Engineering is evolving in the same way than society is doing. Nowadays, data is acquiring a prominence never imagined. In the past, in the domain of materials, processes and structures, testing machines allowed extract data that served in turn to calibrate state-of-the-art models. Some calibration procedures were even integrated within these testing machines. Thus, once the model had been calibrated, computer simulation takes place. However, data can offer much more than a simple state-of-the-art model calibration, and not only from its simple statistical analysis, but from the modeling and simulation viewpoints. This gives rise to the the family of so-called twins: the virtual, the digital and the hybrid twins. Moreover, as discussed in the present paper, not only data serve to enrich physically-based models. These could allow us to perform a tremendous leap forward, by replacing big-data-based habits by the incipient smart-data paradigm.

Sparse model identification by means of data is especially cumbersome if the sought dynamics live in a high dimensional space. This usually involves the need for large amount of data, unfeasible in such a high dimensional settings. This well-known phenomenon, coined as the curse of dimensionality, is here overcome by means of the use of separate representations. We present a technique based on the same principles of the Proper Generalized Decomposition that enables the identification of complex laws in the low-data limit. We provide examples on the performance of the technique in up to ten dimensions.

In recent times a growing interest has arose on the development of data-driven techniques to avoid the employ of phenomenological constitutive models. While it is true that, in general, data do not fit perfectly to existing models, and present deviations from the most popular ones, we believe that this does not justify (or, at least, not always) to abandon completely all the acquired knowledge on the constitutive characterization of materials. Instead, what we propose here is, by means of machine learning techniques, to develop correction to those popular models so as to minimize the errors in constitutive modeling.

Lithium-ion batteries are widely used in the automobile industry (electric vehicles and hybrid electric vehicles) due to their high energy and power density. However, this raises new safety and reliability challenges which require development of novel sophisticated Battery Management Systems (BMS). A BMS ensures the safe and reliable operation of a battery pack and to realize it a model must be solved. However, current BMSs are not adapted to the specifications of the automotive industry, as they are unable to give accurate results at real-time rates and during a wide operation range. For this reason, the main focus of this work is to develop a Hybrid Twin, as introduced in Chinesta et al. (Arch Comput Methods Eng (in press), 2018. so as to meet the requirements of the new generation of BMS. To achieve this, three reduced order model techniques are applied to the most commonly used physics-based models, each one for a different range of application. First, a POD model is used to greatly reduce the simulation time and the computational effort for the pseudo-2D model, while maintaining its accuracy. In this way, cell design, optimization of parameters, and simulation of battery packs can be done while saving time and computational resources. In addition, its real-time performance has been studied. Next, a regression model is constructed from data by using the sparse-Proper Generalized Decomposition (s-PGD). It is shown that it achieves realtime performance for the whole electric vehicle (EV) system with a battery pack. In addition, this regression model can be used in a BMS without issues because of the simple algebraic expression obtained. A simulation of the EV with the proposed approach is demonstrated using the system simulation tool SimulationX (ESI ITI GmbH. Dresden, Germany). Furthermore, the Digital Twin created using the s-PGD does not only allow for real-time simulations, but it can also adapt its predictions taking into consideration the real driving conditions and the real driving cycle to change the planning in real-time. Finally, a data-driven model based on the employment of Dynamic Mode Decomposition techniques is developed to extract an on-line model that corrects the gap between prediction and measurement, thus constructing the first (to our knowledge) hybrid twin of a Li-ion battery able to self-correct from data. In addition, thanks to this model, the above gap is corrected during the driving process, taking into consideration real-time restrictions.

Solutions of partial differential equations can exhibit multiple time scales. Standard discretization techniques are constrained to capture the finest scale to accurately predict the response of the system. In this paper, we provide an alternative route to circumvent prohibitive meshes arising from the necessity of capturing fine-scale behaviors. The proposed methodology is based on a time-separated representation within the standard Proper Generalized Decomposition, where the time coordinate is transformed into a multi-dimensional time through new separated coordinates, each representing one scale, while continuity is ensured in the scale coupling. For instance, when considering two different time scales, the governing Partial Differential Equation is commuted into a nonlinear system that iterates between the so-called microtime and macrotime, so that the time coordinate can be viewed as a 2D time. The macroscale effects are taken into account by means of a finite element-based macro-discretization, whereas the microscale effects are handled with unidimensional parent spaces that are replicated throughout the time domain. The resulting separated representation allows us a very fine time discretization without impacting the computational efficiency. The proposed formulation is explored and numerically verified on thermal and elastodynamic problems.