论文部分内容阅读
油田开发方案设计中 ,产量预报正确与否与油田开发成败息息相关 ,它决定了油田生产规模 ,投资决策等。正确预测油田产量对改善开发生产效果和提高措施作业效益有重要作用 ,油田产量预报方法很多 ,如递减曲线法、神经网络法、Poisson分布法等。各种油气田产量预报模型的应用关键在于参数估计 ,常规参数估计方法为试凑法和最小二乘法 ,工作量大 ,可靠性差 ,所以引入了一种新型遗传算法求取参数最优估计值 ,并通过遗传算法在Γ模型中的应用实例说明了该方法在油气产量预报中的可行性
Oilfield development program design, production forecasting is correct or not and the success or failure of oil development is closely related to it determines the oilfield production scale, investment decisions. Proper prediction of oilfield production plays an important role in improving development and production efficiency and improving operating efficiency. There are many methods for forecasting oilfield production, such as decreasing curve method, neural network method and Poisson distribution method. The key to the application of oil and gas field production forecasting model lies in the parameter estimation. The conventional parameter estimation methods are trial and error method and least square method. The workload is large and the reliability is poor. Therefore, a new genetic algorithm is introduced to obtain the optimal estimation parameters The application of genetic algorithm in Γ model shows the feasibility of this method in oil and gas production forecast