Abstract-The failure detection and identification (FDI) process is viewed as consisting of two stages residual generation and decision making. It is argued that a robust FDI system can be achieved by designing a robust residual generation process. Analytical redundancy, the basis for residual generation, is characterized in terms of a pm'ty space. Using the concept of parity relations, residuals can be generated in a number of ways and the design of a robust residual generation process can be formulated as a minimax optimization problem. An example is included to illustrate this design methodology.I. INTRODUCTION P HYSICAL systems are often subjected to unexpected changes, such as component failures and variations in operating condition, that tend to degrade overall system performance. We will refer to such changes as "failures," although they may not represent the failing of physical components. In order to maintain a high level of performance, it is important that failures be promptly detected and identified so that appropriate remedies can be applied. Over the past decade numerous approaches to the problem of failure detection and identification (FDI) [SI, [6] are some examples. All of these analytical methods require that a dynamic model of some sort be given. The goal of this paper is to investigate the issue of designing FDI systems which are robust to uncertainties i n the models on which they are based.The FDI process essentially consists of two stages: residual generation and decision making. For a particular set of hypothesized failures, an FDI system has the structure shown in Fig. 1. Outputs from sensors are initially processed to enhance the effect of a failure (if present) so that it can be recognized. The processed measurements are called the residuals, and the enhanced failure effect on the residuals is called the signature of the failure. Intuitively, the residuals represent the difference between various functions of the observed sensor outputs and the expected values of these functions in the normal (no-fail) mode. In the absence of a failure residuals should be unbiased, showing agreement between observed and expected normal behavior of the system; a failure signature typically takes the form of residual biases that are characteristic of the failure. Thus, residual generation is based on knowledge of the normal behavior of the system. The actual process of residual generation can vary in complexity. For example, in voting systems [7], [X] the residuals are simply the difrecommended by T. Basar, Past Chairman of the Large Scale Systems.

In this paper we discuss an approach to the detection of incidents on freeways. Our techniques are based on the use of a macroscopic dynamic model describing the evolution of spatialaverage traffic variables (velocities, flows, and densities) over sections of the freeway. With such a model as a starting point we develop two incident detection algorithms based on the multiple model and generalized likelihood ratio techniques. We also describe a new and very simple system for processing raw data from presence-type vehicle detectors to produce estimates of the aggregate variables, which are then in turn used as the input variables to the incident detection algorithms. Simulation results using a microscopic simulation of a two-lane freeway indicate that:(1) our algorithms are robust to the differences between the dynamics of actual traffic and the aggregated dynamics used to design the detection systems; and (2) our methods appear to work as well as existing algorithms in heavy traffic conditions and work better in moderate to light traffic. Areas for future work are outlined at the end of the paper.

No abstract

This paper considers the simultaneous reconstruction of permittivity and conductivity profiles in a cylindrically stratified geometry. It is assumed that both the permittivity and conductivity may only vary in the radial direction, and that the unknown region, to be probed, is an annulus of known thickness. An iterative method for the simultaneous inversion of the radial permittivity and conductivity profiles inside this annulus is presented. This method employs a distorted‐Born approximation at each iteration step. The inherent nonuniqueness of the problem is circumvented by imposing additional constraints that limit the set of feasible solutions. Numerical simulations show that good results can be obtained for smoothly varying profiles even when a limited number of measurements at a single frequency are used.