基于粒子群优化的最小二支持向量机在页岩气井当中的产量预测应用

Application of Least Squares Support Vector Machines Based on Particle Swarm Optimization in Production Forecast of Shale Gas Wells

石油与天然气开发、生产运营期间,在确保质量的前提下,为了尽可能保证企业的利润空间,运营平台的合理管理和规划、页岩气井的生产调度、经济管控、资源分配最优化是提高该指标的重要措施。页岩气井的产量预测是直接影响这一系列措施的重要参考指标之一。基于此背景,将最小二乘支持向量机应用于页岩气产量预测,以期发挥指导页岩气井生产运营的作用。基于此,概述原理性了采用的方法,利用粒子群优化处理实际工程数据,利用MATLAB完成实际数据仿真,并验证预测数据与实际生产数据。结果表明,收敛性好,预测精度和训练速度较高,对实际生产具有一定指导意义。

During the period of oil and gas development, production and operation, on the premise of ensuring quality, in order to ensure the profit space of enterprises as far as possible, reasonable management and planning of operation platform, production scheduling of shale gas wells, economic control and optimization of resource allocation are important measures to improve the index. Production prediction of shale gas wells is one of the important reference indicators that directly affect these measures. Based on this background, least squares support vector machine (LS-SVM) is applied to shale gas production prediction in order to play a guiding role in shale gas well production and operation. Based on this, this paper outlines the principle of the method used, using particle swarm optimization to process actual engineering data, using MATLAB to complete the actual data simulation, and verify the prediction data and actual production data. The results show that it has good convergence, high prediction accuracy and training speed, and has certain guiding significance for practical production.