本文采用Kalman集合过滤方法对生产数据和时间偏移地震波阻抗数据进行了历史拟合。该方法不依赖于油藏模拟器,对油田孔隙度的预测比对渗透率的预测要好,可以利用一系列的油藏模型作为输入,并且通过不断同化观测数据来更新模型,输出适合不确定性分析的大量历史拟合模型。实例表明,通过EnKF方法将时间偏移地震数据和生产数据结合起来描述油藏特征,可以很好地拟合观测数据,并获得正确的模型。 In this paper, the historical data of production data and time-offset seismic impedance data are fitted by Kalman’s set filtering method. This method does not rely on reservoir simulators and predicts oil field porosity better than predicting permeability. A series of reservoir models can be used as input, and the model is updated by assimilating the observed data continually, and the output is suitable for uncertainties Analysis of a large number of historical fit model. The examples show that the combination of time-offset seismic data and production data can be used to describe the reservoir characteristics by the EnKF method, which can well fit the observed data and obtain the correct model.