Storage Operations Research Articles

Numerical Simulation Analysis of the Influence of Interlayer Quantity on the Long-Term Stable Operation of Gas Storage Facilities

Salt rock is considered as an ideal energy storage medium, and compressed air energy storage by a salt cavern can improve the utilisation efficiency of renewable energy. Salt rock in China mostly contains different interlayers, among which mudstone interlayers are the most common. At present, there are relatively few studies on the influence of mudstone interlayers on the long-term stable operation of gas storage. FLAC3D software was used to simulate the long-term operation of salt rock gas storage with different numbers of interlayers in the Yexian area of Pingdingshan. The results show that with the passage of time, the vertical displacement of the surrounding rock of the vertical single-cavity gas storage tank increases gradually. The maximum settlement value at the top of the surrounding rock is always greater than the maximum uplift value at the bottom. The horizontal displacement shows obvious symmetry with the vertical displacement at the top and bottom of the surrounding rock. The effect of the cyclic pressure interval on horizontal displacement is the same as that of vertical displacement. With the increase in the number of interlayers, the volume of the plastic zone gradually increases with the increase in the running time, and the increasing speed shows a growing trend.

Estimation of CO 2 saturation maps from synthetic seismic data using a deep-learning method with a multi-scale approach

In the context of Carbon Capture and Storage (CCS), monitoring the behaviour of CO 2 within subsurface reservoirs is pivotal for efficient and safe storage operations. This study proposes a new approach to enhance the accuracy of predicting CO 2 saturation maps directly from shot gathers, using a combination of deep learning (DL) and feature extraction methods. We employ a U-Net model, specifically tailored to solve regression tasks, and two-dimensional continuous wavelet transform (2D-CWT) to analyse shot gathers at different scales. We introduce a novel approach using multi-channel input data for the DL model, combining shot gathers and CWT images. The DL model was trained and tested using both single channel and multichannel input. The single channel datasets included shot gathers and CWT images at the first scale, while the multi-channel datasets integrated shot gathers with either two or four scales of CWT. We conducted a sensitivity analysis on the number of training epochs to compare the performance of the model with multichannel and single channel input. Additionally, a Transfer Learning approach was implemented to improve performance in noisy conditions by leveraging knowledge from a pre-trained model on noise-free images. Monte Carlo dropout was used to predict pixel-scale variability and provided a prediction variability with a standard deviation ranging from 0.019 to 0.194, enhancing the robustness of CO 2 saturation estimates. Our results suggest that combining feature extraction using 2D-CWT with the U-Net architecture improves the prediction performance of CO 2 saturation in a synthetic CCS model. Additionally, using multi-channel input instead of single channel is more efficient as it requires less training and prediction time to achieve comparable results. This confirms the effectiveness of 2D-CWT as a pre-processing technique, contributing valuable insights and advances in the field of data-driven geophysical inversion problems.

Integrated market scheduling with flexibility options

This work presents a generic optimization framework using advanced mixed-integer programming techniques to integrate day-ahead and balancing markets with distributed energy resources such as storage, electric vehicles, demand response, and Transmission and Distribution System Operators' coordination schemes. The day-ahead model determines optimal initial energy scheduling, while the balancing model optimizes energy and reserve products for three market designs with varying Transmission and Distribution System Operators’ coordination. The framework is applied to the interconnected Greek-Bulgarian-Romanian power system, considering 2030 installed capacities.The model outputs illustrate market coupling scenarios among Romania, Bulgaria, and Greece, highlighting clear price signals and distributed energy resources flexibility. Results reveal the significance of a diverse energy mix for energy security and show that Transmission and Distribution System Operators’ coordination significantly influences ancillary service prices, with reductions of up to 80 % in certain scenarios. Net demand values determine electricity flow direction. Flexibility providers like storage can cover up to 100 % of the upward congestion management requirements and 11 % of upward balancing energy, helping smooth energy allocation from intermittent renewables. The impact of electric vehicle penetration on generation scheduling is minimal. The proposed model offers valuable insights for system operators, market participants, and policymakers, enabling them to provide accurate price signals and optimize resource allocation.The integrated day-ahead and balancing market models support efficient renewable energy integration, emissions reduction, enhanced grid stability, and investment in low-carbon technologies. This aligns with the United Nations Framework Convention on Climate Change goals, contributing to the development of sustainable and resilient energy systems.

Green Order Sorting Problem in Cold Storage Solved by Genetic Algorithm

This study investigates the efficiency of cold storage warehouses and contributes to sustainable supply chain management by integrating eco-friendly practices into storage operations. In facilities for milk and its derivatives, unregulated order handling significantly increases energy consumption due to frequent door openings in the cooler. To address this challenge, we developed a novel mathematical model aimed at optimizing order sequences and minimizing energy costs, addressing a previously unexplored gap in the literature. A genetic algorithm (GA) was employed to solve this model, with careful consideration of carbon emissions generated during the algorithm’s execution. We utilized the Yates notation, an experimental design technique, to systematically optimize the GA’s parameters, ensuring robust and statistically valid results. This methodology enabled a thorough analysis of the factors influencing energy consumption. The findings enhance energy efficiency in cold storage warehouses, leading to reduced carbon dioxide emissions and fostering sustainable practices within supply chain management. Ultimately, this study successfully integrates green practices into cold storage operations, supporting broader sustainability objectives.

Techno-economic optimization and feasibility of PCM-based seasonal thermal energy storage systems for district heating and cooling

Phase change materials (PCM) are an attractive seasonal thermal energy storage solution for load shifting due to relatively high energy density. Nevertheless, the choice of the right design parameters, such as storage size and phase-change temperature, is nontrivial, as these are tightly linked to the expected seasonal operation of the storage. This paper presents a combined design and operation optimization framework for sizing PCM-based seasonal thermal storage, which can capture dynamic behaviour of the storage in sensible and latent phases, and its impact on the efficiency of connected heat pumps and chillers. The proposed method, formulated as a mixed integer quadratically-constrained programming problem, was applied to a case study of heating-dominated district. It was found that the optimal PCM melting temperature equals to the heating demand supply temperature, thus removing the need of a heat pump to discharge the storage. The optimal operation of storage requires a full charging and discharging cycle of both latent and liquid phase. Higher CO2 emission price does not lead to a significant reduction of CO2 emissions, but the storage size does, leading to a higher heating coverage ratio of the storage, and consequently CO2 emissions and cost savings, which can be further enhanced with PV integration. The economic analysis indicates that, unless other benefits such as storage volume reduction are accounted for, PCM cost should drop by a factor 4 to make this solution economically viable for seasonal storage.

Research on the Injection–Production Law and the Feasibility of Underground Natural Gas Storage in a Low-Permeability Acid-Containing Depleted Gas Reservoir

Depleted gas reservoirs are important places for the rebuilding of gas-storage reservoirs. In order to demonstrate the feasibility of constructing and operating such underground gas storage, a low-permeability gas-storage seepage model considering fracture development was developed and established. The model was solved using semi-analytical methods, and the pressure–response characteristics during natural gas injection were analyzed. The impact of gas injection volume on formation pressure has been clarified, and the calculation method for ultimate injection pressure has been determined. Additionally, through numerical simulation methods, the migration law of acidic gas during gas injection, the variation law of produced acidic gas concentration, and the main control factors affecting the concentration of the produced acidic gas were studied. Furthermore, measures to reduce the concentration of the acidic gas produced were proposed. Finally, injection and production plans were designed for typical depleted acidic gas reservoirs, simulating the operation of gas storage for 12 cycles. The results indicate that the quality of natural gas produced in the third cycle can meet the Class II standard for commercial natural gas. Through this study, the feasibility of constructing gas-storage facilities for acidic depleted gas reservoirs has been demonstrated, and injection and production strategies for this type of gas reservoir have been proposed.

Optimization of Offshore Saline Aquifer CO2 Storage in Smeaheia Using Surrogate Reservoir Models

Machine learning-based Surrogate Reservoir Models (SRMs) can replace/augment multi-physics numerical simulations by replicating the reservoir simulation results with reduced computational effort while maintaining accuracy compared with numerical simulations. This research will demonstrate SRMs’ potential in long-term simulations and optimization of geological carbon storage in a real-world geological setting and address challenges in big data curation and model training. The present study focuses on CO2 storage in the Smeaheia saline aquifer. Two SRMs were created using Deep Neural Networks (DNNs) to predict CO2 saturation and pressure over all grid blocks for 50 years. 18 million samples and 31 features, including reservoir static and dynamic properties, build the input data. Models comprise 3–5 hidden layers with 128–512 units apiece. SRMs showed a runtime improvement of 300 times and an accuracy of 99% compared to the 3D numerical simulator. The genetic algorithm was then employed to determine the optimal rate and duration of CO2 injection, which maximizes the volume of injected CO2 while ensuring storage operations’ safety through constraints. The optimization continued for the reproduction of 100 generations, each containing 100 individuals, without any hyperparameter tuning. Finally, the optimization results confirm the significant potential of Smeaheia for storing 170 Mt CO2.

Compositional simulation of coupled transport mechanisms in fractured-vuggy underground gas storage: A case study of LWC in the Sichuan Basin

Compositional models were built to simulate the underground gas storage (UGS) operation, which is characterized by multiple cycles of rapid injection and production. Based on the validated model of the fractured-vuggy underground gas storage Lao Weng-Chang (LWC), this study evaluated the individual and coupling effects of stress sensitivity, relative permeability hysteresis, and high-speed non-Darcy flow on UGS operation. It was found that stress sensitivity is the main controlling factor behind the reduced working gas volume, and it led to a 6.07% decline in working gas volume. Relative permeability hysteresis is the determining factor for the storage capacity, and it led to a 9.05% decline in storage capacity. The high-speed non-Darcy flow only led to a 0.16% reduction in storage capacity but led to a 4.19% decline in working gas volume. A higher stress sensitivity coefficient and increased residual gas saturation both lead to greater losses in both storage capacity and working gas volume. Coupling stress sensitivity and relative permeability hysteresis would have a more pronounced effect on working gas volume than when considered individually. Considering the high-speed non-Darcy flow further decreases working gas volume. The gas production per unit pressure drop will also decrease with the increasing Reynolds number, and there exists a critical value of Re where the decrease significantly accelerates.

Research on Electric Hydrogen Hybrid Storage Operation Strategy for Wind Power Fluctuation Suppression

Research on Electric Hydrogen Hybrid Storage Operation Strategy for Wind Power Fluctuation Suppression

Dynamic feed scheduling for optimised anaerobic digestion: An optimisation approach for better decision-making to enhance revenue and environmental benefits

Anaerobic digestion (AD) offers a sustainable solution for clean energy production, with the potential for significant revenue enhancement through enhanced decision-making. However, the complexity and limited flexibility of AD systems pose challenges in developing reliable optimisation methods. Changing feeding strategies provides opportunities for efficient feedstock utilisation and optimal gas production, especially in volatile gas markets.To provide better decision-making tools in AD for energy production, we propose an integrated site model for the dynamic behaviour of the AD process in a biomethane-to-grid system and optimise production based on predicted gas prices. The model includes methods for optimal feed co-digestion strategies and integrates these results into a scheduling model to identify the optimal feedstock acquisition, feeding pattern, and potential gas storage operation considering feedstock availability, properties, sustainability, and fluctuating gas demand under different pricing variations.The methodology was tested on a 150 tonnes per day farm-scale AD plant in the UK, processing energy crops and manure considering both environmental (global warming potential) and economic objectives. The results showed strong adaptability of the proposed feeding schedule to the general trend of gas prices over time. To address the challenge of immediate price peaks, typically unattainable due to the system's sluggish behaviour and high retention times, the impacts of on-site storage were explored, leading to annual revenue increases ranging from 2 % to 7.4 %, depending on the pricing scheme, which translates to a significant boost in terms of revenue.

Realizing In-Memory Computing using Reliable Differential 8T SRAM for Improved Latency

Traditional von Neumann computing architectures suffer from high energy and lower speed as compared to the requirements of modern applications like those required in neural network accelerators. A modified differential eight transistor (8 + T) static random access memory (SRAM)-based in-memory computing (IMC) structure was presented for realizing bit-wise Boolean logic operations. The 8 + T SRAM-IMC is designed at the 32 nm technology node with throughput of 2.1849, 2.4815, 2.5795, 2.6240, 2.6495, 2.6619, 2.6690, 2.6732, and 2.6749 giga outputs per second for 0.5 to 1.3 V supply voltage range, respectively. The differential 8T cell used to implement logic operations supports NAND and NOR operations with minimal overhead while also performing the standard storage operation with added stability over the conventional 6T and 8T SRAM cells. The SRAM-IMC offers reliable Boolean logic operations by using asymmetric sensing strategy for all process corners, TT, SS, SF, FS, and FF for an operating temperature range of 220 K to 400 K. Monte Carlo simulation considering global threshold voltage deviation of 50 mV was performed for various operating conditions to study the impact of variations on design parameters such as latency.

Discovery and Reconstruction of the Remains of the Beacon-Equipped Hollow Enemy Towers along the Ming Great Wall

Hollow Enemy Towers, as iconic structures of the Ming Great Wall, are renowned for their roles in defense surveillance, weapon storage, and firearm operation. Recent studies have indicated that certain Hollow Enemy Towers along the Ji Town section of the Ming Great Wall also serve the function of Beacon Towers for beacon signaling. However, previous studies have not definitively determined if these towers were distinctively marked, nor have they provided a comprehensive account of their current distribution and original historical appearance. This paper initially examined the historical documentation of white lime markings employed on the outer walls of certain Hollow Enemy Towers, which served as Beacon Towers during the middle and late Ming periods. Utilizing multidisciplinary methodologies, this research identified remains of lime markings of the Beacon-Equipped Hollow Enemy Towers along the Ji Town section of the Ming Great Wall, illustrating their extensive distribution. We analyzed the material composition and construction techniques of the lime mortar. This analysis clarifies the scope of lime plastering on the exterior walls of these towers and offers a point of reference for restoring their original historical appearance. The results make a significant supplement to the types of signaling structures on the Great Wall, enriching existing understanding of the original appearance of the Great Wall’s historical landscape.

Optimization of shared energy storage configuration for village-level photovoltaic systems considering vehicle charging management

Distributed renewable energy is more abundant in rural areas, and a large amount of distributed photovoltaic grid-connected power brings challenges to the stable of the power grid. How to promote the self-generation and self-consumption of distributed renewable energy has become an urgent problem. In this paper, a village-level distributed photovoltaic power generation system including energy storage and electric vehicles is constructed. With the goal of minimizing the photovoltaic grid-connected power and maximizing the annual comprehensive revenue, the planning model of energy storage capacity allocation for village-level distributed power generation system is constructed. Considering the charging management for different numbers of electric vehicles, the optimal energy storage capacity allocation strategy is solved using the improved particle swarm algorithm.Five scenarios are set up as examples to be analyzed.The conclusions are:(1)After the configuration of a reasonable energy storage,the grid-connected generation of the system is significantly reduced,and the annual comprehensive revenue of the system is significantly increased.(2)After considering charging management for different numbers of electric vehicles, the capacity and power of the energy storage configuration can be significantly reduced, thus reducing the energy storage investment cost and operation and maintenance cost, and further significantly improving the annual comprehensive revenue of this system.

Simulation of bench-scale CO2 injection using a coupled continuum-discrete approach

Unintended releases of CO2 from carbon capture and storage operations presents the risk of atmospheric emissions and groundwater or surface water quality impacts. Given the potential impacts, it is valuable to have tools capable of predicting groundwater concentrations and likely pathways of CO2 migration in the subsurface. Traditional multiphase flow models struggle to simulate the discontinuous flow expected at leakage sites. This work applied a coupled continuum-discrete model, ET-MIP, to simulate a bench-scale injection of CO2. Results demonstrate the capability of ET-MIP to accurately capture gas fingering behaviour, and the complexity of multicomponent mass transfer observed in the experiment. Simulations were computationally efficient, allowing for the use of multiple displacement pressure realizations. CO2 migration in the subsurface was shown to be sensitive to mass transfer, as i) increased groundwater velocity can dissolve leaked CO2 prior to reaching the surface and ii) background dissolved gases in the subsurface can impact the rate of upwards gas movement, gas distribution, and the composition and persistence of the gas phase. The sensitivity to mass transfer suggests it may be preferable to monitor for low-solubility gases in the source mixture rather than CO2. These findings are applicable to other gases in the subsurface, such as hydrogen or methane migrating from geoenergy wells.

Comparing green hydrogen and green ammonia as energy carriers in utility-scale transport and subsurface storage

Many of the challenges associated with utility-scale hydrogen transport and storage relate to its low density, high diffusivity, and the risk of hydrogen embrittlement, motivating consideration to integrating ammonia as an energy carrier. Compared to hydrogen, ammonia is more compatible with pipeline materials and delivers energy at higher density. Ammonia is also a mature industry with a greater extent of established pipeline networks and regulations that may accelerate hydrogen transitions and penetration in energy grids. However, converting hydrogen produced by renewable-driven electrolysis into ammonia (and back to hydrogen, depending on end use) complicates logistics, and associated energy and resource demands may offset the green hydrogen's carbon neutrality. This work outlines core considerations for the use of hydrogen vs. ammonia during transport and storage operations, with an emphasis on green hydrogen or green ammonia pathways coupled to pipeline transport and underground storage. We compare tradeoffs in pipeline infrastructure and operations; subsurface storage options; and project economics. We also evaluate round-trip efficiencies (RTE) for both pathways, which indicate that hydrogen is more attractive from an energy efficiency perspective for hydrogen end-use applications due to the efficiency penalties of initial ammonia synthesis and subsequent cracking, but RTE's for ammonia transport and storage are comparable to hydrogen for direct use or ammonia-to-power systems. The tradeoffs presented in this work would need to be considered on a case-by-case basis, but indicate that selective use of ammonia as an energy-dense hydrogen carrier could support decarbonization goals in industry and hydrogen economies.

Research on the collaborative operation strategy of shared energy storage and virtual power plant based on double layer optimization

Large-scale access to distributed energy resources leads to new energy consumption problems and safe operation risks in the power system. Virtual power plants and shared energy storage are effective ways to promote the flexible consumption of distributed energy resources and improve the reliability and economy of power system operation. Based on the concept of sharing economy and considering the complementary characteristics of source and load resources between different virtual power plants, this paper focuses on the optimisation of shared energy storage and multi-virtual power plant operation. Firstly, distributed wind power, distributed photovoltaic and flexible load resources are aggregated into virtual power plants to analyze the cooperative operation mode of shared energy storage and multi-virtual power plant systems. Secondly, a two-layer decision model for shared energy storage configuration and multi-VPP system operation optimisation is constructed, with the upper model solving the optimal energy storage configuration scheme by maximising the revenue of the shared energy storage operator, and the lower model optimising the multi-VPP system operation strategy by minimising the total operation cost, the maximum amount of new energy consumed, and the minimum amount of carbon emission as the multi-objective optimisation strategy. Finally, a simulation analysis is carried out, and the results show that compared with the independent operation mode of each virtual power plant, the model proposed in this paper increases the annual profit of the shared energy storage operator by 7180¥, reduces the operating cost of the VPP system by 7.08 %, improves the rate of renewable energy consumption by 0.82 %, and reduces the annual carbon emission by 54.91 t, which realizes the mutual benefits of the virtual power plant and the shared energy storage.

Loss characteristics and optimization method of a compressed air energy storage radial turbine with nozzle governing

During the operation of the compressed air energy storage (CAES) system, a discrepancy exists between the air storage pressure and the turbine inlet pressure. At the same time, to ensure that the turbine operates efficiently when the system load varies, the turbine needs to be regulated reasonably. In this paper, the influence of three nozzle governing (NG) parameters on turbine performance during the NG process is investigated: the inlet pressure of the NG region, the area of the NG region, and the base pressure. The flow characteristics and loss mechanisms under partial admission conditions and NG conditions are also analyzed. Furthermore, the NG process is optimized by using a transition section. The results show that the loss under the NG condition with low inlet pressure at the NG region is similar to the partial admission loss, both of which are mainly affected by the area of the NG region. In both working conditions, vortex structures appear at the stator outlet of the NG region. At the same time, the mixing of different airflows and the flow separation phenomenon within the rotor leads to a significant increase in entropy production rate. The transition section can attenuate the leakage and reduce the entropy production rate in the radial gap between the stator and rotor, increasing turbine efficiency by up to 1.1 % at maximum when using the transition section.

Stochastic dual dynamic programming for optimal power flow problems under uncertainty

Planning in the power sector has to take into account the physical laws of alternating current (AC) power flows as well as uncertainty in the data of the problems, both of which greatly complicate optimal decision making. We propose a computationally tractable framework to solve multi-stage stochastic optimal power flow (OPF) problems in AC power systems. Our approach uses recent results on dual convex semi-definite programming (SDP) relaxations of OPF problems in order to adapt the stochastic dual dynamic programming (SDDP) algorithm for problems with a Markovian structure. We show that the usual SDDP lower bound remains valid and that the algorithm converges to a globally optimal policy of the stochastic AC-OPF problem as long as the SDP relaxations are tight. To test the practical viability of our approach, we set up a case study of a storage siting, sizing, and operations problem. We show that the convex SDP relaxation of the stochastic problem is usually tight and discuss ways to obtain near-optimal physically feasible solutions when this is not the case. The algorithm finds a physically feasible policy with an optimality gap of 3% and yields a significant added value of 27% over a rolling deterministic policy, which leads to overly optimistic policies and underinvestment in flexibility. This suggests that the common industry practice of assuming direct current and deterministic problems should be reevaluated by considering models that incorporate realistic AC flows and stochastic elements in the data as potentially more realistic alternatives.

Optimization of geological carbon storage operations with multimodal latent dynamic model and deep reinforcement learning

Identifying the time-varying control schemes that maximize storage performance is critical to the commercial deployment of geological carbon storage (GCS) projects. However, the optimization process typically demands extensive resource-intensive simulation evaluations, which poses significant computational challenges and practical limitations. In this study, we presented the multimodal latent dynamic (MLD) model, a novel deep learning framework for fast flow prediction and well control optimization in GCS operations. The MLD model implicitly characterizes the forward compositional simulation process through three components: a representation module that learns compressed latent representations of the system, a transition module that approximates the evolution of the system states in the low-dimensional latent space, and a prediction module that forecasts the flow responses for given well controls. A novel model training strategy combining a regression loss and a joint-embedding consistency loss was introduced to jointly optimize the three modules, which enhances the temporal consistency of the learned representations and ensures multi-step prediction accuracy. Unlike most existing deep learning models designed for systems with specific parameters, the MLD model supports arbitrary input modalities, thereby enabling comprehensive consideration of interactions between diverse types of data, including dynamic state variables, static spatial system parameters, rock and fluid properties, as well as external well settings. Since the MLD model mirrors the structure of a Markov decision process (MDP) that computes state transitions and rewards (i.e., economic calculation for flow responses) for given states and actions, it can serve as an interactive environment to train deep reinforcement learning agents. Specifically, the soft actor-critic (SAC) algorithm was employed to learn an optimal control policy that maximizes the net present value (NPV) from the experiences gained by continuous interactions with the MLD model. The efficacy of the proposed approach was first compared against commonly used simulation-based evolutionary algorithm and surrogate-assisted evolutionary algorithm on a deterministic GCS optimization case, showing that the proposed approach achieves the highest NPV, while reducing the required computational resources by more than 60%. The framework was further applied to the generalizable GCS optimization case. The results indicate that the trained agent is capable of harnessing the knowledge learned from previous scenarios to provide improved decisions for newly encountered scenarios, demonstrating promising generalization performance.

On the Significance of Hydrate Formation/Dissociation during CO2 Injection in Depleted Gas Reservoirs

Summary The study aims to quantitatively assess the risk of hydrate formation within the porous formation and its consequences on injectivity during storage of CO2 in depleted gas reservoirs considering low temperatures caused by the Joule-Thomson (JT) effect and hydrate kinetics. Hydrates formed during CO2 storage operation can occupy porous spaces in the reservoir rock, reducing the rock’s permeability and thus becoming a hindrance to the storage project. The aim was to understand which mechanisms can mitigate or prevent the formation of hydrates. The key mechanisms we studied included water dry-out, heat exchange with surrounding rock formation, and capillary pressure. A semicompositional thermal reservoir simulator is used to model the fluid and heat flow of CO2 through a reservoir initially composed of brine and methane. The simulator can model the formation and dissociation of both methane and CO2 hydrates using kinetic reactions. This approach has the advantage of computing the amount of hydrate deposited and estimating its effects on the porosity and permeability alteration. Sensitivity analyses are also carried out to investigate the impact of different parameters and mechanisms on the deposition of hydrates and the injectivity of CO2. Simulation results for a simplified model were verified with results from the literature. The key results of this work are as follows: (1) The JT effect strongly depends on the reservoir permeability and initial pressure and could lead to the formation of hydrates within the porous media even when the injected CO2 temperature was higher than the hydrate equilibrium temperature; (2) the heat gain from underburden and overburden rock formations could prevent hydrates formed at late time; (3) permeability reduction increased the formation of hydrates due to an increased JT cooling; and (4) water dry-out near the wellbore did not prevent hydrate formation. Finally, the role of capillary pressure was quite complex, as it reduced the formation of hydrates in certain cases and increased in other cases. Simulating this process with heat flow and hydrate reactions was also shown to present severe numerical issues. It was critical to select convergence criteria and linear system tolerances to avoid large material balance and numerical errors.