Data-Driven Spatiotemporal Modeling of Reservoir Pressure Dynamics in a Giant Saturated Oil Field in Libya

This paper aims to explore modern techniques based on artificial intelligence (AI) and data science, in order to produce data-driven workflows to analyze, model, and simulate reservoir pressure dynamics. In this paper, it was investigated a data-driven workflow to model reservoir pressure at any point in space and time from sparse pressure data observed at wells, without building a physics-based numerical model. This workflow was termed as spatiotemporal modelling of reservoir pressure. Spatiotemporal modelling of reservoir pressure was based on a three-step workflow including multivariate analysis of pressure data and relevant explanatory variables (features), pressure modelling and spatiotemporal interpolation. The overall workflow provided a comprehensive method to understand and map the reservoir pressure dynamics using data science tools. Several modelling techniques such as generalized additive models, artificial neural networks and spatiotemporal kriging were investigated for their applicability and accuracy. The workflow was applied to a real oil and gas reservoir case, for which the reservoir pressure prediction accuracy was optimized through a few experiments. The optimum experiment produced highly accurate prediction with a mean absolute error of 26.85 psi measured on the training dataset. Moreover, a portion of data used was kept to evaluate blind test accuracy, which amounted to a mean absolute error of 55 psi, for the optimum case. The proposed data-driven workflow was aimed to improve current methods of reservoir engineering and simulation. The suggested workflow showed high accuracy in reservoir pressure predictions with high efficiency in terms of computational resources and time. Additionally, the proposed workflow was developed using open-source libraries which pose no additional cost to computation, in contrast to extremely expensive industry standard physics-based reservoir simulation software. Finally, this workflow could also be used to model other reservoir variables such as production ratios (Water cut, and Gas-Oil Ratio), contacts (Water-Oil contact and Gas-Oil contact), among others.

Spatiotemporal mapping of major trauma in Victoria, Australia

Background: Spatiotemporal modelling techniques allow one to predict injury across time and space. However, such methods have been underutilised in injury studies. This study demonstrates the use of statistical spatiotemporal modelling in identifying areas of significantly high injury risk, and areas witnessing significantly increasing risk over time. Methods: We performed a retrospective review of hospitalised major trauma patients from the Victorian State Trauma Registry, Australia, between 2007 and 2019. Geographical locations of injury events were mapped to the 79 local government areas (LGAs) in the state. We employed Bayesian spatiotemporal models to quantify spatial and temporal patterns, and analysed the results across a range of geographical remoteness and socioeconomic levels. Results: There were 31,317 major trauma patients included. For major trauma overall, we observed substantial spatial variation in injury incidence and a significant 2.1% increase in injury incidence per year. Area-specific risk of injury by motor vehicle collision was higher in regional areas relative to metropolitan areas, while risk of injury by low fall was higher in metropolitan areas. Significant temporal increases were observed in injury by low fall, and the greatest increases were observed in the most disadvantaged LGAs. Conclusions: These findings can be used to inform injury prevention initiatives, which could be designed to target areas with relatively high injury risk and with significantly increasing injury risk over time. Our finding that the greatest year-on-year increases in injury incidence were observed in the most disadvantaged areas highlights the need for a greater emphasis on reducing inequities in injury.

Modelling of a Stochastic Spatiotemporal Variable in Transport Domain

In this article, building on our previous work, we engage in spatiotemporal modelling of transport demand in the Montreal metropolitan area over the period of six years. We employ classical machine learning and regression models, which predict bike-sharing demand in the form of daily cumulative sums of bike trips for each considered docking station. Hourly estimates of demand are then determined by considering the statistical distribution of demand across individual hours of an average day. In order to capture seasonal and other regular variation of demand, longer-term distribution characteristics of bike trips, such as their average number falling on each day of the week, month of the year, etc., were also used as input attributes. We initially conjectured that weather would be an important source of irregular variation in bike-sharing demand, and subsequently included several available meteorological variables in our models. We validated our models by Hold-Out and 10-Fold Cross-Validation, with encouraging results.

Evaluating Spatial Interdependencies of Sector Coupling Using Spatiotemporal Modelling

In light of global warming and the energy turn, sector coupling has gained increasing interest in recent years, from both the scientific community and politics. In the following article it is hypothesized that efficient multifaceted sector coupling solutions depend on detailed spatial and temporal characteristics of energy demand and supply. Hence, spatiotemporal modelling is used as a methodology of integrated spatial and energy planning, in order to determine favourable sector coupling strategies at the local level. A case study evaluation was carried out for both central and decentral renewable energy sources. Considering the high temporal resolutions of energy demand and supply, the results revealed a feasible operation of a district heating network in the central areas of the case study municipalities. Additionally, building integrated solar energy technologies are capable of providing large amount of excess energy that could serve other demand sectors, such as the mobility sector, or could be used for Power-to-X solutions. It is suggested that sector coupling strategies require spatial considerations and high temporal comparisons, in order to be reasonably integrated in spatial and urban planning.