Determining the potential range of recoverable volumes for a Coalbed Methane (CBM) prospect is a necessary precursor to a successful development plan. Several key best practices were incorporated into a workflow to consistently assess the CBM potential of numerous prospective areas. For each area 3D static models were built based on available structural data. The models were geo-statistically populated with coal properties such as density and ash content. Correlations for other properties including gas content, permeability and Langmuir volume were developed. An analysis of the residual distribution between each correlation and its measurements was used to characterise the uncertainty in each. Several methods were considered to reproduce this uncertainty. These ranged from directly applying discrete trends, to geo-statistical property population. The effect of applying each on the predicted EUR was investigated. Reservoir simulation models of production pilots were built and history matched. Given the complexities of the coal reservoir and the non-uniqueness of the history match, further work was carried out to capture the remaining uncertainty and determine its impact on the model predictions. Experimental design (DOE) was used to generate a population of simulation models that sampled the uncertainty range. By using the measured pilot production as a filter, this population was reduced to include only those that matched the observed production. The final step was to optimise the placement of development wells. An algorithm that traded off the gain in gas recovery obtained by a tighter well spacing, against the increased cost associated with the extra wells was devised. The uncertainty in recovery given by this well spacing was tested using the reservoir simulation models. Although static and dynamic modelling of CBM reservoirs is quickly becoming routine in the industry, the best practices developed while building this workflow are novel solutions to several challenges that still confound the CBM modelling community. These best practices are not unique to the study area and could easily be applied to other areas. As such this paper should provide a useful reference to those about to undertake a CBM modelling project.
Decline curve analysis (DCA) has been a popular technique to forecast production and estimate reserves using production data for about seven decades. The reliability of forecasts and reserves estimates obtained through DCA ultimately depends on the quality of the data collected and on the experience and judgment of the analyst. It is important to know what assumptions are being made during DCA analysis and under which conditions the application is valid. Originally DCA have been applied to conventional oil reservoirs. Later on, mathematical manipulations of rate versus time and cumulative versus rate decline equations defined by Arps (1) as well as combinations with other analysis techniques have extended the application of DCA to conventional and unconventional gas reservoirs. The objective of this work is to investigate the applicability of DCA techniques for reserve estimation and production forecasting of Coal Seam Gas (CSG) wells using rate-cum and rate-time technique, and provide general guidelines and limitations for its use. This paper compares production forecasts from reservoir simulation and DCA. It also presents guidelines for selection of DCA candidates based on field/well maturity and shows the application of DCA in well and reservoir management, i.e. statistical results from Arps DCA analysis can be used to identify production enhancement potential and to generate typical distributions of EUR and decline parameters for specific fields. The results indicate that the ultimate recovery of CSG wells is dependent on such factors as gas content, seam thickness, well type, depth and well inseam length. Correlations developed to estimate ultimate recovery ranges can be used to estimate a well's EUR without the need for intensive dynamic reservoir simulation modeling and would enables efficient well-by-well analysis for development scenarios where it is not otherwise practical to perform DCA on each individual well. The results of this paper can be used as a helpful guide to understand the importance of various parameters to CSG wells elsewhere.
The risk profile for an unconventional resource play differs from a conventional opportunity in that the producibility, per well Estimated Ultimate Recovery (EUR) and Unit Technical Cost (UTC) are more important for identifying potential success than proving the presence of in-place volumes. Unconventional plays are often characterised by a large number of wells, lower density of subsurface data, large geographical extent and corresponding large range of uncertainty in subsurface parameters. The rapid economic screening of well design and spacing parameters for multiple subsurface realisations is integral in the planning stages of large unconventional plays. An additional complexity is the use of horizontal or complex geometry well designs which may limit or complicate the application of full field reservoir simulation methods. Recoverable volumes are strongly dependent on the proposed well design and spacing. These should be systematically evaluated and optimised by identifying the well density beyond which the incremental recovery and commerciality benefit begins to erode due to the extra well costs and/or interference between wells. A method for efficient analysis and comparison of complex well design and well spacing options has been designed to assist in unconventional play planning and evaluation. The method involves the automatic generation and analysis of a large number (thousands) of dynamic reservoir simulation models. The models are analysed systematically for major value drivers to: identify the most efficient well design and optimal spacing factors; select the most economic well designs; assess the impact of subsurface uncertainties; and assist in rig selection and surface planning.
Summary Control systems with feedback controllers are useful in reservoir simulation because they enable the maintenance of desired operating conditions of a field. This, in turn, helps to establish the value of implementing automated mechanisms in the field, and also in determining long-term field operating strategies. A generic controller framework is constructed within a reservoir simulator that enables the usage of different kinds of controller algorithms for managing a variety of field processes. In this study, three field processes are considered. First, average pressure within a reservoir region is maintained by adjusting the voidage-replacement ratio between a group of injectors and producers. Second, control systems are used for the prevention of gas/water coning for single and multiple wells. Finally, the average temperature within a reservoir region is maintained at a critical value by controlling flow into the formation, so as to operate with the desired mobility of heavy oil. Traditional proportional, integral, derivative (PID) controllers, as well as linear and nonlinear fuzzy controllers, are considered. The advantages and disadvantages of the approaches are discussed. Tuning control systems is a difficult process in practice. Several methods for tuning the parameters of these controllers are investigated, and rule-of-thumb values are suggested in this study. Synthetic and real reservoir models are used.
No abstract
The amount of condensed water should be carefully estimated in gas development plans, especially for fields with high CO2 content. This may affect the design of the well, surface facilities such as water treatment equipment and CO2 removal units, and flow assurance and corrosion control strategies. However, reliable prediction of the water content of gas with high CO2 level in contact with formation brine is still a challenge. The actual condensed water production data from a sour gas field offshore peninsular Malaysia has indicated that the condensed water levels are higher than estimated in the field development plan (FDP). The trend of water/gas ratios (WGR) across the field was analyzed versus depth, pressure, temperature and CO2 content of the reservoirs. To isolate the primary source of errors in the measured WGR from production tests, the measured water/condensate splits were verified with liquid/gas ratio (LGR) using calculated CGRs based on equation-of-state models. Salinity measurements of lower than 300 ppm were considered as fresh water. Equilibrium-based predictive tools were used to estimate the condensed water levels according to gas composition, pressure, temperature and water salinity. The results demonstrate significant differences between the EOS model prediction and the actual condensed water production. This might be due to the CO2 content of different reservoirs, in-situ pressure-temperature conditions, formation water salinity, the water vaporization around the wellbore due to the high gas velocity, the dry-out effect, or may be attributed to non-equilibrium concepts. This paper details the investigation of high condensed water production in the field and what cautions that must be taken into account in developing such fields. A new correlation was developed to predict the amount and trend of condensed water in the field with various CO2 levels. The study results can be used as a useful reference during the development planning of similar fields. Introduction The amount of water production from the gas fields prior to water breakthrough can be affected by both thermodynamic and kinetic bahaviour of the fluid flow in the reservoir and more particularly in the near wellbore area. Parameters, which affect the thermodynamic conditions and determine the water content of the gas, are temperature, pressure, formation water salinity, gas composition and CO2 content. Near wellbore area kinetic and non-equilibrium facts such as gas velocity, drawdown, gas stripping, water vaporization and dry out phenomena may also affect the amount of the produced water. All of these parameters must be considered to quantify the produced water prior the water breakrough time. During full field review of a particular sour gas field on the edge of the Malay Basin, it is observed that at the initial stage of reservoir life, many of the field's reservoirs were produced with constant higher water levels than the estimated condensed water based on the equilibrium conditions. Investigation has been undertaken to explain the source and amount of high water production in the field. The high condensed water levels should be considered carefully in the development plans for high CO2 levels gas fields in the region as they may impact the design of surface facilities, such as water treatment equipment, CO2 removal units and corrosion protection for pipelines.
Control systems with feedback controllers are useful in reservoir simulation as they enable the maintenance of desired operating conditions of a field. This in turn helps establish the value of implementing automated mechanisms in the field, and also in determining long term field operating strategies. A generic controller framework is constructed within a reservoir simulator that enables the usage of different kinds of controller algorithms for managing a variety of field processes. In this study, three field processes are considered. First, average pressure within a reservoir region is maintained by adjusting the voidage replacement ratio between a group of injectors and producers. Second, control systems are used for the prevention of gas/water coning for single and multiple wells. Finally, the average temperature within a reservoir region is maintained at a critical value by controlling flow into the formation, so as to operate with the desired mobility of heavy-oil. Traditional Proportional, Integral, Derivative (PID) controllers, as well as linear and nonlinear fuzzy controllers are considered. The advantages and disadvantages of the approaches are discussed. Tuning control systems is a difficult process in practice. Several methods for tuning the parameters of these controllers are investigated, and rule-of-thumb values are suggested in this study. Synthetic and real reservoir models are used.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.