Thanks to predecessor's researches, several reserve estimation methods have been proposed over the last few decades, which are commonly accurate for conventional reservoir. However, for stress-sensitive reservoir, permeability is the function of pressure, which may lead to significant errors when using traditional methods even though the flow is boundary-dominated.
In this paper, errors caused by applying several traditional reserve estimation methods based on production data including Flowing Material Balance, Normalized Rate – Cumulative Curved in stress-sensitive reservoir are quantified with numerical simulation. Then, based on the sensitive-analysis, modification procedures for these traditional methods are proposed and validated with synthetic data. Finally, the modified methods are applied to Sulige gas field, a typical stress-sensitive tight gas reservoir.
It can be concluded from the error analysis that for stress-sensitive gas reservoir, Flowing Material Balance and Normalized Rat-Cumulative Curve would result in underestimation. For all the methods above, the magnitude of error is influenced by formation properties, hydraulic fracture qualities, operation conditions and the level of stress-sensitivity. Besides, the error caused by the latter method is larger than that caused by flowing material balance for the same condition. We also find that Flowing Material Balance and Normalized Rate–Cumulative Curve can be modified by introducing new pseudo-pressure and pseudo-time, which take stress-sensitivity into account. These methods are proved to be accurate by both numerical simulated data and field data.
The technique contributions of this paper including: (1) the error caused by stress-sensitivity for traditional reserve estimation methods are quantified and the influencing factors are investigated; (2) modified ideals for traditional methods are proposed.