Most model-based diagnosis systems, such as GDE and Sherlock, have concerned discrete, static systems such as logic circuits and use simple constraint propagation to detect inconsistencies. However, sophisticated systems such as QSIM and QPE have been developed for qualitative modeling and simulation of continuous dynamic systems. We present an integration of these two lines of research as implemented in a system called QDOCS for multiple-fault diagnosis of continuous dynamic systems using QSIM models. The main contributions of the algorithm include a method for propagating dependencies while solving a general constraint satisfaction problem and a method for verifying the consistency of a behavior with a model across time. Through systematic experiments on two realistic engineering systems, we demonstrate that QDOCS demonstrates the best balance of generality, accuracy, and efficiency among competing methods.
ML ID: 64
As systems like chemical plants, power plants, and automobiles get more complex, online diagnostic systems are becomingly increasingly important. One of the ways to rein in the complexity of describing and reasoning about large systems such as these is to describe them using qualitative rather than quantitative models.Model-based diagnosis is a class of diagnostic techniques that use direct knowledge about how a system functions instead of expert rules detailing causes for every possible set of symptons of a broken system. Our research builds on standard methods for model-based diagnosis and extends them to the domain of complex dynamic systems described using qualitative models.
We motivate and describe out algorithm for diagnosing faults in a dynamic system given a qualitative model and a sequence of qualitative states. The main contributions in this algorithm include a method for propagating dependencies while solving a general constraint satisfaction problem, and a method for verfying the compatibility of a behavior with a model across time. The algorithm can diagnose multiple faults and uses models of faulty behavior, or behavioral modes.
We then demonstrate these techniques using an implemented program called QDOCS and test it on some realistic problems. Through our experiments with a model of the reaction control system (RCS) of the space shuttle and with a level-controller for a reaction tank, we show that QDOCS demonstrates the best balance of generality, accuracy and efficiency among known systems.
ML ID: 49
This paper describes an approach to diagnosis of systems described by qualitative differential equations represented as QSIM models. An implemented system QDOCS is described that performs multiple-fault, fault-model based diagnosis, using constraint satisfaction techniques, of qualitative behaviors of systems described by such models. We demonstrate the utility of this system by accurately diagnosing randomly generated faults using simulated behaviors of a portion of the Reaction Control System of the space shuttle.
ML ID: 46
This paper describes an approach to diagnosis of systems described by qualitative differential equations represented as QSIM models. An implemented system QDOCS is described that performs multiple-fault, fault-model based diagnosis, using constraint satisfaction techniques, of qualitative behaviors of systems described by such models. We demonstrate the utility of this system by accurately diagnosing randomly generated faults using simulated behaviors of a portion of the Reaction Control System of the space shuttle.
ML ID: 39
The problem of learning qualitative models of physical systems from observations of its behaviour has been addressed by several researchers in recent years. Most current techniques limit themselves to learning a single qualitative differential equation to model the entire system. However, many systems have several qualitative differential equations underlying them. In this paper, we present an approach to learning the models for such systems. Our technique divides the behaviours into segments, each of which can be explained by a single qualitative differential equation. The qualitative model for each segment can be generated using any of the existing techniques for learning a single model. We show that results of applying our technique to several examples and demonstrate that it is effective.
ML ID: 34
We describe a method of automatically abducing qualitative models from descriptions of behaviors. We generate, from either quantitative or qualitative data, models in the form of qualitative differential equations suitable for use by QSIM. Constraints are generated and filtered both by comparison with the input behaviors and by dimensional analysis. If the user provides complete information on the input behaviors and the dimensions of the input variables, the resulting model is unique, maximally constrainted, and guaranteed to reproduce the input behaviors. If the user provides incomplete information, our method will still generate a model which reproduces the input behaviors, but the model may no longer be unique. Incompleteness can take several forms: missing dimensions, values of variables, or entire variables.
ML ID: 17
