论文部分内容阅读
机动管道在输送油品和地形条件固定的情况下,管道长度、空压机的排气压力和排气量是影响管道排空作业时间的主要因素,但几种因素对排空时间的影响呈现复杂的非线性关系。以机动管道排空过程的实验数据为基础,分析了排空时间随各影响因素的变化规律。通过分析,发现各因素与排空时间之间具有较好的关联性,用支持向量回归的方法可以对排空时间进行预测。在对样本数据进行训练和预测时,采用基于morlet小波核的支持向量回归方法,并与高斯核的预测效果进行对比,发现小波核比高斯核具有更好的排空时间预测效果,并给出了管道气顶排空时间预报公式。
The length of the pipeline, the pressure of the air compressor and the displacement of the air compressor are the main factors that affect the emptying time of the pipeline when the oil pipelines are transported and the topographic conditions are fixed. However, the influence of several factors on the emptying time Complex non-linear relationship. Based on the experimental data of the emptying process of the maneuvering pipe, the changing rules of the emptying time with various influencing factors were analyzed. Through the analysis, it is found that there is a good correlation between each factor and the emptying time, and the exhaust time can be predicted by the method of support vector regression. In the training and prediction of the sample data, the support vector regression method based on the morlet wavelet kernel is used and compared with the prediction effect of the Gaussian kernel. It is found that the wavelet kernel has a better prediction effect of the emptying time than the Gaussian kernel, Predict the time of gas cap emptying.