Estimator Gain Matrices Research Articles

Improved Results on Guaranteed Generalized $${\mathcal {H}}_{2}$$ H 2 Performance State Estimation for Delayed Static Neural Networks

This paper is concerned with the guaranteed generalized $${\mathcal {H}}_{2}$$ performance state estimation for a class of static neural networks with a time-varying delay. A more general Arcak-type state estimator rather than the Luenberger-type state estimator is adopted to deal with this problem. Based on the Lyapunov stability theory, the inequality techniques and the delay-partitioning approach, some novel delay-dependent design criteria in terms of linear matrix inequalities (LMIs) are proposed ensuring that the resulting error system is globally asymptotically stable and a prescribed generalized $${\mathcal {H}}_{2}$$ performance is guaranteed. The estimator gain matrices can be derived by solving the LMIs. Compared with the existing results, the sufficient conditions presented in this paper are with less conservatism. Numerical examples are given to illustrate the effectiveness and superiority of the developed method over the existing approaches. A comparison between the Arcak-type state estimator and Luenberger-type state estimator is given simultaneously.

A New Result on $H_{\infty }$ State Estimation of Delayed Static Neural Networks.

This brief presents a new guaranteed performance state estimation criterion for delayed static neural networks. To facilitate the use of the slope information about activation function, the estimation error of activation function is separated into two parts for the first time. Then, a novel Lyapunov-Krasovskii functional (LKF) is constructed, which has fully captured the slope information of the activation. Based on the new LKF, a less conservative design criterion of estimator is derived to ensure the asymptotic stability of estimation error system with performance. The desired estimator gain matrices and the performance index are obtained by solving a convex optimization problem. The simulation results show that the proposed method has much better performance than the most recent results.

Event-Triggered State Estimation for Complex Networks With Mixed Time Delays via Sampled Data Information: The Continuous-Time Case.

In this paper, the event-triggered state estimation problem is investigated for a class of complex networks with mixed time delays using sampled data information. A novel state estimator is presented to estimate the network states. A new event-triggered transmission scheme is proposed to reduce unnecessary network traffic between the sensors and the estimator, where the sampled data is transmitted to the estimator only when the so-called "event-triggered condition" is satisfied. The purpose of the problem addressed is to design an estimator for the complex network such that the estimation error is ultimately bounded in mean square. By utilizing Lyapunov theory combined with the stochastic analysis approach, sufficient conditions are established to guarantee the ultimate boundedness of the estimation error in mean square. Then, the desired estimator gain matrices are obtained via solving a convex problem. Finally, a numerical example is given to illustrate the effectiveness of the results.

Robust Fault Detection and Isolation Using Robust l1 Estimation

The application of robust 1 estimation to robust fault detection and isolation (FDI) is considered. This is accomplished by developing a series, or bank, of robust estimators (full-order observers), each of which is designed such that the residual will be sensitive to a certain fault (or faults) while insensitive to the remaining faults. Robustness is incorporated by ensuring that the residual remains insensitive to exogenous disturbances as well as modeling uncertainty. Mixed structured singular value and 1 theories are used to develop the appropriate threshold logic to evaluate the outputs of the estimators used for determining the occurrence and location of a fault. A real-coded genetic algorithm is used to obtain the estimator gain matrices. This approach to FDI is successfully demonstrated using a linearized model of a jet engine.

Optimal low-order controller design via ‘LQG-like’ parametrization

A new LQG-like parametrization scheme is introduced for low-order controller design. Descent-based design methodologies are developed for optimizing an H 2 cost with respect to the controller design parameters, in this case, the reduced-order regulator and estimator gain matrices. In a modified version, the reduced-order model of the plant is also updated. Numerical experiments using large-space structure models show the efficacy of this approach.