论文部分内容阅读
温降是热油管道运行管理中优化输送方案及分析运行能耗的决定因素,针对传热系数难以获取及运行参数波动引起的无法对温降进行准确预测的问题,对埋地热油管道温降与多种运行参数及油品物性之间的相关性进行分析,提出一种基于相关向量机算法(RVM)的埋地热油管道温降预测的新方法.通过对出站油温、出站压力、输量、地表温度、埋深、管长、管径和油品物性与温降之间的内在规律进行学习训练相关向量机,建立埋地热油管道温降预测的相关向量机模型.对东北某输油管道温降进行预测的结果表明,方法与传统的反算插值法相比,预测结果平均相对误差降低4.43%,具有预测精度高、泛化性好等优点,更适用于现场实际工况下的埋地热油管道温降的预测.
Temperature drop is the decisive factor to optimize the transportation scheme in hot oil pipeline operation management and to analyze the energy consumption of operation. In view of the difficulty in obtaining the heat transfer coefficient and the inability to predict the temperature drop caused by fluctuation of operating parameters, And the correlation between various operating parameters and oil properties, a new method based on correlation vector machine (RVM) is proposed to predict the temperature drop of buried hot oil pipeline.Through the analysis of outbound oil temperature, outbound pressure , The output, the surface temperature, the buried depth, the pipe length, the pipe diameter and the physical properties of the oil and the temperature drop between the learning vector training machine-related vector machine model to establish the temperature drop prediction of buried hot oil pipeline vector machine model. The results of a pipeline temperature drop prediction show that compared with the traditional inverse interpolation method, the average relative error of the prediction results is reduced by 4.43%, which has the advantages of high prediction accuracy and good generalization, and is more suitable for on-site actual conditions Buried hot oil pipeline temperature drop prediction.