Andrey Sadovykh

Mohammad Samadi, Luís Miguel Pinho, Andrey Sadovykh et al.

Renewable Energy Communities (RECs) are emerging as a key organizational model for local and global sharing of renewable generation, storage, and flexible loads. Engineering Digital Twins of RECs is made difficult by the heterogeneity of devices, contracts, and runtime data involved. In this paper, we take a first step toward a Model-Based Systems Engineering (MBSE) workflow for REC's Digital Twins. Starting from an industrially-validated REC domain model, we re-express a representative house subset in SysML using the open-source Modelio tool, yielding two Block Definition Diagrams - a device taxonomy and a community organizational view. We then discuss four semantic gaps that plain SysML leaves open and sketch how the SAREF4ENER ontology could be imported as a reference package to close them. Combining SysML with SAREF-based semantics for smart-energy Digital Twins remains largely unexplored, and we position this paper as a first step along that line.

Jesus Gorronogoitia Cruz, Andrey Sadovykh, Dragos Truscan et al.

In this paper, we overview our experiences of developing large set of open source tools in ECSEL JU European project called MegaM@Rt2 whose main objective is to propose a scalable model-based framework incorporating methods and tools for the continuous development and runtime support of complex software-intensive Cyber-Physical Systems (CPSs). We briefly present the MegaM@Rt2 concepts, discuss our approach for open source, enumerate tools and give an example of a tools selection for a specific industrial context. Our goal is to introduce the reader with open source tools for the model-based engineering of CPSs suitable for diverse industrial applications.

Vladimir Ivanov, Andrey Sadovykh, Alexandr Naumchev et al.

Requirements identification in textual documents or extraction is a tedious and error prone task that many researchers suggest automating. We manually annotated the PURE dataset and thus created a new one containing both requirements and non-requirements. Using this dataset, we fine-tuned the BERT model and compare the results with several baselines such as fastText and ELMo. In order to evaluate the model on semantically more complex documents we compare the PURE dataset results with experiments on Request For Information (RFI) documents. The RFIs often include software requirements, but in a less standardized way. The fine-tuned BERT showed promising results on PURE dataset on the binary sentence classification task. Comparing with previous and recent studies dealing with constrained inputs, our approach demonstrates high performance in terms of precision and recall metrics, while being agnostic to the unstructured textual input.

Andrey Sadovykh, Dragos Truscan, Hugo Bruneliere

In this paper, we report on our 5-year's practical experience of designing, developing and then deploying a Model-based Requirements Engineering (MBRE) approach and language in the context of three different large European collaborative projects providing complex software solutions. Based on data collected both during projects execution and via a survey realized afterwards, we intend to show that such an approach can bring interesting benefits in terms of scalability (e.g. large number of handled requirements), heterogeneity (e.g. partners with different types of RE background), traceability (e.g. from the requirements to the software components), automation (e.g. requirement documentation generation), usefulness or usability. To illustrate our contribution, we exemplify the application of our MBRE approach and language with concrete elements coming from one of these European research projects. We also discuss further the general benefits and current limitations of using this MBRE approach and corresponding language.