针对长输液体管道调度员实际生产过程中无法实时有效地判断工况,以及以往理论公式对实际生产的应用存在局限性等问题,提出管道智能化研究,辅助调度员日常工作,降低工作压力和减少工作量,提高工作及决策效率。通过建立管道智能化架构,分析获得的数据挖掘层为核心层,核心层的构建需要根据管道业务需求,由业务专家和数据挖掘专家共同完成。对比理论公式和数据挖掘算法获得各自优势特点,基于实际生产数据的数据挖掘算法可根据业务需求建立相应的预测、识别模型,从而构建数据挖掘层。将数据挖掘算法应用于HY热油管道,建立GA-BP油温预测模型,其分析及预测结果相比理论公式计算更准确,可有效指导HY热油管道工艺调整。可见,基于数据挖掘算法的管道智能化研究可有效提升管道安全优化管控水平,并为管道智能化控制奠定基础。

For the dispatcher's actual production process, it is impossible to judge the working condition in real time.The previous theoretical formula has limitations on the actual production guidance,and the pipeline intelligent research is proposed.In addition, it assists the dispatcher in daily work,gradually reduces the dispatcher's work pressure and workload, and improves work and decision-making efficiency.Through the establishment of the pipeline intelligent architecture, the analysis obtains its technical support layer as the core layer.The construction of the core layer needs to be completed by technical experts and algorithm experts according to the pipeline business requirements.Compared with theoretical formulas and data mining algorithms,data mining algorithms based on actual production data can establish corresponding prediction and recognition models according to business needs.Application of data mining method to HY hot oil pipeline, statistical analysis of oil temperature and ground temperatureand the GA-BP oil temperature prediction model is established.The analysis and prediction results are more accurate than the theoretical formula calculation,which can effectively guide the process adjustment of HY hot oil pipeline.The intelligent research of pipeline based on data mining algorithm can effectively improve the pipeline safety optimization control level and lay a foundation for intelligent control of pipelines.