The Concept of Support for Laser-Based Additive Manufacturing on the Basis of Artificial Intelligence Methods

Abstract

A general concept of software and information support for laser-based additive manufacturing of metal parts from powder compositions is proposed. It is based on an ontological two-level approach to the formation of knowledge about the processes of laser additive manufacturing. For such an approach, the ontology is clearly separated from the knowledge base. So, domain specialists can create and maintain knowledge without intermediaries in terms and representation that they understand. The conceptual architecture of the decision support software for laser-based additive manufacturing processes is presented. Its information and software components are described. Information components are ontologies, databases of laser-based additive manufacturing system components, databases of materials for additive manufacturing, knowledge base and case database. The knowledge base contains formalized information on the settings of laser-based additive manufacturing modes that ensure compliance of the obtained metal parts with the requirements of the current industry-specific guidelines. The case database contains a structured description of the protocols for using laser technological equipment for additive manufacturing of metal parts from powder compositions. Software components are editors for creating and maintaining data and knowledge bases, decision support system based on both knowledge and cases and tool for cases structuring. There are also external tools for mathematical modelling of directed energy deposition physico-chemical processes. When making decisions, it is proposed to use a hybrid approach that combines knowledge engineering methods and case-based search by analogy. The feature of the approach is the continuous updating of the knowledge base due to its improvement by experts and due to its verification in the process of accumulating cases.