Abstract
The process industries rely on various software systems and use a wide range of technologies. Predictive modeling techniques are often applied to data obtained from these systems to build the predictive functions used to optimize the production processes. Therefore, there is a need to provide a proper representation of knowledge and data and to improve the communication between the data scientists who develop the predictive functions and domain experts who possess the expert knowledge of the domain. This can be achieved by developing a semantic model that focuses on cross-sectorial aspects rather than concepts for specific industries, and that specifies the meta-classes for the formal description of these specific concepts. This model should cover the most important areas including modeling the production processes, data analysis methods, and evaluation using the performance indicators. In this paper, our primary objective was to introduce the specifications of the Cross-sectorial domain model and to present a set of tools that support data analysts and domain experts in the creation of process models and predictive functions. The model and the tools were used to design a knowledge base that could support the development of predictive functions in the green anode production in the aluminum production domain.
Funder
Horizon 2020 Framework Programme
Subject
Process Chemistry and Technology,Chemical Engineering (miscellaneous),Bioengineering
Cited by
4 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Cognitive Architecture for Process industries;Proceedings of the 3rd Eclipse Security, AI, Architecture and Modelling Conference on Cloud to Edge Continuum;2023-10-17
2. Developing a Comprehensive Mathematical Model for Aluminium Production in a Soderberg Electrolyser;Energies;2023-08-30
3. A Graph-Based Approach for Representing Water Treatment Process Knowledge;2023 14th International Conference on Information, Intelligence, Systems & Applications (IISA);2023-07-10
4. Knowledge-Based Approaches to Intelligent Data Analysis;Towards Digital Intelligence Society;2020-12-22