Anika Schumann

Learning user preferences to maximise occupant comfort in office buildings

It is desirable to ensure that the thermal comfort conditions in offices are in line with the preferences of occupants. Controlling their offices correctly therefore requires the correct prediction of their thermal sensation which is often determined using the ISO 7730 norm. The latter defines the predicted mean vote, i.e. the mean thermal preferences of an average group of people, based on a number of variables that are either difficult to measure in practice or require the placement of many sensors in the offices of a building, which is very costly. This paper addresses these issues and predicts the comfort preferences of users solely based on the temperature readings and their previous comfort votes. In order to determine how relevant the latter are to a new state a distance measure is defined that quantifies the similarity between two states. Based on that similarity the previous votes are weighted and the expected comfort vote for the new state is determined. The paper concludes with an experimental analysis using real field study data that show under which climatic conditions our approach outperforms existing approaches.

TOWARDS LOW-COST FAULT DIAGNOSIS IN LARGE COMPONENT-BASED SYSTEMS 1

Abstract We address the problem of fault diagnosis in discrete-event systems. Our contribution is the development of a set of specialised diagnosers whose computation is much more realistic than that of the classical diagnoser. A specialised diagnoser is devoted to the diagnosis of one particular type of fault and is based on the observation of only a subpart of the system.

A Decentralised Symbolic Diagnosis Approach

This paper considers the diagnosis of large discrete-event systems consisting of many components. The problem is to determine, online, all failures and states that explain a given sequence of observations. Several model-based diagnosis approaches deal with this problem but they usually have either poor time performance or result in space explosion. Recent work has shown that both problems can be tackled when encoding diagnosis approaches symbolically by means of binary decision diagrams. This paper further improves upon these results and presents a decentralised symbolic diagnosis method that computes the diagnosis information for each component off-line and then combines them on-line. Experimental results show that our method provides significant improvements over existing approaches.

Computing Optimal Tests for Non-deterministic Systems Using DNNF Graphs

The goal of testing is to distinguish between a number of hypotheses about a system-for example, different diagnoses of faults-by applying input patterns and verifying or falsifying the hypotheses from the observed outputs. Optimal distinguishing tests (ODTs) are those input patterns that are most likely to distinguish between hypotheses about non-deterministic systems. Finding ODTs is practically important, but it amounts in general to determining a ratio of model counts and is therefore computationally very expensive. In this paper, we present a novel approach to this problem, which uses structural properties of the system to limit the complexity of computing ODTs. We first construct a compact graphical representation of the testing problem via compilation into decomposable negation normal form. Based on this compiled representation, we show how one can evaluate distinguishing tests in linear time, which allows us to efficiently determine an ODT. Experimental results from a real-world application show that our method can compute ODTs for instances that were intractable for previous approaches.

Constraint-based optimal testing using DNNF graphs

The goal of testing is to distinguish between a number of hypotheses about a system--for example, different diagnoses of faults-- by applying input patterns and verifying or falsifying the hypotheses from the observed outputs. Optimal distinguishing tests (ODTs) are those input patterns that are most likely to distinguish between hypotheses about non-deterministic systems. Finding ODTs is practically important, but it amounts in general to determining a ratio of model counts and is therefore computationally very expensive. n nIn this paper, we present a novel approach to constraint-based ODT generation, which uses structural properties of the system to limit the complexity of computation.We first construct a compact graphical representation of the testing problem via compilation into decomposable negation normal form. Based on this compiled representation, we show how one can evaluate distinguishing tests in linear time, which allows us to efficiently determine an ODT. Experimental results from a real-world application show that our method can compute ODTs for instances that were intractable for previous approaches.

EFFICIENT ON-LINE FAILURE IDENTIFICATION FOR DISCRETE-EVENT SYSTEMS 1

Abstract The paper addresses the problem of diagnosing complex embedded discrete-event systems. Given a flow of observations from the system, the goal is to explain these observations by identifying possible failures. Approaches that efficiently compute the possible failures, like the classical diagnoser approach, suffer from large space complexity and hence cause difficulties in embedding diagnostic activities. This paper presents a method that dramatically reduces the size of the classical diagnoser and keeps its efficiency.

Efficient on-line failure identification for discrete-event systems

: nThe paper addresses the problem of diagnosing complex embedded discrete-event systems. Given a flow of observations from the system, the goal is to explain these observations by identifying possible failures. Approaches that efficiently compute the possible failures, like the classical diagnoser approach, suffer from large space complexity and hence cause difficulties in embedding diagnostic activities. This paper presents a method that dramatically reduces the size of the classical diagnoser and keeps its efficiency.

Efficient On-Line Failure Identification for Discrete-Event Systems11This research was supported by National ICT Australia (NICTA) in the framework of the SuperCom project (Model-Based Supervision of Composite Systems). NICTA is funded through the Australian Government’s Backing Australia’s Ability initiative, in part through the Australian National Research Council.

: nThe paper addresses the problem of diagnosing complex embedded discrete-event systems. Given a flow of observations from the system, the goal is to explain these observations by identifying possible failures. Approaches that efficiently compute the possible failures, like the classical diagnoser approach, suffer from large space complexity and hence cause difficulties in embedding diagnostic activities. This paper presents a method that dramatically reduces the size of the classical diagnoser and keeps its efficiency.

Efficient On-Line Failure Identification for Discrete-Event Systems1

: nThe paper addresses the problem of diagnosing complex embedded discrete-event systems. Given a flow of observations from the system, the goal is to explain these observations by identifying possible failures. Approaches that efficiently compute the possible failures, like the classical diagnoser approach, suffer from large space complexity and hence cause difficulties in embedding diagnostic activities. This paper presents a method that dramatically reduces the size of the classical diagnoser and keeps its efficiency.