Gas Dispersion Model Research Articles

Gas Leakage Risk Management Of Biogas Powerplant Using Aloha Gas Dispersion Modelling And Bowtie Analysis

Biogas, a fuel derived from organic waste, has gained popularity as a sustainable energy source. The adaptability of biogas allows its utilization for electricity and heating. This study aims to evaluate the risk of biogas leakage using ALOHA gas dispersion modeling and Bowtie analysis to identify hazard zones and evaluate the effectiveness of preventive and mitigation measures. The research method using Gas dispersion modeling in ALOHA software can simulate the spread of leaked biogas, identify potential hazard zones, and inform emergency response planning. The results show that gas dispersion reaches up to 296 meters from the leak source in a worst-case scenario without ignition, with significant implications for safety. The ignition scenario produces high thermal radiation, further exacerbating the risk. In conclusion, these findings highlight the intolerable risk of biogas leakage, requiring comprehensive risk mitigation measures. This study recommends the implementation of a robust control system, routine maintenance, and emergency response plan to ensure the safety and sustainability of biogas facility operations

Reconstruction of the Chemical Gas Concentration Distribution Using Partial Convolution-Based Image Inpainting.

An interpolation method, which estimates unknown values with constrained information, is based on mathematical calculations. In this study, we addressed interpolation from an image-based perspective and expanded the use of image inpainting to estimate values at unknown points. When chemical gas is dispersed through a chemical attack or terrorism, it is possible to determine the concentration of the gas at each location by utilizing the deployed sensors. By interpolating the concentrations, we can obtain the contours of gas concentration. Accurately distinguishing the contours of a contaminated region from a map enables the optimal response to minimize damage. However, areas with an insufficient number of sensors have less accurate contours than other areas. In order to achieve more accurate contour data, an image inpainting-based method is proposed to enhance reliability by erasing and reconstructing low-accuracy areas in the contour. Partial convolution is used as the machine learning approach for image-inpainting, with the modified loss function for optimization. In order to train the model, we developed a gas diffusion simulation model and generated a gas concentration contour dataset comprising 100,000 contour images. The results of the model were compared to those of Kriging interpolation, one of the conventional spatial interpolation methods, finally demonstrating 13.21% higher accuracy. This suggests that interpolation from an image-based perspective can achieve higher accuracy than numerical interpolation on well-trained data. The proposed method was validated using gas concentration contour data from the verified gas dispersion modeling software Nuclear Biological Chemical Reporting And Modeling System (NBC_RAMS), which was developed by the Agency for Defense Development, South Korea.

SO2 emission rates and incorporation into the air pollution dispersion forecast during the 2021 eruption of Fagradalsfjall, Iceland

During the low-effusion rate Fagradalsfjall eruption (19 March – 18 September 2021), the emission of sulfur dioxide (SO2) was frequently measured using ground-based UV spectrometers. The total SO2 emitted during the entire eruption was 970 ± 540 kt, which is only about 6% of the SO2 emitted during the similar length Holuhraun eruption (2014–2015). The eruption was divided into five phases based on visual observations, including the number of active vents and the occurrence of lava fountaining. The SO2 emission rate ranged from 44 ± 19 kg/s in Phase 2 to 85 ± 29 kg/s in Phase 5, with an average of 64 ± 34 kg/s for the entire eruption. There was notable variability in SO2 on short timescales, with measurements on 11 August 2021 ranging from 17 to 78 kg/s.SO2 flux measurements were made using scanning DOAS instruments located at different distances from and orientations relative to the eruption site augmented by traverses. Four hundred and forty-four scan and traverse measurements met quality criteria and were used, along with plume height and meteorological data, to calculate SO2 fluxes while accounting for wind-related uncertainties.A tendency for stronger SO2 flux concurrent with higher amplitude seismic tremor and the occurrence of lava fountaining was observed during Phases 4 and 5 which were characterized by intermittent crater activity including observable effusion of lava and gas release interspersed with long repose times. This tendency was used to refine the calculation of the amount of SO2 emitted during variably vigorous activity. The continuous seismic tremor time series was used to quantify how long during these eruption phases strong/weak activity was exhibited to improve the calculated SO2 flux during these Phases.The total SO2 emissions derived from field measurements align closely with results obtained by combining melt inclusion and groundmass glass analyses with lava effusion rate measurements (910 ± 230 kt SO2). Specifically, utilizing the maximum S content found in evolved melt inclusions and the least remaining S content in accompanying quenched groundmasses provides an identical result between field measurements and the petrological calculations. This suggests that the maximum SO2 release calculated from petrological estimates should be preferentially used to initialize gas dispersion models for basaltic eruptions when other measurements are lacking.During the eruption, the CALPUFF dispersion model was used to forecast ground-level exposure to SO2. The SO2 emission rates measured by DOAS were used as input for the dispersion model, with updates made when a significant change was measured. A detailed analysis of one mid-distance station over the entire eruption shows that the model performed very well at predicting the presence of volcanic SO2 when it was measured. However, it frequently predicted the presence of SO2 that was not measured and the concentrations forecasted had no correlation with the concentrations measured. Various approaches to improve the model forecast were tested, including updating plume height and SO2 flux source terms based on measurements. These approaches did not unambiguously improve the model performance but suggest that improvements might be achieved in more-polluted conditions.

Research on leakage detection technology of natural gas pipeline based on modified Gaussian plume model and Markov chain Monte Carlo method

Natural gas pipeline leakage can cause significant economic losses and even jeopardize public safety. Therefore, leakage detection technology is crucial. In this study, the pipeline source term estimation method is proposed to identify the source of natural gas pipeline leakage. The method combines the modified Gaussian plume model, the unmanned aerial vehicle inspection technology equipped with laser methane sensors, and the Markov chain Monte Carlo method based on Bayesian inference. Different from traditional leak detection techniques, this method can simultaneously obtain the location and rate of pipeline leakage. The accuracy of the method is validated by using simulated data from the modified Gaussian plume model, which can be employed to replace experimental tests. Convergence diagnosis and model evaluation are performed on the inversion model in this method, verifying its rationality and feasibility. The impact of system errors on the pipeline source estimation method is analyzed, revealing that the accuracy of inversion can be improved by using prior information and ensuring the reliability of the forward gas dispersion model and monitoring data. The pipeline source term estimation method proposed in this paper can offer technical support for the emergency treatment of natural gas pipeline leakage accidents.

Exploration and Gas Source Localization in Advection-Diffusion Processes with Potential-Field-Controlled Robotic Swarms.

Mobile multi-robot systems are well suited for gas leak localization in challenging environments. They offer inherent advantages such as redundancy, scalability, and resilience to hazardous environments, all while enabling autonomous operation, which is key to efficient swarm exploration. To efficiently localize gas sources using concentration measurements, robots need to seek out informative sampling locations. For this, domain knowledge needs to be incorporated into their exploration strategy. We achieve this by means of partial differential equations incorporated into a probabilistic gas dispersion model that is used to generate a spatial uncertainty map of process parameters. Previously, we presented a potential-field-control approach for navigation based on this map. We build upon this work by considering a more realistic gas dispersion model, now taking into account the mechanism of advection, and dynamics of the gas concentration field. The proposed extension is evaluated through extensive simulations. We find that introducing fluctuations in the wind direction makes source localization a fundamentally harder problem to solve. Nevertheless, the proposed approach can recover the gas source distribution and compete with a systematic sampling strategy. The estimator we present in this work is able to robustly recover source candidates within only a few seconds. Larger swarms are able to reduce total uncertainty faster. Our findings emphasize the applicability and robustness of robotic swarm exploration in dynamic and challenging environments for tasks such as gas source localization.

Accelerated convergence for city-scale flow fields using immersed boundaries and coupled multigrid

Fast and accurate prediction of Chemical, Biological and Radiological (CBR) dispersion patterns and toxic extents is important for public safety as conventional and non-conventional airborne threats released into the built environment can pose serious dangers to human health ranging from difficulty breathing and eye irritation to incapacitation and death. Computational fluid dynamics (CFD) is routinely used for gas dispersion modeling when increased accuracy over traditional Gaussian-based methods is required, although CFD requires greater modelling efforts. The combination of structured Cartesian meshes and immersed boundary methods reduces this effort and has been resurging in popularity over the past decade.Due to a lack of related literature, a simple and robust methodology was developed by the authors to treat immersed boundaries at multigrid levels, and this integration focused on a segregated multigrid approach using structured Cartesian meshes with adaptive mesh refinement. In the current work, a coupled multigrid treatment is adapted to the integrated immersed boundary methodology and adaptive mesh approach and compared to the segregated technique in terms of convergence rates. To accelerate convergence, the coupled approach selectively applies grid refinement at coarse grid levels, and by design, neither methodology detects the immersed boundaries on coarse grid levels.The two multigrid approaches are compared at three flow scales: wind tunnel, mock-urban and full-urban. At the wind tunnel scale, the coupled multigrid solver shows 39.3 times speed up over single grid and 10.4 times improvement over the baseline segregated approach. At the full urban scale, the coupled approach is shown to converge wind fields with a 6.3 times speed improvement over the previous segregated approach and therefore is suitable for CFD-based CBR assessment tools.

Real-time hydrogen release and dispersion modelling of hydrogen refuelling station by using deep learning probability approach

Hydrogen release and dispersion from hydrogen refuelling stations have the potential to cause explosion disaster and bring significant causalities and economic losses to the surroundings. Real-time spatial hydrogen plume concentration prediction is essential for the quick emergency response planning to dissipate such flammable vapor cloud and prevent explosion disaster. Deep learning approaches have recently been applied to real-time gas release and dispersion modeling, however, are ‘over-confident’ for spatial plume concentration and boundary estimation, which could not support the robust decision-makings. This study proposes a hybrid deep probability learning-based spatial hydrogen plume concentration prediction model, namely DPL_H2Plume by integrating deep learning and Variational Bayesian Inference. Numerical model of hydrogen release and dispersion from hydrogen refuelling station is built to construct the benchmark dataset. By using such dataset, two pre-defined parameters, namely Monte Carlo sampling number m = 300 and dropout probability p = 0.1 are determined to ensure the model's tradeoff between inference accuracy and efficiency. Comparison between our proposed model and the state-of-the-art model is also conducted. The results demonstrate that our model exhibits a competitive accuracy of R2 = 0.97 as well as an inference time 3.32 s. In addition, our model gives the comprehensive estimations including not only spatial hydrogen plume concentration but also its uncertainty. Also, our model provides the more accurate estimation at plume boundary compared to the state-of-the-art model. Overall, our proposed model could provide reliable alternative for constructing a digital twin for emergency management of hydrogen refuelling station.

Quantification of CO2 degassing and atmospheric dispersion at Caldeiras da Ribeira Grande (São Miguel Island, Azores)

Caldeiras da Ribeira Grande is one of the degassing areas of Fogo, a trachytic central volcano located at São Miguel Island (Azores archipelago). Recently, new steam emissions, soil CO2 and temperature anomalies developed towards the inhabited area, causing high indoor CO2 values and affecting the vegetation and several small animals that were found dead in depressions and low-ventilated zones. During July–August 2021, a soil CO2 flux survey was carried out on the north flank of the volcano, estimating a soil gas release of at least 40 t d−1 (excluding the contribution of the fumaroles) over an area of ∼0.27 km2. Two populations for the CO2 released were found, highlighting the biogenic and volcanic-hydrothermal origins. General NW-SE diffuse degassing structures (DDS) were identified, in agreement with the tectonic lineaments previously recognized in the area. In this regard, we investigated the passive gas dispersion in the atmosphere at Caldeiras da Ribeira Grande performing a model validation aimed to estimate the fumarolic gas flux at source and the potential hazard for human and animal lives posed by CO2. Numerical simulations were carried out with the DISGAS-2.3, a 3D Eulerian advection-diffusion model, and the relative outputs processed through the VIGIL-1.3 workflow able to provide probabilistic long-term CO2 concentration maps, considering a meteorological variability over the last 30 years (1991–2020) taken from the ECMWF ERA5 reanalysis dataset. A best-fit between observed and simulated CO2 concentrations allowed us to estimate the total gas flux of the area (∼209 t d−1) obtained by scaling the soil CO2 gas flux by a factor 30. Such an estimate is composed of ∼174 t d−1 as unknown fumarolic and ∼ 35 td−1 as diffuse contribution, in a good agreement with measurements.Although the present-day CO2 concentration at 0.3 m height cannot be considered to raise serious concerns for human health, we reasonably infer that the death of small animals may be due to local conditions of CO2 accumulation or to the presence of H2S. The current study highlights the relevance of coupling gas flux maps, concentration data, and gas dispersion modeling to obtain robust estimation of gas fluxes, including the fumarolic contribution, and identify zones potentially impacted by dangerous concentrations of volcanic gases, which are relevant for land-use planning and hazard assessment in case of renewed escalations of volcanic activity.

BI-IEnKF coupling model for effective source term estimation of natural gas leakage in urban utility tunnels

As an effective way to facilitate the increasing demand for reliable infrastructure, energy supply and sustainable urban development, underground utility tunnels have been developed rapidly in recent years. Due to the widespread distribution of utility tunnels, the safe operation of natural gas pipelines accommodated in utility tunnels has caused great concern considering fire, explosion, and other coupling consequences induced by the gas pipeline leakage. However, the limited information on leakage source terms in accidental leakage scenarios could preclude timely consequence assessment and effective emergency response. In this study, a BI-IEnKF coupling source term estimation (STE) model is developed, with the combination of gas dispersion model, Bayesian inference (BI) and iterative ensemble Kalman filter (IEnKF) method, to achieve the effective source term estimation (including leakage location and leakage rate) and gas concentration distribution prediction. The newly developed model is first evaluated by the twin experiment with good reliability and accuracy. Furthermore, three contributing factors affecting the performance of the developed BI-IEnKF coupling STE model were investigated to assist parameter selection for practical use. Additionally, the novel application of mobile sensors serving as an alternative for fixed sensors is explored, and an application framework is sequentially given to guide the deployment of the developed coupling model in utility tunnels. The results show that the developed model has great performance in accuracy, efficiency and robustness, as well as the potential to be applied in actual utility tunnel scenarios. This study can provide technical supports for safety control and emergency response in the case of natural gas pipeline leakage accidents in utility tunnels. Also, it could be helpful to reasonable references for gas lekage monitoring system design.

Stationary and drone-assisted methane plume localization with dispersion spectroscopy

This work presents stationary and mobile retroreflector-based remote sensing techniques for methane leak localization and quantification using chirped laser dispersion spectroscopy equipped with a custom laser transceiver capable of continuous tracking of a flying drone and coupled with inverse atmospheric gas dispersion modeling. The techniques demonstrate the ability to localize leaks as low as 0.13 g CH4·s−1, which are up to 25 times smaller than those typically observed at natural gas facilities, as well as actively track a moving retroreflector mounted on a lightweight (∼250 g) drone to enable spatial plume reconstruction. This system exhibited a 2.3 ppm-m sensitivity over pathlengths of 40–150 m. Source localization to within ±7 m is demonstrated using a modified horizontal radial plume mapping technique with a stationary retroreflector grid. Meanwhile, the mobile system utilizing a drone-mounted retroreflector is able to localize a controlled release within ±1 m of its source location and estimate leak rates using inversion techniques assuming type B Gaussian plume stability class within ±30% error with respect to the actual low flow rate releases.

A Modified k-ε Turbulence Model for Heavy Gas Dispersion in Built-Up Environment

For hazard assessment purposes, the dispersion of gases in complex urban areas is often a scenario to be considered. However, predicting the dispersion of heavy gases is still a challenge. In Germany, the VDI Guideline 3783, Part 1 and 2 is widely used for gas dispersion modelling. Whilst Part 1 uses a gauss model for calculating the dispersion of light or neutrally buoyant gases, Part 2 uses wind tunnel experiments to evaluate the heavier-than-air gas dispersion in generic built up areas. In practice, with this guideline, it is often not possible to adequately represent the existing obstacle configuration. To overcome this limitation, computational fluid dynamics (CFD) methods could be used. Whilst CFD models can represent obstacles in the dispersion area correctly, actual publications show that there is still further research needed to simulate the atmospheric flow and the heavy gas dispersion. This paper presents a modified k-ε-turbulence model that was developed in OpenFOAM v5.0 (England, London, The OpenFOAM Foundation Ltd Incorporated) to enhance the simulation of the atmospheric wind field and the heavy gas dispersion in built-up areas. Wind tunnel measurements for the dispersion of neutrally buoyant and heavy gases in built-up environments were used to evaluate the model. As a result, requirements for the simulation of the gas dispersion under atmospheric conditions have been identified and the model showed an overall good performance in predicting the experimental values.

Estimation of the influence of protective behaviour on expected injuries in transport accidents with hazardous materials

The release of hazardous materials due to transport accidents can have major health consequences for bystanders. The number of casualties will partly be determined by their protective behaviour.This study describes a three-step approach to predict health consequences of protective behaviour (for example, hiding or escaping) of bystanders in the first minutes of a transport accident with hazardous material. First, a discrete choice experiment (DCE) was used to predict protective behaviour. Second, a gas dispersion model (SeReMo) was used to estimate the distribution of casualties for different protective behaviours. Third, results of the DCE and SeReMo were combined to estimate the distribution of casualties in the population-at-risk. This approach was applied to a hypothetical marine accident with different hazardous material scenarios on a large waterway in a close vicinity of a beach/quay with bystanders.An important finding of our study was that in general a short reaction time and escaping in cross-wind direction, as protective behaviours, are of vital importance to reduce the number of casualties. A scenario with a short reaction time and a visible cloud towards the beach/quay resulted in a protective behaviour with the largest reduction of casualties.A dynamic risk assessment approach considering that people threatened by hazardous material are not ‘stationary observers’, but will exhibit protective behaviour, and a risk assessment that takes into account empirical information on expected protective behaviour will present a more realistic estimate of the number and severity of casualties when a large transport accident with hazardous materials would occur.

Determination of methane emissions from biogas plants, using different quantification methods

Reliable and comparable quantification methods are needed for assessing the effectiveness of the biogas production and utilisation process in mitigating methane (CH4) emissions as well as improving the database for emission inventories. The objective of this study was to compare and validate CH4 emissions quantified at two agricultural biogas plants, for up to three days, using diverse on-site (two teams) and off-site methods (three teams), including differential absorption lidar (DIAL), tracer gas dispersion (TDM) and inverse dispersion modelling (IDM). For plant 1, with a constant combined heat and power (CHP) load, the average emission factor varied from 0.3% CH4 (on-site approaches) to 1.2% CH4 (off-site approaches). On-site approaches underestimated overall emissions due to many small (unquantified) CH4 leakages. All methods observed comparable average emission factors for plant 2, ranging between 1.9 and 2.2% CH4. In this case, the majority of emissions emanated from just a few sources. However, correcting the significant influence of the varying CHP load during the measurement campaign revealed significant differences between TDM and IDM (DIAL did not participate). It was demonstrated that TDM and IDM could recover the emission rate from a known point source (controlled release of CH4 via a small diffuser) within an accuracy of 93 ± 15% (TDM) and 92 ± 17% (IDM) under favourable and similar conditions.

Locating and Quantifying Methane Emissions by Inverse Analysis of Path-Integrated Concentration Data Using a Markov-Chain Monte Carlo Approach.

The action to reduce anthropogenic greenhouse gas emissions is severely constrained by the difficulty of locating sources and quantifying their emission rates. Methane emissions by the energy sector are of particular concern. We report results achieved with a new area monitoring approach using laser dispersion spectroscopy to measure path-averaged concentrations along multiple beams. The method is generally applicable to greenhouse gases, but this work is focused on methane. Nineteen calibrated methane releases in four distinct configurations, including three separate blind trials, were made within a flat test area of 175 m by 175 m. Using a Gaussian plume gas dispersion model, driven by wind velocity data, we calculate the data anticipated for hundreds of automatically proposed candidate source configurations. The Markov-chain Monte Carlo analysis finds source locations and emission rates whose calculated path-averaged concentrations are consistent with those measured and associated uncertainties. This approach found the correct number of sources and located them to be within <9 m in more than 75% of the cases. The relative accuracy of the mass emission rate results was highly correlated to the localization accuracy and better than 30% in 70% of the cases. The discrepancies for mass emission rates were <2 kg/h for 95% of the cases.

Numerical modelling of wind-influenced above sea gas dispersion and explosion risk analysis due to subsea gas release on multileveled offshore platform

Above-sea gas dispersion due to subsea release could lead to explosion on nearby offshore facility. Less congested single-deck offshore platform has been considered in past for numerical modelling of above-sea gas dispersion and deflagration. Offshore platforms are designed compactly and comprise of multiple decks due to space restrictions. The approaching flammable gas may disperse differently at each deck, and high overpressures could develop in event of explosion in confined decks causing severe damages. Therefore, investigating effect of above-sea gas release and dispersion and resulting vapour cloud explosion on equipment damage and human fatality in complex multi-deck platform is worth investigating. This paper presents above-sea gas dispersion and explosion modelling on realistic multileveled fixed offshore platform under varying wind speeds using FLACS. Simplified steps are provided to model gas release from large area sea surface gas pool along with atmospheric dispersion and subsequent explosion in FLACS. It was found that generally, lower decks had near-stoichiometry concentrations thus being most susceptible to explosion consequences. High wind speed increased flammable cloud accumulation, where 50-60% of decks were filled at 3m/s wind speed whereas 100% filled under 7m/s. It is suggested to have adequate ventilation on lower platform decks specially under high wind circumstances to avoid possible explosion hazard due to cloud accumulation. Congested and confined regions with low flammable concentration produced more overpressure than the areas with same concentration but no or low congestion and confinement. Stoichiometric concentration together with congestion produced incredibly high localised overpressure reaching 3.55-4 barg. The probability of equipment damage and human fatality was reaching 1 for explosions at base and deck 2 under 7 m/s wind speed. Since roof was open, the probability of equipment damage and human fatality was least.

Consequence assessment considerations for toxic natural gas dispersion modeling

The numerical simulation of gas dispersion and estimation of consequence impact is of importance in Oil and Gas industry's process safety management. For natural gas fields with toxic components like Hydrogen sulfide, the toxicity impact zone drives business decisions related to equipment design, facility siting, layout, land use planning, and emergency response measures. Proprietary tools or empirical models which are calibrated using experiment database are often used for carrying out consequence modeling.The selection of a software tool and a suitable dispersion model, based on the cloud behavior, at the source of dispersion is critical for the impact zone estimation. It is observed that, the fluid phase and the cloud density are key for determining the appropriate dispersion model. Incorrect parameter selection could lead to an inaccurate consequence impact zone estimation. This in turn could result in disproportionate process risk management efforts especially for toxic impacts from exposure to a very low concentration.This paper discusses the results from consequence modeling studies done for a selected set of toxic natural gas release events related to onshore pipeline transfer using approved software. The study analyses the modeling inputs, parameters and determines the key release source terms and atmospheric parameters that impacts the estimation of impact zones. The study determined that the natural gas dispersion behaviour is dependent on the natural gas molar mass and the composition of Hydrogen sulfide. The study provides the guidance on overcoming uncertainty in dispersion modeling through sensitivity assessments and lists key parameters to be subjected for toxic natural gas dispersion modeling sensitivity analysis.

VIGIL: A Python tool for automatized probabilistic VolcanIc Gas dIspersion modeLling

Probabilistic volcanic hazard assessment is a standard methodology based on running a deterministic hazard quantification tool multiple times to explore the full range of uncertainty in the input parameters and boundary conditions, in order to probabilistically quantify the variability of outputs accounting for such uncertainties. Nowadays, different volcanic hazards are quantified by means of this approach. Among these, volcanic gas emission is particularly relevant given the threat posed to human health if concentrations and exposure times exceed certain thresholds. There are different types of gas emissions but two main scenarios can be recognized: hot buoyant gas emissions from fumaroles and the ground and dense gas emissions feeding density currents that can occur, e.g., in limnic eruptions. Simulation tools are available to model the evolution of critical gas concentrations over an area of interest. Moreover, in order to perform probabilistic hazard assessments of volcanic gases, simulations should account for the natural variability associated to aspects such as seasonal and daily wind conditions, localized or diffuse source locations, and gas fluxes. Here we present VIGIL (automatized probabilistic VolcanIc Gas dIspersion modeLling), a new Python tool designed for managing the entire simulation workflow involved in single and probabilistic applications of gas dispersion modelling. VIGIL is able to manage the whole process from meteorological data processing, needed to run gas dispersion in both the dilute and dense gas flow scenarios, to the post processing of models’ outputs. Two application examples are presented to show some of the modelling capabilities offered by VIGIL.

Risk mapping of Chlorine gas dispersion from a storage plant in a dense urban area of Lahore

In recent times, with the increasing population in urban areas, the adverse events happening due to explosion and dispersion of toxic chemicals has increased. Events like toxic gas dispersion can cause severe environmental issues that endanger human safety and health. Consequence hazard modeling and vulnerable population assessment are critical to predict and minimize the losses. Chlorine is not only utilized in water treatment plants as a disinfectant, but it ca n also lead to some serious concerns to human health. In this paper, the modeling of chlorine release has been investigated for chlorine storage plant by using Heavy gas dispersion model. The modeling results showed that the cloud of chlorine is about 1.9 miles across just downwind of the release for accidental release of chlorine in the summer; however, the probability of fatality is 100 % in a whole year in a distance of 0.36 miles from the Storage Place. As a significant result, the land around the chlorination unit covering a range of approximately 1.9 miles is vulnerable in all wind directions and in the case of South-West direction of the wind, vulnerable population is highly dense, risk prevention in that region should be accounted for. Affected Population and areas at risk are calculated, which illustrates the toxically impacted areas and the population in need of immediate help and evacuation. Such studies can serve as a useful tool for decision-makers to prepare an emergency plan in case of accidental releases.

Probabilistic real-time deep-water natural gas hydrate dispersion modeling by using a novel hybrid deep learning approach

Computational Fluid Dynamic (CFD) has been widely used for the gas release and dispersion modeling, which however could not support real-time emergency response planning due to its high computation overhead. Surrogate models offer a potential alternative to rigorous computational approaches, however, as the point-estimation alternatives, the existing neural network-based surrogate models are not able to quantify the uncertainty of the released gas spatial concentration. This study aims to fill a gap by proposing an advanced hybrid probabilistic Convolutional-Variational Autoencoder-Variational Bayesian neural network (Conv-VAE-VBnn). Experimental study based on a benchmark simulation dataset was conducted. The results demonstrated the additional uncertainty information estimated by the proposed model contributes to reducing the harmful effect of too ‘confidence’ of the point-estimation models. In addition, the proposed model exhibits competitive accuracy with R2 = 0.94 compared and real-time capacity with inference time less than 1s. Latent size Nz = 2, noise σz=0.1 and Monte Carlo sampling number m = 500 to ensure the model’s real-time capacity, were also given. Overall, our proposed model could provide a reliable alternative for constructing a digital twin for emergency management during the exploration and exploitation of marine natural gas hydrate (NHG) in the near future.

Improved quantitative risk model for integrity management of liquefied petroleum gas storage tanks: Mathematical basis, and case study

AbstractChemical, petrochemical, and refinery sectors have been facing tougher safety, environmental and mechanical integrity regulations as well as challenges associated with the need for cost reduction to improve competitiveness. Risk‐based Inspection (RBI) is a cost‐effective approach to manage operational risks by making an informed decision on inspection frequency, extent of inspections, and types of non‐destructive testing. This paper presents a comprehensive practical risk model that can simulate all components of liquefied petroleum gas (LPG) tanks, that is, roof, shell, nozzles, and bottom plates. The probability of failure (PoF) model was developed based on failure frequency data from four different sources to determine the adequate general failure frequencies, moreover, it addresses the fatigue effect due to cyclic loading. Reviewing past losses of containment incidents revealed that fires and explosion are the most likely outcomes, therefore, this risk model introduced area affected consequence of failure (CoF) simulation. The CoF model addresses all the main components of the LPG tank and deploys a practical gas dispersion model to estimate the vapor cloud explosion (VCE) and flash fire consequence affected areas. Furthermore, modified mass release equations are presented to account for the decline in the release rate with time to determine the pool fire area and leak duration. Consequence areas are determined based on serious personnel injuries and component damage from thermal radiation and explosions. To validate the risk model, a proof of concept assessment was performed on seven tanks.