Material-Balance Techniques for Coal-Seam and Devonian Shale Gas Reservoirs With Limited Water Influx

A New Evaluation Method of Recoverable Reserves and Its Application in Carbonate Gas Reservoirs

The propagation pattern of pressure drawdown effectively reflects the recoverable reserves range around the gas well and serves as a crucial basis for development strategies. However, it is not easy to detect the pressure propagation boundary near the producing well, especially in low-permeability reservoirs where the drainage radius is small. Physical simulation experiments can serve as a crucial method as the whole pressure profile and gas rate can be obtained in real time. Using long core plugs with permeabilities of 2.300 mD, 0.486 mD, and 0.046 mD, physical simulation experiments were carried out under varying initial water saturation (Swi) conditions of 0%, 20%, 40%, and 55% to observe the dynamic variations in pressure profiles of the core plugs during pressure depletion. Based on the material balance equation and pressure profile characteristics of the core plugs, a method for evaluating recoverable reserves within a well-spacing radius through laboratory experiments was proposed and performed. Mechanism analysis was conducted based on mercury injection tests, and suggestions for enhancing gas recovery were presented. Research findings indicate that lower permeability, higher initial water saturation, and higher abandonment gas rates result in reduced reserve utilization range and degree. Under abandoned gas rate conditions, for type I and II rocks, the pore radius is primarily distributed between 0.1 and 1 μm, the pressure drawdown can reach the well-spacing radius of 600 m, and the ultimate recovery efficiencies are more than 70.6%. For type III rocks, the pore radius mainly falls below 0.1 μm, the drainage radius is smaller than 10 m with Swi greater than 40%, and the ultimate recovery is below 10%. This paper provides an experimental method for recoverable reserves evaluation while formulating gas reservoir development strategies before well testing.

Study on development methods of different types of gas wells in tight sandstone gas reservoirs

Reasonable production allocation of tight sandstone gas reservoirs is an important basis for efficient development of gas wells. Taking Block XX in Ordos Basin as an example, the modified flowing material balance equation was established considering the variation of gas viscosity and compression coefficient, the advantages and disadvantages of the method were discussed, and a reasonable production allocation process for gas wells was developed. The results show that: ① The commonly used flow material balance method ignores the change of natural gas compression coefficient, viscosity and deviation coefficient in the production process. The slope of the relationship curve between bottom hole pressure and cumulative production and the slope of the relationship curve between average formation pressure and cumulative production are not equal After considering this change. Compared with the results calculated by the material balance method, the results calculated by the flow material balance method are smaller. ② The production of 660 gas wells in the study area during stable production period is verified. Compared with the open flow method, the dynamic reserve allocation method is better, with an error of 0.06%. ③ The new method in this paper is used to allocate production for different types of gas wells. The cumulative production of different types of gas wells shows different degrees of increase. The I, II, III and IV types of gas wells increase by 32.26%, 30.29%, 23.58% and 25.07% respectively. This study provides technical support for dynamic reserve calculation and reasonable production allocation of gas wells in the study area, and has important guiding significance for the formulation of reasonable development plan and economic and efficient development of tight sandstone gas reservoirs.

Modified Flowing Material Balance Equation for Shale Gas Reservoirs

To determine original gas-in-place, this study establishes a flowing material balance equation based on the improved material balance equation for shale gas reservoirs. The method considers the free gas in the matrix and fracture, the dissolved gas in kerogen, and the pore volume occupied by adsorbed phase simultaneously, overcoming the problem of incomplete consideration in the earlier models. It also integrates the material balance method with the flowing material balance method to obtain the average formation pressure, eliminating the problem with the previous method where shutting down of wells was needed to monitor the formation pressure. The volume of the adsorbed gas on the ground is converted into volume of the adsorbed phase in the formation using the volume conservation method to characterize the pore volume occupied by the adsorbed phase, which solves the problem of the previous model that the adsorbed phase was neglected in the pore volume. The model proposed in this study is applied to the Fuling Shale Gas Field in southwest China and compared with other flowing material balance equations, and the results show that the single-well control area calculated by the model proposed in this study is closer to the real value, indicating that the calculations in this study are more accurate. Furthermore, the calculations show that the dissolved gas takes up a large fraction of the total reserves and cannot be ignored. The sensitivity analyses of critical parameters demonstrate that (a) the greater the porosity of the fracture, the greater the free gas storage; (b) the values of Langmuir volume and TOC can significantly affect the results of the reservoir calculation; and (c) the adsorbed phase occupies a smaller pore volume when the Langmuir volume is smaller, the Langmuir pressure is higher, or the adsorbed phase density is higher. The findings of this study can provide better understanding of the necessity to take into account the dissolved gas in the kerogen, the pore volume occupied by the adsorbed phase, and the fracture porosity when evaluating reserves. The method could be applied to the calculation of pressure, recovery of free gas phase and adsorbed phase, original gas-in-place, and production predictions, which could help for better guidance of reserve potential estimations and development strategies of shale gas reservoirs.

A Multivariate Long Short-Term Memory Neural Network for Coalbed Methane Production Forecasting

Owing to the importance of coalbed methane (CBM) as a source of energy, it is necessary to predict its future production. However, the production process of CBM is the result of the interaction of many factors, making it difficult to perform accurate simulations through mathematical models. We must therefore rely on the historical data of CBM production to understand its inherent features and predict its future performance. The objective of this paper is to establish a deep learning prediction method for coalbed methane production without considering complex geological factors. In this paper, we propose a multivariate long short-term memory neural network (M-LSTM NN) model to predict CBM production. We tested the performance of this model using the production data of CBM wells in the Panhe Demonstration Area in the Qinshui Basin of China. The production of different CBM wells has similar characteristics in time. We can use the symmetric similarity of the data to transfer the model to the production forecasting of different CBM wells. Our results demonstrate that the M-LSTM NN model, utilizing the historical yield data of CBM as well as other auxiliary information such as casing pressures, water production levels, and bottom hole temperatures (including the highest and lowest temperatures), can predict CBM production successfully while obtaining a mean absolute percentage error (MAPE) of 0.91%. This is an improvement when compared with the traditional LSTM NN model, which has an MAPE of 1.14%. In addition to this, we conducted multi-step predictions at a daily and monthly scale and obtained similar results. It should be noted that with an increase in time lag, the prediction performance became less accurate. At the daily level, the MAPE value increased from 0.24% to 2.09% over 10 successive days. The predictions on the monthly scale also saw an increase in the MAPE value from 2.68% to 5.95% over three months. This tendency suggests that long-term forecasts are more difficult than short-term ones, and more historical data are required to produce more accurate results.

Effect of Kerogen Thermal Maturity on Methane Adsorption Capacity: A Molecular Modeling Approach

The presence of kerogen in source rocks gives rise to a plethora of potential gas storage mechanisms. Proper estimation of the gas reserve requires knowledge of the quantities of free and adsorbed gas in rock pores and kerogen. Traditional methods of reserve estimation such as the volumetric and material balance approaches are insufficient because they do not consider both the free and adsorbed gas compartments present in kerogens. Modified versions of these equations are based on adding terms to account for hydrocarbons stored in kerogen. None of the existing models considered the effect of kerogen maturing on methane gas adsorption. In this work, a molecular modeling was employed to explore how thermal maturity impacts gas adsorption in kerogen. Four different macromolecules of kerogen were included to mimic kerogens of different maturity levels; these were folded to more closely resemble the nanoporous kerogen structures of source rocks. These structures form the basis of the modeling necessary to assess the adsorption capacity as a function of the structure. The number of double bonds plus the number and type of heteroatoms (O, S, and N) were found to influence the final configuration of the kerogen structures, and hence their capacity to host methane molecules. The degree of aromaticity increased with the maturity level within the same kerogen type. The fraction of aromaticity gives rise to the polarity. We present an empirical mathematical relationship that makes possible the estimation of the adsorption capacity of kerogen based on the degree of polarity. Variations in kerogen adsorption capacity have significant implications on the reservoir scale. The general trend obtained from the molecular modeling was found to be consistent with experimental measurements done on actual kerogen samples. Shale samples with different kerogen content and with different maturity showed that shales with immature kerogen have small methane adsorption capacity compared to shales with mature kerogen. In this study, it is shown for the first time that the key factor to control natural gas adsorption is the kerogen maturity not the kerogen content.

Forecasting of Coalbed Methane Daily Production Based on T-LSTM Neural Networks

Accurately forecasting the daily production of coalbed methane (CBM) is important forformulating associated drainage parameters and evaluating the economic benefit of CBM mining. Daily production of CBM depends on many factors, making it difficult to predict using conventional mathematical models. Because traditional methods do not reflect the long-term time series characteristics of CBM production, this study first used a long short-term memory neural network (LSTM) and transfer learning (TL) method for time series forecasting of CBM daily production. Based on the LSTM model, we introduced the idea of transfer learning and proposed a Transfer-LSTM (T-LSTM) CBM production forecasting model. This approach first uses a large amount of data similar to the target to pretrain the weights of the LSTM network, then uses transfer learning to fine-tune LSTM network parameters a second time, so as to obtain the final T-LSTM model. Experiments were carried out using daily CBM production data for the Panhe Demonstration Zone at southern Qinshui basin in China. Based on the results, the idea of transfer learning can solve the problem of insufficient samples during LSTM training. Prediction results for wells that entered the stable period earlier were more accurate, whereas results for types with unstable production in the early stage require further exploration. Because CBM wells daily production data have symmetrical similarities, which can provide a reference for the prediction of other wells, so our proposed T-LSTM network can achieve good results for the production forecast and can provide guidance for forecasting production of CBM wells.

Multi-factor controls on initial gas production pressure of coalbed methane wells in Changzhi-Anze block, Central-Southern of Qinshui Basin, China

Adsorption and desorption of coalbed methane are generally at a dynamic equilibrium state under the undisturbed coal reservoir. However, as the reservoir pressure drops to a certain value during the extraction of coalbed methane, the equilibrium state is destroyed and thus more coalbed methane desorbs and escapes from coal to wellbore. Here the corresponding bottom-hole fluid pressure is called initial gas production pressure (IGPP) in the development practice of coalbed methane wells. This paper, which has taken Changzhi-Anze block in the central-southern part of Qinshui basin as the study object, addresses the distribution characteristic and control factors of IGPP of coalbed methane wells and then explores the key factors affecting IGPP using grey correlation analysis theory. The results indicate that IGPP varies from 1.09 MPa to 6.57 MPa, showing a distribution law with high in the middle and low in the west and east of the study area, which presents a similar distribution characteristic with burial depth, thickness, coal rank, gas content, original reservoir pressure, and in-situ stress. Further, through grey correlation analysis, it concludes that the correlation degrees of control factors affecting IGPP of coalbed methane wells in the descending order are decline rate of working fluid level, water yield before gas production, reservoir pressure, coal thickness, coal rank, minimum horizontal principal stress, burial depth, and gas content. Among these factors, engineering factors, including decline rate of working fluid level and water yield before gas production, present a key controlling effect, because they can determine the smooth migration pathway directly during initial water production. And another key factor, original reservoir pressure also builds strong and positive correlation with IGPP under the interaction of other geology and reservoir factors, revealing the capability of gas desorption and the transmission of pressure drops.

Application of temperature-dependent adsorption models in material balance calculations for unconventional gas reservoirs

Langmuir isotherm is the most common adsorption model used in the prediction of gas adsorption in most shale and coal bed methane reservoirs. However, due to the underlying assumption of single temperature, it fails to model gas adsorption where temperature differential exists in the reservoir. To address this shortcoming, temperature-dependent gas adsorption models have been incorporated into material balance calculations for accurate prediction of original gas in place as well as determining both average reservoir pressure and future performance in coal/shale gas reservoirs. The material balance equation has been expressed as a straight line with both Bi-Langmuir and Exponential models used in prediction of gas adsorption rather than the Langmuir isotherm. With this methodology, several adsorption capacities can be obtained at multiple temperatures which will allow for better estimation of original gas in place and future gas production. The results from this works show that temperature-dependent gas adsorption models can be used in place of Langmuir isotherm to account for the effect of temperature variations and more accurate representation of the adsorption of gas in coal/shale gas reservoirs.

Volumetric Measurements of Methane-Coal Adsorption and Desorption Isotherms—Effects of Equations of State and Implication for Initial Gas Reserves

This study presents the effects of equations of state (EOSs) on methane adsorption capacity, sorption hysteresis and initial gas reserves of a medium volatile bituminous coal. The sorption experiments were performed, at temperatures of 25 °C and 40 °C and up to 7MPa pressure, using a high-pressure volumetric analyzer (HPVA-II). The measured isotherms were parameterized with the modified (three-parameter) Langmuir model. Gas compressibility factors were calculated using six popular equations of state and the results were compared with those obtained using gas compressibility factors from NIST-Refprop® (which implies McCarty and Arp’s EOS for Z-factor of helium and Setzmann and Wagner’s EOS for that of methane). Significant variations were observed in the resulting isotherms and associated model parameters with EOS. Negligible hysteresis was observed with NIST-refprop at both experimental temperatures, with the desorption isotherm being slightly lower than the adsorption isotherm at 25 °C. Compared to NIST-refprop, it was observed that equations of state that gave lower values of Z-factor for methane resulted in “positive hysteresis”, (one in which the desorption isotherm is above the corresponding adsorption curve) and the more negatively deviated the Z-factors are, the bigger the observed hysteresis loop. Conversely, equations of state that gave positively deviated Z-factors of methane relatively produced “negative hysteresis” loops where the desorption isotherms are lower than the corresponding adsorption isotherms. Adsorbed gas accounted for over 90% of the calculated original gas in place (OGIP) and the larger the Langmuir volume, the larger the proportion of OGIP that was adsorbed.

Cross-formational flow of water into coalbed methane reservoirs: controls on relative permeability curve shape and production profile

Coalbed methane (CBM) wells tend to produce large volumes of water, especially when there is hydraulic connectivity between coalbed and nearby formations. Cross-formational flow between producing coal and adjacent formations can have significant production and environmental implications, affecting economic viability of production from these shallow reservoirs. Such flows can also affect how much gas can be removed from a coalbed prior to mining and thus can have implications for methane control in mining as well. The aim of this paper is to investigate the impact of water flow from an external source into coalbed on production performance and also on reservoir variables including cleat porosity and relative permeability curves derived from production data analysis. A reservoir model is constructed to investigate the production performance of a CBM well when cross-formational flow is present between the coalbed and the overlying formation. Results show that cleat porosity calculated by analysis of production data can be more than one order of magnitude higher than actual cleat porosity. Due to hydraulic connectivity, water saturation within coalbed does not considerably change for a period of time, and hence, the peak of gas production is delayed. Upon depletion of the overlying formation, water saturation in coalbed quickly decreases. Rapid decline of water saturation in the coalbed corresponds to a sharp increase in gas production. As an important consequence, when cross-flow is present, gas and water relative permeability curves, derived from simulated production data, have distinctive features compared to the initial relative permeability curves. In the case of cross-flow, signatures of relative permeability curves are concave downward and low gas permeability for a range of water saturation, followed by rapid increase afterward for water and gas, respectively. The results and analyses presented in this work can help to assess the impact of cross-formational flow on reservoir variables derived from production data analysis and can also contribute to identifying hydraulic connectivity between coalbed and adjacent formations.