Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

KnowTD-An Actionable Knowledge Representation System for Thermodynamics

Luisa Vollmer, Sophie Fellenz, Fabian Jirasek, Heike Leitte, Hans Hasse

TL;DR

This work demonstrates that actionable thermodynamic knowledge can be encoded in an ontology and accessed by a reasoner to produce explainable, correct solutions for problems involving ideal-gas, closed systems. The KnowTD framework combines a thermodynamics ontology with a problem-oriented reasoning engine, using a graph-based representation to set up equations and traverse solution paths, with results and explanations preserved. A Lego-like mental model, a modular LinkML-implemented ontology, and a sequential, single-equation traversal provide a tractable starting point, validated by 13 case problems and a beta software tool. The approach shows promise for transferring domain knowledge to machines, enabling automation, education, and future expansion to include material databases and real-fluid behavior while maintaining a focus on correctness guarantees.

Abstract

We demonstrate that thermodynamic knowledge acquired by humans can be transferred to computers so that the machine can use it to solve thermodynamic problems and produce explainable solutions with a guarantee of correctness. The actionable knowledge representation system that we have created for this purpose is called KnowTD. It is based on an ontology of thermodynamics that represents knowledge of thermodynamic theory, material properties, and thermodynamic problems. The ontology is coupled with a reasoner that sets up the problem to be solved based on user input, extracts the correct, pertinent equations from the ontology, solves the resulting mathematical problem, and returns the solution to the user, together with an explanation of how it was obtained. KnowTD is presently limited to simple thermodynamic problems, similar to those discussed in an introductory course in Engineering Thermodynamics. This covers the basic theory and working principles of thermodynamics. KnowTD is designed in a modular way and is easily extendable.

KnowTD-An Actionable Knowledge Representation System for Thermodynamics

TL;DR

This work demonstrates that actionable thermodynamic knowledge can be encoded in an ontology and accessed by a reasoner to produce explainable, correct solutions for problems involving ideal-gas, closed systems. The KnowTD framework combines a thermodynamics ontology with a problem-oriented reasoning engine, using a graph-based representation to set up equations and traverse solution paths, with results and explanations preserved. A Lego-like mental model, a modular LinkML-implemented ontology, and a sequential, single-equation traversal provide a tractable starting point, validated by 13 case problems and a beta software tool. The approach shows promise for transferring domain knowledge to machines, enabling automation, education, and future expansion to include material databases and real-fluid behavior while maintaining a focus on correctness guarantees.

Abstract

We demonstrate that thermodynamic knowledge acquired by humans can be transferred to computers so that the machine can use it to solve thermodynamic problems and produce explainable solutions with a guarantee of correctness. The actionable knowledge representation system that we have created for this purpose is called KnowTD. It is based on an ontology of thermodynamics that represents knowledge of thermodynamic theory, material properties, and thermodynamic problems. The ontology is coupled with a reasoner that sets up the problem to be solved based on user input, extracts the correct, pertinent equations from the ontology, solves the resulting mathematical problem, and returns the solution to the user, together with an explanation of how it was obtained. KnowTD is presently limited to simple thermodynamic problems, similar to those discussed in an introductory course in Engineering Thermodynamics. This covers the basic theory and working principles of thermodynamics. KnowTD is designed in a modular way and is easily extendable.
Paper Structure (26 sections, 4 figures)

This paper contains 26 sections, 4 figures.

Table of Contents

  1. Introduction
  2. Knowledge representation
  3. Making knowledge actionable
  4. KnowTD
  5. Overview
  6. A mental model: the Lego®-analogy
  7. Current scope of KnowTD
  8. Ontology
  9. Elements
  10. Basic classifications
  11. Relationships and hierarchy
  12. Representing KnowTD as a knowledge graph
  13. Equations in the ontology
  14. Representation of thermodynamic processes
  15. Implementation of the ontology
  16. ...and 11 more sections

Figures (4)

  • Figure 1: Scheme of the actionable knowledge representation system KnowTD, including the ontology and the reasoner.
  • Figure 2: Subgraph of the current KnowTD knowledge graph showing some important concepts, variables, and attributes and their relationships. Dashed arrows correspond to is_a relationships. The list-like entries correspond to has_a relationships, orange entries in the list refer to a concept. The shown attributes control which of the is_a relationships applies.
  • Figure 3: Process classes of closed systems.
  • Figure 4: User interaction diagram of KnowTD. The system assists the user in translating the given problem to the input of KnowTD. It uses the knowledge of the ontology to create instances of the concepts.