out in the oil distribution system in South China. The results show that compared to the business-as-usual scheme, the new scheme saves transportation cost by 3.48%, increases pipeline turnover by 5.7%, and reduces energy consumption and emissions by 7.66% and 6.77%. It is proved that the proposed method improves the revenue of the whole system, achieves fair revenue distribution, and also improves the energy and environmental benefits of the oil supply chain.