Published on: 08/16/2017
This is a new post of our series of publications regarding to Reservoir Engineering in Kraken’s blog. Here, one can expect a succinct discussion about Decline Curve Analysis (DCA) and its importance in forecasting production. Afterwards, the readers are encouraged to subscribe to have access to a more detailed technical contents within DCA practise. These contents will also be showcasing the use of the efficient Python scripting capabilities in Kraken to setup and apply mathematical equations in analysing process workflows and studies.
The production of hydrocarbon fields goes through various distinct stages. When an oil reserve is discovered, an appraisal well is drilled to evaluate the reservoir potential. The next phase refers to the first oil produced and indicates the build-up phase onset. As production continues, its rate reaches a plateau and then finally arrives at the depletion starting point. An example of a theoretical production curve, summarizing the main stages of a field life is shown below, as presented by (Robelius, 2007):
In this context, the flow estimation can be performed by conducting reservoir data analysis together with drilling and development plans and economic studies. However, oftentimes such data may be restricted and not easily accessible; thus, it becomes imperative to devise simplified models to overcome this limitation. The most common methods used to forecast production and hydrocarbon reserves are: depletion rate analysis and depletion curves. These are also used to perform preliminary studies as the computation time required is significantly smaller when compared to other methods. Also, these methods can be applied to perform history-matching when sufficient data is unavailable for analytical methods.
The decline rate corresponds to the decrease in oil production over time, and can be modeled on a monthly or annual basis or any other preferred frequency by the modeler. The decline is typically controlled by both natural dynamics (reservoir withdrawal vs. pressure depletion) and operational factors. Other factors that can also play a role in controlling the decline include political and economic constraints which can cause indirect impact. Although these circumstances are very pertinent, they are not incorporated into the design of production forecast workflows and decision-making.
In order to slow down or delay the production downfall, it means, extending the production plateau, an efficient reservoir management plan will have to be put in place. Fields where water and/or gas are injected to maintain the reservoir pressure (secondary recovery) are recurring examples of slowing down production decline. However, if this technique is not effective, oil production falls again and water production increases as more water is injected. This will lead to a high water cut – the ratio of water volume compared to the total liquid produced – and is commonly observed in mature fields.
Decline curve analysis is a technique routinely used to estimate hydrocarbon reserves under pre-established conditions. It consists of an empirical process based on historical observations and well performance, and it is applied when the historical production has a reasonably established trend. DCA’s importance to the oil industry is vital, especially in the acquisition and divestiture sector, as it is useful for calculating discounted future cash flows and other relevant economic variables. The projected cash flow will determine the viability of a development plan.
Decline curves fitting considers stable conditions – when a well has reached its boundary-dominated flow -, it means, pressure transient has reached all of the boundaries and the static pressure is declining at the boundary. The central assumption is that the established factors will govern the forecast uniformly. Among them, it is worth mentioning reservoir factors, such as pressure depletion, number of producing wells, drive mechanisms, saturation changes, and relative permeability, as well as operation conditions (separator pressure, compressors, tubing size and artificial lift methods). In the case of oil recovery improvements, such as infill, fracturing and fluid injections, the decline curve analysis is implemented to evaluate reservoir performance by comparing production outputs before and after these methods application.
There are several models related to the decline curve equation. None of these methods allows for analysis to be carried out in the transient flow region given that the reservoir conditions are unstable. According to (Arps, 1945), the following equation was proposed and commonly used today as the basis for setting up Decline Curve Analysis modeling:
where b and d are empirical constants related to production data. When d = 0, the equation yields an exponential decline model, and when d = 1 the equation is said to be a harmonic model. When 0 < d < 1, a hyperbolic model is defined. Further, some studies on unconventional players have shown that, in shale wells, the factor d can be greater than 1 (Okouma et., al, 2012).
The following graph compares the different decline curve behaviors:
Other techniques can be used to predict oil production, such as Water Oil Ratio (WOR) and Water Cut (WC) extrapolation. These parameters are regularly used for the evaluation and prediction of waterflood performance. The relationship of WOR (or WC) and oil production allows for the extrapolation of a straight-line as a technique to determine recovery factor.
In that light, Kraken allows users to perform Decline Curve studies by using different methodologies, from Exponential Decline to Water-Cut and Water-Oil Ratio forecasts, through historical data extrapolation. The software has its own Python API (Application Programming Interface) which enables creating a specific script to run such analysis and then visualize and manipulate output plots. Check it out by subscribing here and explore Kraken’s post-processing capabilities.
Decline rate corresponds to the decrease in oil production over time, and it is typically controlled by both natural dynamics and operational factors. These factors are utilized to forecast production and to support decision-making. Analyzing production data inaccurately can lead to severe financial consequences for companies; thus, these mathematical approaches are broadly used due to their key role for various branches of the industry. To find out more, download this free White Paper.