No abstract

Studies have found that the Permian is another important stratum for petroleum exploration except the Jurassic coal measures within Turpan-Hami Basin recently. However, the knowledge of the depositional environments and its petroleum geological significances during the Middle-Late Permian is still limited. Based on the analysis about the sedimentological features of the outcrop and the geochemical characteristics of mudstones from the Middle Permian Taerlang Formation

Accurate assessment of uncertainty of production performance is critical for successful planning and assets development. Particularly in deepwater scenarios, where the uncertainty in the fluid and reservoir characterization is of high level for the difficulties for well testing, fluid and core sampling. Therefore, specific methods are needed for efficient uncertainty quantification of production for deepwater reservoirs with limited information. In this paper, we introduce a history matching method to assimilate static geological data and production data base on the ensemble Kalman filter (EnKF). The EnKF is independent of simulators, and is suitable for uncertainty assessment, reservoir monitoring and performance prediction. We tested this method with a deepwater case. We analyzed the effects of initial ensembles and production history. After that, the uncertainty of production prediction is quantified and the posterior distribution of cumulative production can be estimated. The EnKF is shown to be efficient in updating fluid and reservoir heterogeneity. By sequentially assimilating observed data, the EnKF is suitable for reservoir monitoring and performance prediction. The results indicate that history match with plausible geology can be improved with the improvement in the generation of the initial ensemble. History matching with longer history can narrow the range of ultimate recovery distribution, thus the uncertainty can be decreased. And the results also show that the updated amounts of parameters in the ensemble is larger in the first iteration, and it will get smaller gradually with more data been assimilated. After the history matching, the ultimate recovery can be obtained with each set of parameters in the ensemble, the uncertainty can be quantified with the statistical frequency distribution of recovery. The proposed methodology provides a practical means to assess uncertainty in history matching for deepwater fields. This method can also be used to project and risk management.

From a general review, most petrophysical models applied for the conventional logging interpretation imply that porosity, permeability, or water saturation mathematically have a linear or nonlinear relationship with well logs, and then arguing the prediction of these three parameters actually is accessible under a regression of logging sequences. Based on this knowledge, ensemble learning technique, partially developed for fitting problems, can be regarded as a solution. Light gradient boosting machine (LightGBM) is proved as one representative of the state-of-the-art ensemble learning, thus adopted as a potential solver to predict three target reservoir characters. To guarantee the predicting quality of LightGBM, continuous restricted Boltzmann machine (CRBM) and Bayesian optimization (Bayes) are introduced as assistants to enhance the significance of input logs and the setting of employed hyperparameters. Thereby, a new hybrid predictor, named CRBM-Bayes-LightGBM, is proposed for the prediction task. To validate the working performance of the proposed predictor, the basic data derived from the member of Chang 8, Jiyuan Oilfield, Ordos Basin, Northern China, is collected to launch the corresponding experiments. Additionally, to highlight the validating effect, three sophisticated predictors, including k-nearest neighbors (KNN), support vector regression (SVR), and random forest (RF), are introduced as competitors to implement a contrast. Since ensemble learning models universally will cause an underfitting issue when dealing with a small-volumetric dataset, transfer learning in this circumstance will be employed as an aided technique for the core predictor to achieve a satisfactory prediction. Then, three experiments are purposefully designed for four validated predictors, and given a comprehensive analysis of the gained experimented results, two critical points are concluded: (1) compared to three competitors, LightGBM-cored predictor has capability to produce more reliable predicted results, and the reliability can be further improved under a usage of more learning samples; (2) transfer learning is really functional in completing a satisfactory prediction for a small-volumetric dataset and furthermore has access to perform better when serving for the proposed predictor. Consequently, CRBM-Bayes-LightGBM combined with transfer learning is solidly demonstrated by a stronger capability and an expected robustness on the prediction of porosity, permeability, and water saturation, which then clarify that the proposed predictor can be viewed as a preferential selection when geologists, geophysicists, or petrophysicists need to finalize a characterization of sandy-mud reservoirs.