The increasing complexity of Smart Grid (SG) systems demands standardized approaches for developing and deploying applications across diverse grid configurations. This paper systematically analyzes requirements for a generic SG application description language using Grounded Theory Methodology (GTM). Through a novel semi-automated analysis pipeline leveraging Large Language Models (LLMs), several hundred use cases from European and US use case repositories were analyzed. Our findings reveal that data exchange and communication infrastructure is the core category, with seven key requirement areas, including real-time data exchange, control mechanisms, and system architecture. The analysis identified critical mapping requirements between abstract descriptions and concrete implementations, emphasizing the need for flexible yet standardized representation approaches. The results provide a comprehensive foundation for developing a description language that can effectively capture the complexity of SG applications while ensuring universal applicability across different grid configurations.

This paper introduces an innovative approach for the description and mapping of applications in the smart grid domain. By utilizing a skill-based engineering approach, not only flexibility and changeability will increase, but also the need for describing the same application multiple times for different grids decays. As a first use-case, Fault Location, Isolation, and Service Restauration (FLISR) as an example application and a real-world grid in Vietnam have been chosen. By showing how this novel approach could be applied to these two entities, its potential is demonstrated. The IEC 61850-based approach uses a two-step procedure for the mapping. After defining the application's requirements, a mapping algorithm can check if the application can be deployed to a specific grid. In the second step, the actual mapping is described to deploy the application to the devices in the field automatically. Having the requirement description, a recommender system can suggest possible adaptions to the grid in case the check fails.

Exchanging technical documents in the building automation domain is a complicated process. Files are distributed either as drawings, spreadsheets, or text documents. Each stakeholder has to re-enter the data into their own system, and changes are revised manually, often even without revision control. This paper presents a uniform exchange format based on the ’graphical’ standard VDI 3814. To increase acceptance, the industry standards XSD and XML were chosen. As a result of this work, a model is provided that covers the concepts and exchange files provided in the VDI 3814 standard. Given a supporting tool, data can be entered, revised, and exchanged automatically. Based on this unified representation, it is subsequently possible to transfer the data into one of the already existing ontologies in this domain by using model transformations. Some of these ontologies are also referred to in this paper.

Due to the increasing complexity of power grids, the development complexity of smart grid applications has continued to rise in recent years. Smart grid applications are currently being developed for specific power grids and not in a grid-independent way causing significant repetitive engineering effort. This paper gives an overview of the state of the art of application development in the smart grid domain, as well as approaches that try to cope with the mentioned problem. In particular, solutions are examined where the grid-independent application is modeled using a functional description. These approaches are then reviewed for limitations and weaknesses in order to make suggestions on how an integrated framework for the development of such applications could look like. Finally, an overview of research questions to reach this goal is presented.

257