Introduction

Methodology

The proposed production analysis workflow consists of three major stages as shown in Fig. 2. They are (1) Stage 1: the initial production analysis, (2) Stage 2: the production evaluation using a probabilistic approach, and (3) Stage 3: parametric sensitivity.


Fig. 2

Proposed workflow

In the initial production analysis stage, two wells in an unconventional shale gas reservoir were studied. The appropriate rate transient analysis is performed using IHS Harmony and Kappa Saphir, two software packages specially developed for the rate and pressure transient analysis. The software packages were used to analyze the available production and pressure data.

In the probabilistic evaluation stage, the probability distribution function is defined to obtain the completion parameters. Some parameters are more precise, while others have some level of uncertainty. For example, the well length and the reservoir temperature are more precise than the rock permeability and the fracture half-length. Reservoir parameters, such as permeability, porosity, initial pressure, and the reservoir extent, were evaluated. Completion parameters, including the number of fracture clusters, horizontal well length, well spacing, fracture height, fracture half-length, and fracture conductivity were also assessed. The main objective of the probabilistic evaluation stage is to reduce the uncertainty of unknown well and reservoir parameters. After the probability distributions of the parameters are defined, Monte Carlo simulation was used to construct the stochastic samplings of the parameters. Monte Carlo picks random samples for each parameter. The samples are then used to run the Composite Model numerous times. Thus, the uncertainties of the collected parameters are adequately reflected. The model used in this paper considers the homogenous porosity, isotropic dual region permeability, well location in the center, constant pressure step, and shale lithology. The two main outcomes of the previous steps are the probabilistic production forecast and the probability distribution of the final parameters. The probabilistic forecast captures the likely production range of the subject well. The probability distribution of the final parameters would redefine the initial probability in light of the production forecast. In order to accomplish this task, @Risk (MS Excel add-in) or IHS Harmony was used. @Risk can define the probability distribution of each collected parameter and create Monte Carlo samplings. IHS Harmony can run the composite model and provide results.

In the sensitivity analysis stage, the impact of the well and reservoir parameters on gas production is investigated. A workflow similar to that used in the probabilistic evaluation is used, except that the probabilistic distribution of the final reservoir parameters is used. The sensitivity of the well and reservoir parameters would be repeatedly verified to obtain the production. In order to accomplish these tasks, IHS Harmony was used as the main tool.