由于地质的复杂性和资料的不完备性,地下储层表现为“灰箱”系统,表征结果存在着不确定性.基于系统论和科学哲学的方法,探讨了地下储层表征中不确定性的成因类型、内涵及解决方案,并论述了地下储层表征过程中所应重视的科学思维方法.地下储层表征中存在两类基本的不确定性类型:①随机不确定性,主要由局部预测整体(如应用多井资料进行井间储层预测)以及第二性资料的非完全映射(如应用地震资料预测储层)所致；②模糊不确定性,主要由储层概念模式认知不足所致.为了使表征结果逼近地质实际,除了正确应用和创新相关理论和方法外,需要具备科学的表征理念和思维,包括总体把握、分级约束,多科一体、地质为核,发散求索、聚焦论证,辩证思维、综合分析；同时,应正确理解和处理宏观与微观、绝对与相对、静态与动态、映射与多解、局部与整体、模糊与置信、确定与随机、个别与一般、复杂与简约、手工与自动等矛盾统一体的关系.%Due to the complexity of geology and the incompleteness of the data, subsurface reservoirs are shown as grey box systems. Thus, great uncertainty exists in the reservoir characterization. The genetic types, the essence of the uncertainty and the possible solutions for uncertainty reduction were discussed based on the system theory and philosophy of science. The scientific methodology that should be implemented in the reservoir characterization procedure was also addressed. Uncertainties were grouped into two types according to their fundamental difference. One was random uncertainty that was caused by predicting the entirety from the fraction, such as predicting the inter-well reservoir from well data. Another reason of random uncertainty was non-complete mapping from secondary data such as predicting the reservoir from seismic data. The second type of uncertainty was named as fuzzy uncertainty that was caused by insufficient cognition of the reservoir conceptual model. In order to make the reservoir model more close to the geological reality, the correctly implementation of the new related theories and methods should be guided by a more scientific characterization philosophy. Some of the philosophical characterization principles were proposed. The characterization procedure should be perceived in an overall view, and the geological analysis should be hierarchically constrained. The characterization would integrate multiple disciplines while it was geologically focused. The geologists should follow a divergent sucking and reasoning as well as a point-focused demonstration. The investigation will comply with dialectical thinking and comprehensive analysis. Meanwhile, some relationships in contradictory entities, such as macroform and microform, certainty and ambiguity, static and dynamic, mapping and divergence, fraction and entirety, fuzzy and concrete, certainty and randomness, specialty and generalization, complexity and simplicity, hand-operation and automation, should be correctly understood and handled competently and successfully.
展开▼
