Neural networks for nonlinear state estimation

Estimating the state of a nonlinear stochastic system (observed through a nonlinear noisy measurement channel) has been the goal of considerable research to solve both filtering and control problems. In this paper, an original approach to the solution of the optimal state estimation problem by means of neural networks is proposed, which consists in constraining the state estimator to take on the structure of a multilayer feedforward network. Both non-recursive and recursive estimation schemes are considered, which enable one to reduce the original functional problem to a nonlinear programming one. As this reduction entails approximations for the optimal estimation strategy, quantitative results on the accuracy of such approximations are reported. Simulation results confirm the effectiveness of the proposed method.     

Active state estimation for nonlinear systems: a neural approximation approach.

In this paper, we consider the problem of actively providing an estimate of the state of a stochastic dynamic system over a (possibly long) finite time horizon. The active estimation problem (AEP) is formulated as a stochastic optimal control one, in which the minimization of a suitable uncertainty measure is carried out. Toward this end, the use of the Renyi entropy as an information measure is proposed and motivated. A neural control scheme, based on the application of the extended Ritz method (ERIM) and on the use of a Gaussian sum filter (GSF), is then presented. Simulation results show the effectiveness of the proposed approach.     

Global asymptotic stabilization of the attitude and the angular rates of an underactuated non-symmetric rigid body

The paper deals with the global stabilization of both the attitude and the angular velocities of an underactuated rigid body. First a stability theorem is proven for a class of systems; subsequently, the equations describing the physics of the rigid body are presented, showing that the rigid body belongs to the considered class of systems, and a sufficient condition for the application of the theorem to the stability of the rigid body equilibrium is pointed out. Finally, some simulation results are reported showing the effectiveness of the proposed methodology.     

Passive Attack Detection for a Class of Stealthy Intermittent Integrity Attacks

This paper proposes a passive methodology for detecting a class of stealthy intermittent integrity attacks in cyber-physical systems subject to process disturbances and measurement noise. A stealthy intermittent integrity attack strategy is first proposed by modifying a zero-dynamics attack model. The stealthiness of the generated attacks is rigorously investigated under the condition that the adversary does not know precisely the system state values. In order to help detect such attacks, a backward-in-time detection residual is proposed based on an equivalent quantity of the system state change, due to the attack, at a time prior to the attack occurrence time. A key characteristic of this residual is that its magnitude increases every time a new attack occurs. To estimate this unknown residual, an optimal fixed-point smoother is proposed by minimizing a piece-wise linear quadratic cost function with a set of specifically designed weighting matrices. The smoother design guarantees robustness with respect to process disturbances and measurement noise, and is also able to maintain sensitivity as time progresses to intermittent integrity attack by resetting the covariance matrix based on the weighting matrices. The adaptive threshold is designed based on the estimated backward-in-time residual, and the attack detectability analysis is rigorously investigated to characterize quantitatively the class of attacks that can be detected by the proposed methodology. Finally, a simulation example is used to demonstrate the effectiveness of the developed methodology.      

Distributed Fault Diagnosis With Overlapping Decompositions: An Adaptive Approximation Approach

This technical note deals with the problem of designing a distributed fault detection methodology for distributed (and possibly large-scale) nonlinear dynamical systems that are modelled as the interconnection of several subsystems. The subsystems are allowed to overlap, thus sharing some state components. For each subsystem, a local fault detector is designed, based on the measured local state of the subsystem as well as the transmitted variables of neighboring states that define the subsystem interconnections. The local detection decision is made on the basis of the knowledge of the local subsystem dynamic model and of an adaptive approximation of the interconnection with neighboring subsystems. The use of a specially-designed consensus-based estimator is proposed in order to improve the detectability of faults affecting variables shared among different subsystems. Simulation results provide an evidence of the effectiveness of the proposed distributed fault detection scheme.     

Robust fault isolation for a class of non-linear input?output systems

The design and analysis of fault diagnosis methodologies for non-linear systems has received significant attention recently. This paper presents a robust fault isolation scheme for a class of non-linear systems with unstructured modelling uncertainty and partial state measurement. The proposed fault diagnosis architecture consists of a fault detection and approximation estimator and a bank of isolation estimators. Each isolation estimator corresponds to a particular type of fault in the fault class. A fault isolation decision scheme is presented with guaranteed performance. If at least one component of the output estimation error of a particular fault isolation estimator exceeds the corresponding adaptive threshold at some finite time, then the occurrence of that type of fault can be excluded. Fault isolation is achieved if this is valid for all but one isolation estimator. Based on the class of non-linear systems under consideration, fault isolability conditions are rigorously investigated, characterizing the class of non-linear faults that are isolable by the proposed scheme. Moreover, the non-conservativeness of the fault isolability conditions is illustrated by deriving a subclass of nonlinear systems and faults for which this condition is also necessary for fault isolability. A simulation example of a simple robotic system is used to show the effectiveness of the robust fault isolation methodology.     

Superlattices as Characterisation Tool for the Beginning of PS Formation

Initial stage of porous silicon (PS) formation has been studied in an original way. Multilayer structures constituting of very thin layers of low porosity and thick layers of high porosity have been fabricated and characterised by optical tools and electron microscopy. The non linear behaviour resulting in a change in the dissolution velocity has been quantified by using a stack layer structure. Finally using thermal oxidation it has been shown that, due to the selective oxidation as a function of the porosity, porous silicon can be used to produce a Si/SiO₂ like structure.     

Nonlinear modeling of complex large-scale plants using neuralnetworks and stochastic approximation

This paper deals with a general methodology for system greybox identification. As is well-known, the tuning of accurate models of real plants (obtained, for instance, by using the physical knowledge of the plants and the technicians' expertise), on the basis of the measures provided by the available sensors, remains a challenge. In this paper, a tuning methodology for complex large-scale models, is presented. The proposed technique is based on the suitable use of neural networks and specific stochastic-approximation algorithms. It is therefore possible to design a simulator that can be connected in parallel with a real plant, thus providing the plant technician with information about inaccessible variables that are useful for supervision purposes. The proposed methodology is applied to a section of a real 320 MW power plant. Simulation results on the tuning algorithm show the effectiveness of the approach