Marina Zanella

Diagnosis of large active systems

Abstract This paper presents a modular technique, amenable to parallel implementation, for the diagnosis of large-scale, distributed, asynchronous event-driven (namely, active) systems. An active system is an abstraction of a physical system that can be modeled as a network of communicating automata. Due to the distributed nature of the class of systems considered, and unlike other approaches based on synchronous composition of automata, exchanged events are buffered within communication links and dealt with asynchronously. The main goal of the diagnostic technique is the reconstruction of the behavior of the active system starting from a set of observable events. The diagnostic process involves three steps: interpretation, merging, and diagnosis generation. Interpretation generates a representation of the behavior of a part of the active system based on observable events. Merging combines the result of several interpretations into a new, broader interpretation. The eventual diagnostic information is generated on the basis of fault events possibly incorporated within the reconstructed behavior. In contrast with other approaches, the proposed technique does not require the generation of the, possibly huge, model of the entire system, typically, in order to yield a global diagnoser, but rather, it allows a modular and parallel exploitation of the reconstruction process. This property, to a large extent, makes effective the diagnosis of real active systems, for which the reconstruction of the global behavior is often unnecessary, if not impossible.

Diagnosis of a class of distributed discrete-event systems

Discrete-event modeling can be applied to a large variety of physical systems, in order to support different tasks, including fault detection, monitoring, and diagnosis. The paper focuses on the model-based diagnosis of a class of distributed discrete-event systems, called active systems. An active system, which is designed to react to possibly harmful external events, is modeled as a network of communicating automata, where each automaton describes the behavior of a system component. Unlike other approaches based on the synchronous composition of automata and on the off-line creation of the model of the entire system, the proposed diagnostic technique deals with asynchronous events and does not need any global diagnoser to be built. Instead, the current approach features a problem-decomposition/solution-composition nature whose core is the online progressive reconstruction of the behavior of the active system, guided by the available observations. This incremental technique makes effective the diagnosis of large-scale active systems, for which the one-shot generation of the global model is almost invariably impossible in practice. The diagnostic method encompasses three steps: (1) reconstruction planning; (2) behavior reconstruction; and (3) diagnosis generation. Step 1 draws a hierarchical decomposition of the behavior reconstruction problem. Reconstruction is made in Step 2, where an intensional representation of all the dynamic behaviors which are consistent with the available system observation is produced. Diagnosis is eventually generated in Step 3, based on the faulty evolutions incorporated within the reconstructed behaviors. The modular approach is formally defined, with special emphasis on Steps 2 and 3, and applied to the power transmission network domain.

Diagnosis of discrete-event systems from uncertain temporal observations

Observations play a major role in diagnosis. The nature of an observation varies according to the class of the considered system. In static systems, an observation is the value of a variable at a single time point. In dynamic continuous systems, such a value is observed over a time interval. In discrete-event systems, an observation consists of a sequence of temporally ordered events. In any case, what is observed is assumed not to be ambiguous. This certainty principle, whilst being a useful simplification for a variety of contexts, may become inappropriate for a wide range of real systems, where the communication between the system and the observer is either bound to generate spurious messages, to randomly lose messages, or to lose temporal constraints among them. Consequently, the observation may be underconstrained. To cope with this uncertainty, a number of principles affecting both the observations and the modeled behavior of a system are introduced, that are independent of any specific processing technique. Furthermore, the notion of an uncertain temporal observation for discrete-event systems is introduced and accommodated within a graph whose nodes are labeled by uncertain messages, while edges define a partial temporal ordering among messages. This way, an uncertain observation implicitly defines a finite set of observations in the traditional sense. Thus, solving an uncertain diagnostic problem amounts to solving at one time several traditional diagnostic problems. The notion of an uncertain observation is further generalized to that of a complex observation. Both notions can be exploited by any diagnostic approach pertinent to discrete-event systems. Complex observations are contextualized in the framework of diagnosis of active systems and substantiated by a sample application in the domain of power transmission networks.

A bridged diagnostic method for the monitoring of polymorphic discrete-event systems

Diagnosis of discrete-event systems (DESs) is a challenging problem that has been tackled both by automatic control and artificial intelligence communities. The relevant approaches share similarities, including modeling by automata, compositional modeling, and model-based reasoning. This paper aims to bridge two complementary approaches from these communities, namely, the diagnoser approach and the active system approach, respectively. The more significant shortcomings of such approaches are, on the one side, the need for the generation of the global system model and, on the other, the lack of monitoring capabilities. The former makes the application of the diagnoser approach prohibitive in real contexts, where the system model is too large to be generated, even offline. The latter requires the completion of the system observation before starting the diagnostic task, thereby, making the monitoring of the system. impossible. The bridged diagnostic method subsumes, to a large extent on the peculiarities of the two approaches and is capable of coping with an extended class of DESs that integrate both synchronous and asynchronous behavior. The bridge is built by extending the active system approach by means of several enhanced techniques, which eventually, allow the efficient monitoring of polymorphic DESs. Upon the occurrence of each system message, two pieces of diagnostic information are generated, namely, the snapshot and historic diagnostic sets. While the former accounts for the faults pertinent to the newly generated message only, the latter is based on the whole sequence of messages yielded by the system during operation.

Flexible diagnosis of discrete-event systems by similarity-based reasoning techniques

Diagnosis of discrete-event systems (DESs) may be improved by knowledge-compilation techniques, where a large amount of model-based reasoning is anticipated off-line, by simulating the behavior of the system and generating suitable data structures (compiled knowledge) embedding diagnostic information. This knowledge is exploited on-line, based on the observation of the system behavior, so as to generate the set of candidate diagnoses (problem solution). This paper makes a step forward: the solution of a diagnostic problem is supported by the solution of another problem, provided the two problems are somewhat similar. Reuse of model-based reasoning is thus achieved by exploiting the diagnostic knowledge yielded for solving previous problems. The technique still works when the available knowledge does not fit the extent of the system, but only a partition of it, that is, when solutions are available for subsystems only. In this case, the fragmented knowledge is exploited in a modular way, where redundant computation is avoided. Similarity-based diagnosis is meant for large-scale DESs, where the degree of similarity among subsystems is high and stringent time constraints on the diagnosis response is a first-class requirement.

On Multisets in Database Systems

Database systems cope with the management of large groups of persistent data in a shared, reliable, effective, and efficient way. Within a database, a multiset (or bag) is a collection of elements of the same type that may contain duplicates. There exists a tight coupling between databases and multisets. First, a large variety of data models explicitly support multiset constructors. Second, commercial relational database systems, even if founded on a formal data model which is set-oriented in nature, allows for the multiset-oriented manipulation of tables. Third, multiset processing in databases may be dictated by efficiency reasons, as the cost of duplicate removal may turn out to be prohibitive. Finally, even in a pure set-oriented conceptual framework, multiset processing may turn out to be appropriate for optimization of query evaluation. The mismatch between the relational model and standardized relational query languages has led researchers to provide a foundation to the manipulation of multisets. Other research has focused on extending the relational model by relaxing the first normal form assumption, giving rise to the notion of a nested relation and to a corresponding nested relational algebra. These two research streams have been integrated within the concept of a complex relation, where different types of constructors other than relation coexist, such as multiset and list. Several other database research areas cope with multiset processing, including view maintenance, data warehousing, and web information discovery.

The Active Layer

Active databases support the creation and execution of active rules. Active rules follow the event-condition-action paradigm, according to which, they autonomously react to events occurring on the data, by evaluating a condition, and by executing a reaction whenever the condition evaluates to true.

Monitoring of Active Systems With Stratified Uncertain Observations

In monitoring-based diagnosis of active systems, the observation is fragmented over time: at the occurrence of each fragment, the internal representation of the observation received so far is updated, new monitoring states are estimated, and a new set of candidate diagnoses is output. When the observation is temporally uncertain, a problem arises about the dependability of the monitoring output: Two consecutive sets of diagnoses, relevant to two consecutive observation fragments, may be unrelated to one another, and, even worse, they may be unrelated to the actual diagnosis. To cope with this problem, the notion of monotonic monitoring is introduced, which is supported by specific constraints on the fragmentation of the uncertain temporal observation, leading to the notion of stratification. Stratified observations support monotonic monitoring of active systems.

Software Prototyping in Data and Knowledge Engineering

Preface. Acknowledgements. 1. The Prototyping Approach to Software Development. 2. Overview of Database Technology. 3. Overview of Knowledge-Based Technology. 4. Data and Knowledge Intensive Systems. 5. The Prototyping Hierarchy. 6. The Relational Layer. 7. The Extended Relational Layer. 8. The Deductive Layer. 9. The Object Layer. 10. The Active Layer. 11. Prototyping Techniques Integration. 12. Mapping Prototypes to Relational Databases. References. Index.

Incremental Determinization of Finite Automata in Model-Based Diagnosis of Active Systems

Generating a deterministic finite automaton (DFA) equivalent to a nondeterministic one (NFA) is traditionally accomplished by subset-construction (SC). This is the right choice in case a single transformation is needed. If, instead, the NFA is repeatedly extended, one transition each time, and the DFA corresponding to each extension is needed in real-time, SC is bound to poor performances. In order to cope with these difficulties, an algorithm called incremental subset-construction (ISC) is proposed, which makes up the new DFA as an extension of the previous DFA, avoiding to start from scratch each time, thereby pursuing computational reuse. Although conceived within the application domain of model-based diagnosis of active systems, the algorithm is general in nature, hence it can be exploited for incremental determinization of any NFA. Massive experimentation indicates that, while comparable in space complexity, incremental determinization of finite automata is, in time, far more efficient than traditional determinization by SC.