Search

Abstract In this paper, two iterative algorithms are proposed to solve stochastic algebraic Riccati matrix equations arising in the linear quadratic optimal control problem of linear stochastic systems with state-dependent noise. In the first algorithm, a standard Riccati matrix equation needs to be solved at each iteration step, and in the second algorithm a standard Lyapunov matrix equation needs to be solved at each iteration step. In the proposed algorithms, a weighted average of the estimates in the last and the previous steps is used to update the estimate of the unknown variable at each iteration step. Some properties of the sequences generated by these algorithms under appropriate initial conditions are presented, and the convergence properties of the proposed algorithms are analyzed. Finally, two numerical examples are employed to show the effectiveness of the proposed algorithms. [ABSTRACT FROM AUTHOR]

• Active fault-tolerant strategy is introduced to distributed state estimation scheme. • Intermittent fault (IF) detector is designed and the detectability of IF is analyzed. • Estimation errors can remain the prescribed performance under the influence of IF. In this paper, the problem of distributed fault-tolerant state estimation is studied for stochastic systems over sensor networks with intermittent sensor faults. Compared with the traditional state estimation algorithms, the distinct advantage of fault-tolerant state estimation is that the estimator can keep good performance whether sensor faults occur or not. Different from the previous literature concerning with distributed fault diagnosis, the distributed fault diagnosis problem is investigated in this paper for intermittent faults, whose appearing time, disappearing time and magnitude are all nondeterministic. The distributed fault-tolerant state estimation scheme is constructed, in which the appearing time and disappearing time of intermittent faults are detected, intermittent faults are estimated and compensated. By means of the matrix inequality technique, the H ∞ performance of state estimation errors is guaranteed by properly choosing the estimator parameters. Finally, two examples are provided to demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

• It is the first time to investigate the problem of decentralized finite-time tracking control with prescribed settling time for a class of p -normal form stochastic large-scale systems with output interconnections existing in both the drift and diffusion terms. • It is the first time to investigate the connective stability problem for stochastic large-scale systems when the system structure changes. • Differing from the existing results, this paper proposes a new mathematical treatment algorithm to deal with the interconnections of stochastic large-scale systems, such that the decentralized tracking control is achieved. • The proposed control strategy is more convenient for practical application than the general finite-time control. • Due to the complexity of the system mathematics model, this paper can better reflect the application of mathematics. This paper aims to solve the decentralized finite-time connective tracking control problem for p -normal form stochastic large-scale systems with output interconnections existing in both the drift and diffusion terms. By means of the stochastic system theory and a prescribed finite-time performance function (PFTPF), a novel design scheme is presented for the decentralized finite-time connective tracking controllers with an arbitrarily prescribed settling time. The connective stability problem of stochastic large-scale systems is investigated for the first time. In addition, a new solution for the decentralized tracking control problem of stochastic large-scale systems is presented via a novel mathematical treatment algorithm. The proposed controllers can ensure that the tracking errors converge to a predetermined region within an arbitrarily prescribed settling time and the controlled system is connectively bounded stable in probability. Three simulation examples are presented to exhibit the performance and the superiority of the new control strategy. [ABSTRACT FROM AUTHOR]

• For optimal control, the stochastic information with current iterative results is used to solve the stochastic algebraic Riccati equation, which has stochastic characteristics. • Calculate the increment instead of the value itself to improve the calculation accuracy. • Newton's method with line search solve the potentially disastrous problem of the first Newton step. • The convergence and even quadratic convergence of the proposed incremental Newton's iterative algorithm have been discussed. In this paper, a novel incremental Newton's iterative algorithm for investigating the optimal control problem of Itô stochastic systems is presented. Newton's method is employed under the Fréchet derivative framework to iteratively solve a stochastic algebraic Riccati equation. Under moderate conditions, the convergence and even quadratic convergence of the proposed incremental Newton's iterative algorithm are discussed, respectively. In addition, the Newton's method is extended to the one with linear search. In the end, numerical results are given to demonstrate the effectiveness and superiority of the proposed algorithms. [ABSTRACT FROM AUTHOR]

In this paper we present the convergence and stochastic stability analysis of a set of PSO variants: those that differ with the classical PSO in the statistical distribution of the three PSO tuning parameters: inertia weight, local and global acceleration. We provide an analytical expression for the upper limit of the second order stability regions (the so called USL curves) of the particle trajectories that can be applied to most of these PSO algorithms. Thus, this work generalizes to this set the result found in the literature for the classical PSO. We apply this analysis to some of these variants. Finally, numerical experiments have been performed that confirm the known fact that the best algorithm performance is provided tuning the PSO parameters close to the USL curve. [ABSTRACT FROM AUTHOR]

This paper proposes a generalized extended stochastic gradient (GESG) algorithm for estimating the parameters of a class of Wiener nonlinear autoregressive moving average systems using the gradient search. In order to improve the convergence rates of the GESG algorithm, a multi-innovation GESG algorithm is derived. The simulation results indicate that the proposed algorithms can effectively estimate the parameters of a class of output nonlinear systems. [ABSTRACT FROM AUTHOR]

• We consider the constrained stochastic convex optimization by projected subgradient algorithm. • A new convergence analysis approach is developed to prove the validity of the proposed algorithm. • Convergence rate is given without specifying the gains. This paper discusses a kind of convex optimization problem considering noises from information transmission in multi-agent systems. Different from previous works, we focus on the objective function which is a summation of strictly L 0 (F) -convex functions under random inner space. Our system is described by Itô formula, which leads to that it is hard to calculate second-order derivative when designing the projected subgradient algorithm. It is shown that all states in stochastic system will converge to the unique optimal state in the polyhedric set constraint by adopting projected subgradient algorithm and the convergence rate is also investigated. Numerical examples are provided to demonstrate the results. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents an ant colony optimization (ACO) algorithm for continuous functions based on novel pheromone updating. At the end of the each iteration in the proposed algorithm, pheromone is updated according to percentiles which determine the number of ants to track the best candidate solution. It is performed by means of solution archive and information provided by previous solutions. Performance of the proposed algorithm is tested on ten benchmark problems found in the literature and compared with performances of previous methods. The results show that ACO which is based on novel pheromone updating scheme (ACO-NPU) handles different types of continuous functions very well and can be a robust alternative approach to other stochastic search algorithms. [Copyright &y& Elsevier]

Abstract: A recursive algorithm for nonlinear dynamic system identification is studied in this paper. The structure and parameters can be obtained simultaneously and recursively not only for the linear-in-the-parameters models, but applicable to general nonlinear system. It is based on the ‘innovation’ idea and net information contribution criteria. Using the recursive formulae for computing the Moore–Penrose inverse of a matrix, it is possible to combine innovation calculation with structure term determination and parameters estimation. The recursive algorithm has the advantage of computational simplicity. Simulation examples are given which show the proposed method is numerically more stable than the approach existed. [Copyright &y& Elsevier]

Abstract: For multi-input, single-output output-error systems, a difficulty in identification is that the information vector in the identification model obtained contains unknown inner/intermediate variables; thus the standard least squares methods cannot be applied directly. In this paper, we present a multi-innovation least squares identification algorithm based on the auxiliary model; its basic idea is to replace the unknown inner variables with their estimates computed by an auxiliary model. Convergence analysis indicates that the parameter estimation error converges to zero under persistent excitation. The algorithm proposed has significant computational advantage over existing identification algorithms. A simulation example is included. [Copyright &y& Elsevier]