DATADVANCE User Conference 2021 | VIRTUAL 

As of today, our current way of life is changing. The world is becoming digital. This general trend is being reflected in the industry, and, as such, the Energy Sector is following the momentum. For the latter, the challenge is certainly higher since this is to be managed in parallel of the Energy Transition and the drastic (and short term) changes in the way to produce / store / convey energy.

The purpose of the talk will be consequently to share some insight on how TechnipFMC is managing its digital transformation. The focus will be done on both digital strategy and solution which are being deployed and developed to change the way Energy systems are designed and operated, keeping in mind that the end objective is to improve performance (and reliability) for our Clients. As part of this strategy, the light will be put on how pSeven Enterprise (and more generally speaking AI-based solution) is being embedded within our digital product to speed-up, automate and optimize the engineering on some of our core business activities.

The talk will finally conclude on some of the new challenges we are facing on in the development of Renewable Energy solution and, in particular, how AI-based solution will help us to make technology possible and cost competitive.

VoltAero is taking electric aircraft to an entirely new level, paving the way for a new era of regional air transportation. Benefitting from 80-plus years of combined pioneering expertise, VoltAero is developing a truly unique general aviation airplane with hybrid-electric propulsion for safe, quiet, efficient and eco-friendly flight. With seating capacities from four to 10 passengers, Cassio aircraft will offer 3.5 hours of flight autonomy – enabling flight distances of up to 1,200 km.

The benefits of moving beyond an iterative manual design process and leveraging automatic design space exploration techniques have been proven by many engineering organizations. By programmatically exploring more design versions earlier in the product development process, organizations increase product performance and quality, decrease time-to-market and capture their engineering IP more reliably. Yet the barriers to leveraging design space exploration in more organizations and product development processes remain significant: The traditional setup and orchestration of an effective software and hardware stack to execute programmatic optimization workflows require significant expertise, time and financial investments.

With a new generation of cloud-native engineering software tools emerging, end-to-end design space exploration workflows can be set up and executed without administering any of the required “backend” tools such as CAD, Simulation or the Optimization software itself. David Heiny will demonstrate such a workflow developed by Datadvance and SimScale in collaboration and showcased with an electronics cooling example. The design of a water cooled Insulated-Gate Bipolar Transistor (IGBT) cold plate is explored and optimized to maximize heat transfer from the IGBT to the coolant fluid at low mass flow rates. The workflow runs 100% in the cloud - from the optimization software over CAD generation to the CFD simulation.

The EU and the Japanese government announced a ban on the new gasoline vehicles by 2035. New battery development will greatly affect the future of EV. The characteristics of Lithium-ion secondary batteries , which is one of the key factor for EV are in a trade-off relationship, and it is need to search optimal design considering the balance. In this session, we will discuss an approach to efficiently handle such complex multi-objective optimization problems and search for optimal solutions.

In this talk we give an example of successful application of pSeven product of Datadvance to a particular problem of optimization of the oil production stimulation technology. Hydraulic fracturing has been used routinely in the field since 1947 to stimulate oil (and, later, shale gas) production. The design of a pumping schedule is usually obtained with the help of commercial simulators based on first principles continuum mechanics, where the model needs validation itself. We took an alternative route based on real field data on reservoir properties, well, fracturing design and production. Digital database now includes more than 5000 wells from 23 oilfields, with 6600+ fracturing operations in total. Starting with ~400 parameters characterizing each well, we end up with about 40 key governing parameters used as input features for forward modeling. An inverse problem (selecting an optimum set of fracturing design parameters to maximize production) is formulated as optimizing a high dimensional black box approximation function constrained by boundaries and solved with surrogate-based optimization from pSeven, which demonstrated better performance than alternatives. A recommendation system containing all of the above methods is designed to advise a production stimulation engineer on an optimized fracturing design.

We will also discuss the outlook for the future, where we need to optimize technologies with respect to the set of ESG metrics: minimize carbon footprint (Scope 1 emissions), environmental impact (reduce the amount of chemicals) and minimize solid material injected underground (Scope 3 emissions of supply chain). Net zero is a data integration and data analytics problem, where access to good quality data and advanced analysis instruments becomes crucial to provide a well-grounded advice to the industry.

All three types of Underground Gas Storages (UGS) – depleted fields, aquifers, and salt caverns - are exploited by Storengy, the European leader in the gas storage industry. To summarize: the gas cycling over the UGS portfolio is periodic, with gas injection over the summer (high gas stock), and gas production over the winter period.

Focusing on aquifer gas storages, some chemical reactions might happen in the underground porous media (depending on variables Xi), which might modify the characteristics of the UGS produced gas and challenge a direct re-injection in the transportation system, where gas must fulfill some technical specifications.

It is therefore crucial to have reliable forecasts of such UGS produced gas characteristics (response Y). Although the physics is well understood, no numerical model is currently available to mimic such phenomena.From a raw dataset, we have capitalized on the data analysis and Model builder features of pSeven to:

• Assess and rank the impact on Y of the different Xi,
• Build some reliable models able to forecast the gas key characteristic Y over some periods of interest.

This pSeven application is complementary to other internally developed approaches and, by cross-checking, reinforces our confidence in the obtained forecasts.

In this project, we want to identify a mathematical method to:

• Create virtual sensors based on machine learning applied to multi-physics systems
• Implement the hereabove virtual sensors in an embed software
• Constrain the surrogate model to ensure compatibility with closed loop control

The chosen method is the Response Surface Model of GTApprox. The main reasons are:

• The mathematical simplicity (quadratic function)
• Small memory and CPU usage
• Consistency with the physics modelled

Nowadays, the performance of numerical tools allows studying in silico the influence of operating parameters of industrial processes. This is true even for complex processes like hot rolling involving heat exchanges with ambient and rolls, and strongly non-linear material behavior of the rolled steel.

In this work, numerical simulation has been specifically utilized to reduce the risk of surface defects during the rolling of a microalloyed grade in the LUNA rolling mill (ABS). In a stand identified as critical, the influence of the roll groove design on the surface damage is particularly investigated. First, the material model is calibrated in ABS Centre Métallurgique thanks to hot compression tests in temperature and strain-rate ranges that are consistent with the rolling conditions. Then, an optimization workflow is implemented in the pSeven® software. In this workflow, the optimizer drives the parameters that define the roll geometry. The geometry update using the CAD software SolidWorks® and the hot rolling simulation using Forge® are automated within the workflow, as well as the postprocessing of computations. The initial temperature field of the billet is simulated using the Digital Twin of the LUNA rolling mill including all the previous rolling passes and interpass times. After 50 runs, an optimal groove design has been determined to limit the risk of surface defects.

Pumping outfit is one of the main components for hydraulic networks and systems. Its efficiency and operational reliability determine the performance of the whole systems. At design stage it’s very important to pay attention to the assessment of the effectiveness of pumping outfit and make sure that its parameters are optimized. For these purposes, enterprises use various CAE systems, which provide different approaches to assess efficiency.

pSeven platform comprises sophisticated optimization algorithms and integration capabilities to build complex workflows and carry out Multiphysics analysis. These capabilities allow using pSeven as a single platform for the problem-oriented optimization, which includes different commercial and ad-hoc CAE systems. The report presents examples of pSeven integration with various CAE systems for both fast and comprehensive optimization of pumping outfit.