The complex behaviour of large discrete event systems makes such systems difficult to diagnose. Using decentralised techniques helps limit combinatorial explosion but is not sufficient. Often, the complexity of the diagnosis is dependent on how components in the system are connected and the number of connections between them. We propose to augment a decentralised junction tree-based approach by ignoring some connections on the system. This helps reduce the complexity, and hence the cost, of the diagnostic reasoning required. However accuracy of the diagnosis is also reduced.

Two key impediments for the commercial success of model-based diagnosis (MBD) include (a) a lack of component libraries for more cheaply and efficiently building MBD models, and (b) a failure to integrate the development of embedded MBD code within the design process. This article addresses both of these impediments by providing a formal framework that integrates component-based design with MBD modeling. The proposed framework extends the consistency-based theory of MBD with a component-based design theory based on contracts. The contributions of the article are as follows:

This paper presents a methodology to quantify the uncertainty in fatigue crack growth prognosis, applied to structures with complicated geometry and subjected to variable amplitude multi-axial loading. Finite element analysis is used to address the complicated geometry and calculate the stress intensity factors. Multi-modal stress intensity factors due to multi-axial loading are combined to calculate an equivalent stress intensity factor using a characteristic plane approach. Crack growth under variable amplitude loading is modeled using a modified Paris law that includes retardation effects.

Timed failure propagation graph (TFPG) is a causal model that captures the causal and temporal aspects of failure propagation in a wide variety of engineering systems. In this paper we investigate the problem of failure prognosis within the TFPG model settings. The paper introduces a formal definition for system reliability based on measures of failure criticality, proximity between alarm observations, and plausibility of the estimated current system condition.

This paper extends model-based diagnosis (MBD) (Reiter87,deKleer87) to systems which convert, move and process material. Examples of such systems are printers, refineries and food processing plants. Such plants present two challenges to model-based diagnosis: (1) the plant may process 100s-1000s of items per minute so retaining full details of behavior of all past objects is impractical, and (2) complex multi-way interactions can occur among components operating on the same object. We address the first challenge by synopsizing past behavior in a data structure of fixed size.

Opportunities exist to apply nonlinear filtering to model-based prognostics in order to provide a systematic way of dealing with the propagation of system damage at some future time, whenever imprecise diagnostic information is obtained. Central to the prognostics problem is the ability to properly capture and manage uncertainties when predicting remaining useful life of a particular component of interest. The goal of this paper is to present a foundation for prediction and filtering of the failure process using nonlinear prognostic models and exact (finite-dimensional) filters.

Prognosis and Health Management methodologies require efficient parameter estimation approaches to enable systematic system reconfiguration and adaptive control to accommodate faulty behaviors, and to predict future system states. However, accurate and timely on-line parameter estimation of complex, nonlinear systems is difficult and can be computationally expensive. In this work, we propose a more efficient technique for on-line parameter estimation in TRANSCEND. This new approach is based on previous works on model decomposition and dependency compilation.