Hydraulic fracturing is a stimulation technique in which the formation is fractured using high pressure exerted by a fluid. The induced fracture increases the permeability of the formation by providing conductive channels to the formation which results in improved fluids productivity.

Hydraulic fracturing is a common practice in oil and gas, particularly in the development of unconventional low porosity and low permeability reservoirs. However, as the hydraulic fracturing technique is costly, considerable preparations efforts must be made before executing the fracturing operation including simulating the intended fracture model.

A simulation model of a hydraulic fracturing assists in forecasting and controlling the intended fractures that are to be induced. Although the simulation model can be helpful, it may not exactly mimic or predict the actual initiated fractures due to the complex nature of the actual fracturing process. Thus, the simulated model and the actual fracture might differ in many ways which results in an uncertainty in the simulated fracture model.

Therefore, in order to reduce uncertainty, initial data input and assumptions made before and during the fracturing simulation need to be precise in order to obtain accurate simulation results. The growth of a single fracture is often assumed during the simulation of hydraulic fracturing which maybe incorrect as multiple fractures may initiate at the start or middle of the actual fracturing treatment and can have significant effect on the simulated fracturing results.

This paper proposes a method to minimize the difference between fracturing simulation and actual fracture treatment results by utilizing sensitivity tests to the main fracturing parameters. Thus, the initial actual fracturing results were used to detect the occurrence of multiple fractures where the latter was considered to enhance the upcoming simulation accuracy of the proposed treatments. The analysis of high net pressure data during the actual treatment indicates the possible presence of multiple fractures where history matching between actual treatment and simulation results data can give an estimate on when and how many multiple fractures were initiated during the fracturing treatment.

As a result, the data analysis showed that multiple fractures initiation had a significant effect on the fracture simulation results and the assumption of a single fracture during hydraulic fracturing should be discarded unless it is confirmed to be the case. Geological settings of the reservoir and the presence of natural fractures were also found to cause multiple fractures initiation during the treatments, and therefore, the reservoir data and description need to be determined properly before attempting the simulation of a fracturing treatment.

This content is only available via PDF.
You do not currently have access to this content.