**9.2 Classical finite difference reservoir simulation**

 A dynamic model using a finite difference compositional reservoir simulator is built around a two well pad by using previously derived 3D properties (**Figure 4**) as well as the final hydraulic fracture geometry and its resulting conductivity (**Figure 13B**). Another derived key input is the inter-well permeability resulting from the interaction between the hydraulic and natural fractures and captured by the strain resulting from the geomechanical simulation using the MPM 2D plane strain framework. This strain is converted into an effective inter-well permeability (**Figure 14B**) using the approach shown in [36] where a calibration factor that relates strain to the matrix effective permeability is estimated through history matching. This calibration factor was very quickly estimated and a match was found for both wells A and B (**Figure 15**) for oil, gas and water rates using a bottom hole pressure. The major drawback of using a finite difference reservoir simulator is the intensive numerical computation required even in today's new generation parallel reservoir simulators. For example, the case described in this section required a 6-hour run on a good workstation. For unconventional reservoirs, we could trade the accuracy for a faster run time.

#### **9.3 Unconventional reservoir simulation using fast marching method (FMM)**

The motivation and the unique features of the Fast-Marching Method (FMM) simulator needed for unconventional reservoirs were described in Ouenes et al. [6] and Paryani et al. [18]. The input 3D models (**Figure 4**) and hydraulic fracture geometry (**Figure 14**) were input in the FMM simulator along with the PVT

 (Pressure, Volume, Temperature) and other inputs. The resulting pressure depletion at the end of the simulation derived from the FMM simulator (**Figure 16B**) shows the same features as those seen in the pressure depletion estimated in the classical reservoir simulator (**Figure 16A**). The same conclusions can be seen when examining the pressure distribution in a cross-section view as shown in **Figure 17**.

## **Figure 14.**

*(A) fracture geometry and conductivity resulting from the stimulation of two wells and (B) interwell permeability resulting from the interaction between the hydraulic and natural fractures.* 

### **Figure 15.**

*History matching of oil (green), gas (red), and water (blue) by using the bottom hole pressure (BHP) as a constraint. Notice the good match of both well measurements A and B, achieved very easily and quickly by using one single history matching parameter.* 

### **Figure 16.**

*(A) Areal view of the pressure depletion from finite difference reservoir simulator compared to the (B) pressure derived from the fast marching method simulator.* 

*Surface Drilling Data for Constrained Hydraulic Fracturing and Fast Reservoir Simulation… DOI: http://dx.doi.org/10.5772/intechopen.84759* 

### **Figure 17.**

*(A)–(D) Cross section view of the pressure depletion from a finite difference reservoir simulator compared to the (E)–(H) pressure derived from the fast marching method (FMM) simulator.* 

 There is a major difference between the classical finite difference reservoir simulation run and the one using the FMM: using the same computer, the full-scale heterogeneous model using the compositional finite difference reservoir simulator requires six (6) hours run time due to the large number of components while the FMM multi-phase black oil simulator results were derived in less than 1 minute.

 With such a rapid evaluation tool and robust workflow that leverages the multiple constraints derived from the use of surface drilling data, the complex balance between finding the optimal Net Present Value (NPV) per well or per section could be easily estimated in few days or even hours. Using the current industry tools to achieve the same objective will take many weeks if not months and will have a large uncertainty if no well logs or seismic are available as it is very usual the case in unconventional reservoirs where well data and seismic are sacrificed at the altar of cost cutting measures. Fortunately, the surface drilling data provides a reasonable alternative that enables the entire reservoir modeling and management workflow.
