Production Enterprise is PDS Group's SaaS solution for the surveillance and optimisation of well-centric, production processes for Oil and Gas and related Industries. It embeds a number of key innovations that allow operators to realise the full potential of digital transformation technologies.
The product has been designed from the ground up to enable operators to manage and optimise their subsurface assets as continuous, semi-autonomous processes. Maximisation of production is achieved by optimising well performance, by detecting and diagnosing anomalies at the earliest possible opportunity, and by ensuring timely and effective intervention. To enable this, processes that can be automated should be automated. Industrial Internet of Things (IIoT) and predictive analytics technologies are key to providing improved insights, but surveillance engineers must be able to combine these improved insights with their hard-won understanding of the historical behaviour of their assigned assets.
Discover Production Enterprise
Production Enterprise is PDS Group's SaaS solution for the surveillance and optimisation of well-centric, production processes for Oil and Gas and related Industries. It embeds a number of key innovations that allow operators to realise the full potential of digital transformation technologies.
The product has been designed from the ground up to enable operators to manage and optimise their subsurface assets as continuous, semi-autonomous processes. Maximisation of production is achieved by optimising well performance, by detecting and diagnosing anomalies at the earliest possible opportunity, and by ensuring timely and effective intervention. To enable this, processes that can be automated should be automated. Industrial Internet of Things (IIoT) and predictive analytics technologies are key to providing improved insights, but surveillance engineers must be able to combine these improved insights with their hard-won understanding of the historical behaviour of their assigned assets.
Production Enterprise Workbench
The optimisation of production from subsurface assets is highly dependent on the knowledge, skills and experience of the specialised engineers engaged in surveillance. Production Enterprise allows them to work closely with their surveillance system to evolve, apply, automate and share analysis tools that incorporate their specialised expertise; to codify and share knowledge with their colleagues, and to drive best practices throughout the organisation.
Production Enterprise includes a core set of well-centric, production engineering workflows and other content to help operators understand subsurface behaviour and pinpoint the root cause of issues related to production changes such as: build-up of flow impairments, e.g. scale, asphaltines or hydrates; or the onset of artificial lift or EOR anomalies in water, steam, and gas injection settings.
This "out-of-the-box" content may be customised and expanded upon by the surveillance engineer using the Production Enterprise Workbench. The Workbench is used to run, customise and create new content. It also serves as an analytical tool, providing easy access to curated data with powerful analytical and visualisation capabilities. Developed content can easily be shared with other users, within asset team and with peers across the organisation, supporting bottom-up learning and top-down dissemination of emergent best practices.
Production Enterprise leverages cutting-edge and complementary innovations in connectivity, instrumentation and machine intelligence. These enable it to handle higher rate, higher volume data streams, combining them with rich sources of contextual information in real time to deliver valuable insights actionable by machines, by humans, and by both working in collaboration.
Digital Well Twin
The Digital Well Twin is a foundational element of Production Enterprise. It is a virtual replica (or model) of the physical well that is able to reflect the asset’s configuration and context, and its historical and current states and conditions. It can be used to generate and test inferences about current and future behaviors.
It is:
Live – working with streaming production data to mirror the operational cadence of the well, and to sense, alert and respond at the earliest opportunity to defer or avoid sub-optimal performance or downtime.
Intelligent – using the wealth of data and its internal model to infer hidden parameters of the well, predict and diagnose behaviors, and identify opportunities for optimisation.
Whereas the models underlying most Digital Twins can draw from a robust scientific and engineering explanation of how their real-world counterpart should behave and why, and from technology that can measure real-world behavior in accordance with modelling needs, this is simply not the case for Digital Twins of subsurface assets such as wells. In most cases, a computational physics-based model of a well is limited in fidelity and utility by our inability to measure all required real world properties with sufficient accuracy and with adequate spatial or temporal resolution. The resulting model can only suggest plausible explanations for observed behaviours, and inferences can only be made subject to these inherent unknowns and uncertainties. Under such conditions, model results and their fidelity must be interpreted in a wider context guided by the experience of the expert surveillance engineer.
The innovation within our Digital Well Twin is the way we tackle this problem. Firstly, our Digital Well Twin embodies a Hybrid Well Model which is a physics-based well model extended with machine learning techniques. This is a new type of well model specifically designed for semi-autonomous use in a surveillance system - for example self-tuning to changing local conditions and supporting automatic recalibration. Secondly, by embedding it within Production Enterprise, we deliver a Digital Well Twin that can evolve under the guidance of the surveillance engineer to provide the best possible support for production surveillance and optimisation.
The Hybrid Well Model is an ongoing research activity being undertaken in collaboration with the Industrially Focused Mathematical Modelling Centre for Doctoral Training at the Mathematical Institute of the University of Oxford (InfoMM CDT). Its aim is to couple innovations in mathematical modelling, computational fluid dynamics and data sciences to create robust models for the Digital Well Twin and ultimately for a Digital Field Twin.
The Digital Well Twin is a foundational element of Production Enterprise. It is a virtual replica (or model) of the physical well that is able to reflect the asset’s configuration and context, and its historical and current states and conditions. It can be used to generate and test inferences about current and future behaviors.
It is:
Live – working with streaming production data to mirror the operational cadence of the well, and to sense, alert and respond at the earliest opportunity to defer or avoid sub-optimal performance or downtime.
Intelligent – using the wealth of data and its internal model to infer hidden parameters of the well, predict and diagnose behaviors, and identify opportunities for optimisation.
Whereas the models underlying most Digital Twins can draw from a robust scientific and engineering explanation of how their real-world counterpart should behave and why, and from technology that can measure real-world behavior in accordance with modelling needs, this is simply not the case for Digital Twins of subsurface assets such as wells. In most cases, a computational physics-based model of a well is limited in fidelity and utility by our inability to measure all required real world properties with sufficient accuracy and with adequate spatial or temporal resolution. The resulting model can only suggest plausible explanations for observed behaviours, and inferences can only be made subject to these inherent unknowns and uncertainties. Under such conditions, model results and their fidelity must be interpreted in a wider context guided by the experience of the expert surveillance engineer.
The innovation within our Digital Well Twin is the way we tackle this problem. Firstly, our Digital Well Twin embodies a Hybrid Well Model which is a physics-based well model extended with machine learning techniques. This is a new type of well model specifically designed for semi-autonomous use in a surveillance system - for example self-tuning to changing local conditions and supporting automatic recalibration. Secondly, by embedding it within Production Enterprise, we deliver a Digital Well Twin that can evolve under the guidance of the surveillance engineer to provide the best possible support for production surveillance and optimisation.
The Hybrid Well Model is an ongoing research activity being undertaken in collaboration with the Industrially Focused Mathematical Modelling Centre for Doctoral Training at the Mathematical Institute of the University of Oxford (InfoMM CDT). Its aim is to couple innovations in mathematical modelling, computational fluid dynamics and data sciences to create robust models for the Digital Well Twin and ultimately for a Digital Field Twin.
We use strictly necessary cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.