Search

This paper investigates the vibration attenuation problem of a non-linear full-car suspension system and aims to stabilize the vehicle attitude to provide a good ride quality. First, with respect to heave motion, pitch motion and roll motion, the full-car suspension system is separated into three interconnected subsystems. For each subsystem, corresponding motion-based controllers are designed to attenuate the vibrations of the sprung mass. A non-linear tracking differentiator is used to track the reference signal and to obtain its derivative. An extended state observer is established to estimate the total disturbance, which includes all the uncertainties and the external disturbance. Based on the principle of active disturbance rejection control, proportional–derivative and fuzzy proportional–derivative controllers are designed to control the resulting linear system with total disturbance compensation. Finally, four actuator forces are computed online using the three motion-based controllers obtained. Simulations are carried out in different road conditions; the results illustrate the merits of the proposed control method.

In this paper, an H∞ controller with actuator saturation consideration is proposed to attenuate the vibration of periodic piecewise vibration systems. Based on a continuous Lyapunov function with a time-varying Lyapunov matrix, the H∞ performance index of periodic piecewise vibration systems is studied first. On the basis of the obtained H∞ criterion, the conditions of designing a state-feedback active vibration controller are proposed in matrix inequality form with actuator saturation taken into account. Because of the nonconvexity of the conditions, a corresponding algorithm to compute the controller gain is developed as well. A representative numerical example is used to verify the effectiveness of the proposed method.

In this paper, we study the suspension performances with an adaptive inerter under the assumption that the inertance may be adjusted in real-time. A quarter-car model with an inerter installed in parallel with a spring and a damper is considered. First, for a given suspension system, a study of the effects of using a fixed inerter is provided. It shows the difficulty in choosing a fixed inerter to satisfy the vehicle performances at the sprung mass natural frequency without deteriorating significantly at the unsprung mass natural frequency. Then, a state-feedback H2 controller for an active suspension system is designed. The active force is approximated by an inerter with adjustable inertance. Ride quality, suspension deflection and tyre deflection performances are considered independently as the prime objective. Simulation results show that comparing with the passive suspension with fixed inerter, for ride quality and tyre deflection, the suspension with adaptive inerter can achieve improvement at the sprung mass natural frequency at the expense of a relatively small deterioration at the unsprung mass natural frequency. For suspension deflection, a better performance can be achieved at high frequencies with almost no loss of performance at other frequencies. This method is easier to implement in practice than the optimized passive suspension with a fixed inerter, which calls for complicated layout.

This paper investigates stability and guaranteed cost of time-triggered Boolean networks (BNs) based on the semitensor product of matrices. The time triggering is generated by mode-dependent average dwell-time switching signals in the BNs. With the help of the copositive Lyapunov function, a sufficient condition is derived to ensure that the considered network is globally stable under a designed average dwell-time switching signal. Subsequently, an infinite time cost function is further discussed and its bound is presented according to the obtained stability result. Numerical examples are finally given to show the feasibility of the theoretical results.

For model reduction, the approximation performance sometimes needs to be enhanced over a specific frequency range. Motivated by this fact, the paper investigates generalized $$H_{\infty }$$H? model reduction for stable two-dimensional (2-D) discrete systems represented by the Roesser model and the Fornasini---Machesini local state-space model, respectively. The generalized $$H_{\infty }$$H? norm of 2-D systems is introduced to evaluate the approximation error over a specific finite frequency (FF) domain. In light of the 2-D generalized Kalman---Yakubovich---Popov lemmas, sufficient conditions in terms of linear matrix inequalities are derived for the existence of a stable reduced-order model satisfying a specified generalized $$H_{\infty }$$H? level. Several examples are provided to illustrate the effectiveness and advantages of the proposed method. Compared with most of the existing 2-D model reduction results, the proposed method has the following merits: (1) The generalized $$H_\infty $$H? model reduction problems are considered for both important types of 2-D models, and no structural assumption is made for the plant model, so that our method has a broader applicable scope. (2) The reduced-order model is guaranteed to be stable, and an upper bound on the generalized $$H_{\infty }$$H? error is provided. Moreover, no frequency weighting function is needed. (3) The proposed method is applicable for 2-D model reduction with multiple FF specifications.

In this paper, the novel design of a side-slip angle observer and an observer-based stability controller for a vehicle with a non-linear tyre model and a varying forward speed is presented. The Takagi–Sugeno fuzzy modelling technique is first applied to represent the vehicle’s lateral dynamics with the non-linear Dugoff tyre model and varying speed. The observer and the observer-based controller are then constructed using the measured yaw rate and the estimated premise and state variables and are designed to be robust against the model and parameter uncertainties. The conditions for designing such an observer and an observer-based controller are derived in terms of linear matrix inequalities. Numerical simulations on a non-linear eight-degree-of-freedom vehicle dynamics model and experimental data are conducted to validate the effectiveness of the proposed approach. The comparison of results with those from other existing approaches shows that the designed observer can accurately estimate the vehicle’s side-slip angle and the controller can effectively control this side-slip angle regardless of the variation in the vehicle’s longitudinal velocity.

This paper presents a multi-objective control method with wheelbase preview for active vehicle suspension. A four-degree-of-freedom half-car model with active suspension is considered in this study. H ∞ norm and generalized H 2 norm are used to improve ride quality and ensure that hard constraints are satisfied. Disturbances at the front wheel are obtained as preview information for the rear wheel. Static output-feedback is utilized in designing controllers, the solution is derived by iterative linear matrix inequality (ILMI) and cone complementarity linearization (CCL) algorithms. Simulation results confirm that multi-objective control with wheelbase preview achieves a significant improvement of ride quality (a maximum 27 percent and 60 percent improvement on vertical and angular acceleration, respectively) comparing with that of control without preview, while suspension deflections, tyre deflections and actuator forces remaining within given bounds. The extent of the improvement on the ride quality for different amount of preview information used is also illustrated.

This paper presents a velocity-dependent multi-objective control method to solve the problem of preview control with velocity uncertainty. A half-car model is considered, the Pade approach is used to approximate the preview delay so as to rewrite it in a finite-dimensional description. It is assumed that the forward velocity of the vehicle resides in an interval and can be measured in real time. The controller, whose gain matrix depends on the information of the velocity, is designed by utilizing the polynomial parameter-dependent idea. The design procedure is developed based on homogeneous polynomial parameter-dependent matrices with arbitrary degree, with quadratic and linear parameter-dependent frameworks as special cases of this method. As the degree increases, the proposed controller can yield less conservative result. A linear parameter-dependent (LPV) controller has also been proposed for the system under time-varying velocity. Simulation results illustrate the usefulness and the advantages of the proposed methods.

This paper puts forward a novel optimization approach for Multiple Active Tuned Mass Dampers (MATMDs) system under seismic and wind-induced building vibration. A model of an n -storey building with MATMD system is established and a joint optimization method is used to obtain an optimal state-feedback controller gain and the parameters of the MATMD system. A mixed H 2 / H ∞ / GH 2 control is employed to attenuate the seismic and wind-induced vibration of the building with the constraints of the actuating forces and strokes of the masses. Genetic algorithm (GA) is used to search for the optimal parameters and obtain the corresponding controller gain. Two illustrative examples are presented in this paper. In the first comparison, the GA-based approach can obtain a better set of parameters and achieve better control performance. When comparing with an Active Tuned Mass Damper (ATMD), an MATMD system can achieve similar control effects with much smaller acting forces.

In this paper, a novel monitoring approach is developed for nonlinear processes based on available measurements. To cope with complicated process nonlinearity, we implement total projection to latent structures (T-PLS) in each local model created by the locally weighted projection regression (LWPR) algorithm. Under the framework of locally weighted learning, four improved test statistics are established to detect potential process faults. The test statistics are not only capable to monitor the abnormal changes in the relevant subspaces, but also with thresholds suitable for non-Gaussian measurements. The effectiveness of our proposed approach is further demonstrated by a numerical nonlinear case.

This paper is concerned with the problem of passivity analysis for uncertain continuous-time neural networks with mixed time-varying delays. The mixed time-varying delays consist of both discrete and distributed delays, in which the discrete delays are assumed to be varying within a given interval. In addition, the uncertainties are assumed to be norm-bounded. By employing a novel Lyapunov–Krasovskii functional, new passivity delay-interval-dependent criteria are established to guarantee the passivity performance. When estimating an upper bound of the derivative of the Lyapunov–Krasovskii functional, we handle the terms related to the discrete and distributed delays appropriately so as to develop less conservative results. These passivity conditions are presented in terms of linear matrix inequalities, which can be easily solved via standard numerical software. Some numerical examples are given to illustrate the effectiveness of the proposed method.

This paper investigates the problem of robust stability for a class of stochastic interval neural networks with discrete and distributed time-varying delays. The discrete delays are assumed to be varying within a given interval, while the parameter uncertainties are assumed to be bounded in some given compact sets. Based on the Ito differential formula and stochastic stability theory, delay-range-dependent criteria for stochastic interval neural networks with time-varying delays are derived to guarantee robust global asymptotic stability in mean square. In this paper, when estimating the upper bound of the derivative of Lyapunov functionals, we are concerned with better handling of terms related to the discrete and distributed delays and establishing less conservative results. These robust stability conditions are presented in terms of linear matrix inequalities and can be efficiently solved via standard numerical software. An important feature of the results proposed in this paper is that the stability conditions are dependent on the upper and lower bounds of the discrete delays. Numerical examples are given to illustrate the effectiveness of the proposed method.

In this paper, stability and stabilization of Boolean networks with stochastic delays are studied via semi-tensor product of matrices. The stochastic delays, randomly attaining finite values, are modeled by Markov chains. By utilizing an augmented method, the considered Boolean network is first converted into two coupled Markovian switching systems without delays. Then, some stochastic stability results are obtained based on stability results of positive systems. Subsequently, the stabilization of Boolean networks with stochastic delays is further investigated, and an equivalent condition for the existence of feedback controllers is provided in terms of a convex programming problem, which can be easily solved and also conveniently applied to design controller gains. Finally, numerical examples are given to illustrate feasibility of the obtained results.

In the common practice of designing an attitude tracker for an aerospacecraft, one transforms the Newton-Euler rotation equations to obtain the dynamic equations of some chosen inertial frame based attitude metrics, such as Euler angles and unit quaternions. A Lyapunov approach is then used to design a controller which ensures asymptotic convergence of the attitude to the desired orientation. Although this design methodology is pretty standard, it usually involves singularity-prone coordinate transformations which complicates the analysis process and controller design. A new, singularity free error feedback method is proposed in the paper to provide simple and intuitive stability analysis and controller synthesis. This new body frame based method utilizes the concept of Euleraxis and angles to generate the smallest error angles from a body frame perspective, without coordinate transformations. Global tracking convergence is illustrated with the use of a feedback linearizing PD tracker, a sliding mode controller, and a model reference adaptive controller. Experimental results are also obtained on a quadrotor platform with unknown system parameters and disturbances, using a boundary layer approximated sliding mode controller, a PIDD controller, and a unit sliding mode controller. Significant tracking quality is attained.

The key performance indicator (KPI) has an important practical value with respect to the product quality and economic benefits for modern industry. To cope with the KPI prognosis issue under nonlinear conditions, this paper presents an improved incremental learning approach based on available process measurements. The proposed approach takes advantage of the algorithm overlapping of locally weighted projection regression (LWPR) and partial least squares (PLS), implementing the PLS-based prognosis in each locally linear model produced by the incremental learning process of LWPR. The global prognosis results including KPI prediction and process monitoring are obtained from the corresponding normalized weighted means of all the local models. The statistical indicators for prognosis are enhanced as well by the design of novel KPI-related and KPI-unrelated statistics with suitable control limits for non-Gaussian data. For application-oriented purpose, the process measurements from real datasets of a proton exchange membrane fuel cell system are employed to demonstrate the effectiveness of KPI prognosis. The proposed approach is finally extended to a long-term voltage prediction for potential reference of further fuel cell applications.

This paper investigates the stability and L 2 -gain problems for a class of continuous-time periodic piecewise linear systems with possibly non-Hurwitz subsystems. First, the exponential stability of periodic piecewise systems is studied by allowing the Lyapunov function to possibly non-monotonically decreasing over a period. A sufficient condition is established in terms of matrix inequalities. In light of the proposed Lyapunov function, the L 2 -gain criterion is derived for periodic piecewise linear systems as well.

This paper analyses the generalized H 2 performance for a class of continuous-time periodic piecewise linear systems with possibly non-Hurwitz subsystems. A sufficient criterion is proposed in terms of matrix inequalities by employing a time-varying Lyapunov function which is formulated with periodic a continuous linear time-varying Lyapunov matrix. The Lyapunov function deceases non-monotonically over a period. Also, by using a Lyapunov function with common decreasing rates respectively for all Hurwitz subsystems and all non-Hurwitz subsystems over their operating intervals, a simplified stability criterion is established. A numerical example is given to illustrate the effect and feasibility of the proposed techniques.

The paper investigates generalized H ∞ model reduction for two-dimensional (2-D) systems represented by the Roesser model and the Fornasini-Machesini local state-space model, respectively. The generalized H ∞ norm of 2-D systems is introduced to evaluate the approximation error over a specific finite frequency (FF) domain. In light of the 2-D generalized Kalman-Yakubovich-Popov lemmas, sufficient conditions in terms of linear matrix inequalities are derived for the existence of a stable reduced-order model satisfying a specified generalized H ∞ level. Several examples are provided to illustrate the effectiveness and advantages of the proposed method. Compared with most of the existing results, the proposed method has the following merits: 1) Both important types of 2-D models are considered in a unified framework, and no structural assumption is made for the plant model. 2) An upper bound on the generalized H ∞ error can be obtained, and no weighting function is needed. 3) The proposed method is applicable to multiple FF specifications.

The paper presents a study on the direct voltage control of a magnetorheological (MR) damper for application in vehicle suspensions. As MR damper dynamics is highly nonlinear, the direct control system design for an MR damper is difficult. Representing an MR damper by a Takagi-Sugeno (TS) fuzzy model enables the linear control theory to be directly applied to design the MR damper controller. In this paper, first the MR damper dynamics is represented by a TS fuzzy model, and then an H∞ controller that considers the suspension performance requirements and the constraint on the input voltage for the MR damper is designed. Furthermore, considering the case that not all the state variables are measurable in practice, the design of an H∞ observer with immeasurable premise variables and the design of a robust controller are proposed, respectively. Numerical simulations are used to validate the effectiveness of the proposed approaches. © 2013 IOP Publishing Ltd.

In the process industry, the key performance indicator (KPI)-related prediction and fault diagnosis are important steps to guarantee the product quality and improve economic benefits. A popular monitoring method as it has been, the partial least squares (PLS) algorithm is sensitive to outliers in training datasets, and cannot efficiently distinguish faults related to KPI from those unrelated to KPI due to its oblique projection to the input space. In this paper, a novel robust data-driven approach, named advanced partial least squares (APLS), is presented to handle process outliers under an improved framework of PLS. By means of a weighting strategy, APLS can remove the impact of outliers on process measurements and establish a more accurate model than PLS for fault diagnosis in the monitoring scheme, whose effectiveness has been verified through the Tennessee Eastman (TE) benchmark process. Simulation results demonstrate that the proposed approach is suitable not only for the KPI-related process prediction but also for the diagnosis of KPI-related faults.

A novel approach is presented for fully automated boundary extraction and rectification of bony tissue from planar CT data. The approach extracts and rectifies feature boundary in a hierarchical fashion. It consists of a fuzzy multilevel thresholding operation, followed by a small void cleanup procedure. Then a binary morphological boundary detector is applied to extract the boundary. However, defective boundaries and undesirable artifacts may still be present. Thus two innovative anatomical knowledge based algorithms are used to remove the undesired structures and refine the erroneous boundary. Results of applying the approach on lumbar CT images are presented, with a discussion of the potential for clinical application of the approach.