基于油气开采的海上油田中长期电力负荷预测

Medium and long term power load prediction of offshore oil field based on oil and gas exploitation

关键词:  海上油田  负荷预测  强特征量  盲数  逐步回归

Key words:  offshore oil and gas fields  power load forecasting  strong characteristic quantity  blind number  stepwise regression

摘要:

中长期电力负荷预测是电力系统规划设计的理论依据,精确的负荷预测可以减少海上油田平台的安装空间,节省投资费用和运行费用。分析海上油气生产规模、开采工艺、集输工艺等环节的电力负荷需求和发展趋势,提出应用逐步回归分析法进行负荷预测特征量的强相关筛选;基于盲数理论将强特征量盲数化,建立负荷预测盲数化回归模型。以某海上油田区块的历史数据为例,应用所提出的预测方法与其他预测方法进行对比。结果表明,基于强特征量的盲数回归模型的负荷预测精度更高,并给出了未来10 a的中长负荷预测结果。

Abstract:

Medium and long term power load prediction is the theoretical basis of power system planning and design. Accurate load prediction can reduce the installation space of offshore oil platform, and save investment and operation cost. Based on the analysis of the power load demand and future development trend of offshore oil and gas production scale, in the process of production technology, gathering and transportation technology and so on, the stepwise regression analysis method was proposed to select the strong correlation characteristic quantity of load forecasting. Based on the blind number theory, the strong characteristic quantity was blinded, and the blind number regression model of load forecasting was established. Finally, taking the historical data of an offshore oilfield block as an example, the forecasting results of the proposed method were compared with those of other forecasting methods. The result shows that, the load forecasting accuracy of the blind number regression model based on strong characteristic quantity is higher, and the medium and long-term load forecasting results in the next 10 years are given.