A protocol to formalise function verbs to support conservation-based model checking

Abstract

This paper proposes, demonstrates, and validates a protocol to formalise mechanical function verb definitions to support automated analysis of early design concepts, specifically, physics-based model checking using the conservation laws. Present graph-based function structures rely on controlled vocabularies of verbs and nouns to model functions and flows. Currently, all vocabularies define these terms in plain text, allowing inconsistent interpretation and term usage in models, thus making the models unsuitable for formal computational reasoning. To address this limitation, two types of model-level consistenciestopological and conservational -are identified through model examination as critical requirements for physics-based reasoning. A protocol is then demonstrated to evolve three verbs -Branch, Distribute, and Convert_Energy -from their textual definitions into formal classes that include the necessary information elements to support this reasoning. These verbs are used to construct function models, which are then used to simulate reasoning by computer algorithms, thus validating that the definitions indeed support conservation-based model checks at the conceptual level. Objective and scopeThe overall motivation of this research is to develop a formal representation of mechanical functions in order to develop software tools that allow modelling and analysis of design concepts using automated reasoning that is presently not supported. As a first step in this direction, this paper presents a protocol to derive formal definitions of function verbs from their existing informal definitions that will ensure two basic types of model-level consistencies: (1) topological and (2) conservational. The first ensures that the count and types of input and output flows associated with the instances of a given verb in a function structure are consistent. The second ensures that each function in the model obeys material and energy conservation laws. This research is motivated by the possibility of supporting a concept-level analysis in design. As the critical objective of conceptual design is to synthesise design solutions, much of the *Corresponding author.

Identifying requirements for physics-based reasoning on function structure graphs

“…Advances in this direction have been made in the past years. Specifically, Sen et al (2011a) describes a protocol to derive formal definitions of existing function verbs, Sen et al (2011b) illustrates the potential of automated reasoning using this representation, and Sen (2012) presents a formal representation of functions that partially meets these requirements. Finally, Sen (2012) illustrates a software tool, Concept Modeler, which addresses the requirements of coverage, consistency, validity, domain neutrality, qualitative reasoning, and quantitative reasoning.…”

“…Thus, physics-based objectivity is not realized in this definition. The recent literature proposes two new verbs, energize material and de-energize material, which describe the atomic actions of adding and removing energy or work from material flows (Sen et al, 2011a(Sen et al, , 2013. These verbs could capture the distinction between the two cases.…”

Formalizing Technology Descriptions for Selection During Conceptual Design

“…Evidently, other efforts have been made towards similar goals. Mostly, systems are described by functions/behaviors, leading to the topic of functional decomposition [5][6][7][8][9][10], which is mainly used in design synthesis. This allows automatic selection of components to fulfill certain requirements.…”

Automatically inferring technology compatibility with an ontology and graph rewriting rules

“…The same holds for AirCADia (Guenov et al 2016), which breaks down the system description into a functional and logical domain and proceeds by mapping functions to means to arrive at a system synthesis. Sen, Summers, and Mocko (2011, 2013b, 2013a use function-structure graphs to describe the behaviour of a system enabling physics-based reasoning on it. Although effective, it appears to only be applicable to mechanical and electrical engineering domains, while continuum mechanics seem to pose a problem to this approach.…”