Electrical energy consumption forecasting of crude oil pipeline plays a critical role in energy consumption target setting, batch scheduling, unit commitment. For actual crude oil pipelines, the construction of electrical energy consumption forecasting model is challenging due to its uncertain, nonlinearity, intermittency, fluctuations and complicated characteristics. It is difficult to describe the non-linear characteristics of electrical energy consumption forecasting by traditional methods. Therefore, a novel hybrid electrical energy consumption forecasting system based on the combination of support vector machine (SVM) and improved particle swarm optimization (IPSO) is proposed, which includes four parts: a data preprocessing part, optimization part, forecasting part, and evaluation part. In the preprocessing stage, in order to avoid large deviation caused by sampling stochasticity of small samples, the training set and the test set are divided by the stratified sampling method. During the modeling phase, the non-linear relationship in electrical energy consumption forecasting is efficiently represented by SVM, and the algorithm of IPSO is developed to optimize the parameters of constructed support vector machine regression. According to the established IPSO-SVM electrical energy consumption forecasting model, the evaluation criteria, the hypothesis testing and the stability analysis of the forecasting model are introduced for a comprehensive assessment of the system. To verify the forecasting effectiveness of the system, four forecasting cases are presented in this study, comparing the above evaluation indicators of IPSO-SVM with that of eight state-of-the-art prediction methods of GA-SVM, PSO-SVM, SA-SVM, DE-SVM, FOA-SVM, GSA-SVM, SVM and LR, the effectiveness of IPSO-SVM algorithm is evaluated. For the operation data of four crude oil pipelines in China, the results indicate that the proposed IPSO-SVM hybrid model has the best forecasting accuracy and stability than other benchmark models, and the forecast results most closely match the actual data, which shows its superior ability in terms of forecasting accuracy. As a result, it is concluded that the proposed approach can be an efficient technique for crude oil pipeline electrical energy consumption forecasting.


A Hybrid Method Based on Svm Integrated Improved Pso  Algorithm for Electrical Energy Consumption Forecasting of Crude Oil Pipeline

Division / Sponsors: 
Pipeline Systems Division

Paper Proceedings: https://asmedigitalcollection.asme.org/IPC/IPC2020/volume/84447

IPC2020 - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.