---
title: "A Diagnosis Method of Dynamic System Using the Knowledge on System Description"
authors:
  - "Naoyuki Yamada"
  - "Hiroshi Motoda"
doi: ""
url: "https://ijcai.org/proceedings/1983/54"
pdf: "https://ijcai.org/proceedings/1983/0054.pdf"
venue: "Proceedings of the Eighth International Joint Conference on Artificial Intelligence"
year: 1983
pages: "225-228"
track: "Expert Systems"
keywords:
---

# A Diagnosis Method of Dynamic System Using the Knowledge on System Description

Naoyuki Yamada, Hiroshi Motoda

*Proceedings of the Eighth International Joint Conference on Artificial Intelligence*, pages 225-228.

## Abstract

A method is proposed for automatic diagnosis of a dynamic system. It basically uses only the knowledge on system description and does not require any knowledge concerning failure causality. The diagnosis consists of the four steps: expectation value computation, suspects computation, suspects discrimination using observable data and suspects discrimination by test generation. Linear input resolution is used with explicit metalevel control. This method can diagnose in principle all kinds of failures that are logically diagnosable if the system description is appropriate. The capability of the method is demonstrated by an application example of a nuclear reactor feed-water system.