Sy Dzung Nguyen

A new neuro-fuzzy training algorithm for identifying dynamic characteristics of smart dampers

This paper proposes a new algorithm, named establishing neuro-fuzzy system (ENFS), to identify dynamic characteristics of smart dampers such as magnetorheological (MR) and electrorheological (ER) dampers. In the ENFS, data clustering is performed based on the proposed algorithm named partitioning data space (PDS). Firstly, the PDS builds data clusters in joint input?output data space with appropriate constraints. The role of these constraints is to create reasonable data distribution in clusters. The ENFS then uses these clusters to perform the following tasks. Firstly, the fuzzy sets expressing characteristics of data clusters are established. The structure of the fuzzy sets is adjusted to be suitable for features of the data set. Secondly, an appropriate structure of neuro-fuzzy (NF) expressed by an optimal number of labeled data clusters and the fuzzy-set groups is determined. After the ENFS is introduced, its effectiveness is evaluated by a prediction-error-comparative work between the proposed method and some other methods in identifying numerical data sets such as ?daily data of stock A?, or in identifying a function. The ENFS is then applied to identify damping force characteristics of the smart dampers. In order to evaluate the effectiveness of the ENFS in identifying the damping forces of the smart dampers, the prediction errors are presented by comparing with experimental results.

Design of active suspension controller for train cars based on sliding mode control, uncertainty observer and neuro-fuzzy system

This paper focuses on building a controller for active suspension system of train cars in the case that the sprung mass and model error are uncertainty parameters. The sprung mass is always varied due to many reasons such as changing of the passengers and load or impacting of wind on the operating train while an unknown difference between the suspension model used for survey and the real suspension system also always exists. The controller is built based on an adaptive neuro-fuzzy inference system (ANFIS), sliding mode control, uncertainty observer (NFSmUoC) and a magnetorheological damper (MRD) which can be seen as an actuator for applying active force. A nonlinear uncertainty observer (NUO), a sliding mode controller (SMC) together with an inverse model of the MRD are designed in order to calculate the current value by which the MRD creates the required active control force u(t). An ANFIS and measured MR-damper-dynamic-response data sets are used to identify the MRD as an inverse MRD model (ANFIS-I-MRD). Based on dynamic response of the suspension, firstly the active control force u(t) is calculated by NUO and SMC, in which the impact of the uncertainty load on the system is estimated by the NUO. The ANFIS-I-MRD is then used to estimate applied current for the MRD in order to create the calculated active control force to control vertical vibration status of the train cars. Simulation surveys are carried out to evaluate the effectiveness of the proposed method.

Design of a new adaptive neuro-fuzzy inference system based on a solution for clustering in a data potential field

In this study, we propose a new method for building adaptive neuro-fuzzy inference systems (ANFIS) via datasets. In order to improve the performance of conventional ANFIS to handle noisy data, we focus on ameliorating the cluster-data space established from a given dataset. To achieve this, we propose a weighted clustering process in the joint input-output data space. Thus, during the clustering process, the cluster with the smallest potential distance, which is a combination of the Euclidean distance and the size of the clusters, has priority when obtaining the surveyed sample. Based on this principle, we formulate a new algorithm for synthesizing an ANFIS via the proposed data potential field, called ANFIS-PF, which has the following features: it establishes a data potential field that covers the entire initial data space, a cluster-data space is built based on the generated data potential field, and the ANFIS is synthesized using this cluster-data space. Finally, we performed experiments using datasets with and without noise to demonstrate the effectiveness of the proposed method in several applications, including dynamic-response noisy datasets obtained from a magnetorheological damper.

Establishing ANFIS and the use for predicting sliding control of active railway suspension systems subjected to uncertainties and disturbances

The effectiveness of control of the active railway suspension system (ARSS) using a magnetorheological damper (MRD) with unknown track profile and load depends deeply on (1) the control strategy and the ability to adapt to noise, (2) system’s response delay compared with the real status of track profile impacting on it, and (3) uncertainty of the model used to describe the ARSS and external disturbance. Deriving from these, in order to improve the control efficiency, in this paper, we focus on three following factors. The first is to improve the accuracy of the MRD model. The second is to establish the ability to predict the track-profile’s status to update adaptively the optimal parameters of the control system. Finally, it is to build an uncertainty and disturbance observer (DUO) to compensate for noise. A novel algorithm for fuzzy C-means clustering (FCM) in an overlapping data space deriving from the Kernel space and the data potential field named PKFCM is proposed. Based on the PKFCM, the inverse MRD model is established as well as the design of a fuzzy-based predicting sliding controller (FPSC) for the ARSS is performed which is always updated by the optimal parameters adapting to the status of track profile. The stability of the FPSC is proved theoretically while its performance is estimated by numerical surveys.

An optimal design of interval type-2 fuzzy logic system with various experiments including magnetorheological fluid damper

This paper focuses on optimal design of an interval type-2 fuzzy logic system (IT-2FLS) to cope with uncertainty issue of training set and noisy data. Content of the solution is depicted based on the proposed algorithm to optimally design an IT-2FLS from a dataset, named OD-T2FLS. The major concept of the OD-T2FLS is a combination of a useful method of clustering data space to establish a type-1 fuzzy logic system (T-1FLS) and an appropriate way to transform the T-1FLS into an IT-2FLS as well as to optimally adjust parameters of the IT-2FLS. Firstly, an improved algorithm to establish an adaptive neuro-fuzzy inference system (ANFIS), named IM-ENFS, is presented. Based on the given dataset, clustering in the join input–output data space is realized to establish a cluster-data space. Using the IM-ENFS for this cluster-data space, together with the cluster-data space optimized, an ANFIS having a role as an optimal T1-FLS is also established. Parameters of the optimized T-1FLS are then used to build the initial structure of IT-2FLS. Subsequently, this IT-2FLS is optimally adjusted based on the well-known genetic algorithm. Finally, to demonstrate the effectiveness of the proposed OD-T2FLS, experiments including magnetorheological fluid damper are realized based on two different statuses of data sources, with and without noise.

ANFIS deriving from jointed input-output data space and applying in smart-damper identification

Display Omitted Reflecting the relation between the convergent capability of ANFIS and especial features of the created cluster data space.Providing solutions for improving the convergent ability of ANFIS.Proposing an improved-configuration for ANFIS.Proposing a novel algorithm for building ANFIS.Establishing smart damper models based on ANFIS. For approximation of unknown mapping f: XY expressing a given database via an adaptive Neuro-Fuzzy inference system (ANFIS), ANFISs convergent capability is quite sensitive to the data features. In order to deal with this, this paper focus on ameliorating quality of cluster data space (CDS) used to establish the ANFIS. Firstly, we formulate and prove CDS-related necessary conditions for an approximation expressing an initial data space (IDS) convergent. Based on this theory basis, we propose a fuzzy system typed ANFIS associated with two solutions for establishing the CDS from the IDS, which focus on preventing, seeking and exterminating critical data samples in the CDS. In order to deploy these, we also present an improved structure of ANFIS. These aspects are described via a novel offline identification algorithm named ANFIS-JS for building ANFIS in a jointed input-output data space (JDS) deriving from the IDS. The results obtained via several surveys, including identifying smart dampers, magnetorheological damper (MRD) and electrorheological damper (ERD), show that the convergent stability and response accuracy are the main advantages of the ANFIS-JS.

Recurrent Mechanism and Impulse Noise Filter for Establishing ANFIS

In many real applications, building and updating adaptive neuro-fuzzy inference system (ANFIS) based on noisy measuring data sources need to be performed such that the filtering impulse noise (IN) from the initial datasets (IDSs) and establishing the ANFIS via the filtered IDS are carried out simultaneously. Focused on this purpose, in this paper, a novel recurrent mechanism as well as a solution for filtering IN based on Lyapunov stability theory is proposed to establish an adaptive online IN filter (AOINF). Using the AOINF, kernel fuzzy-C-means clustering method, and the least mean squares method, a cluster data space deriving from the filtered IDS is created to which the ANFIS is then formed. The recurrent mechanism executes filtering IN to build ANFIS and using the ANFIS as an updated-filter to filter IN synchronously until either the ANFIS converges to the desired accuracy or a stop condition is satisfied. Surveys, including identifying dynamic response of a magnetorheological damper via measuring datasets, are performed to evaluate the proposed method.

Nonlinear adaptive control based on fuzzy sliding mode technique and fuzzy-based compensator

It is difficult to efficiently control nonlinear systems in the presence of uncertainty and disturbance (UAD). One of the main reasons derives from the negative impact of the unknown features of UAD as well as the response delay of the control system on the accuracy rate in the real time of the control signal. In order to deal with this, we propose a new controller named CO-FSMC for a class of nonlinear control systems subjected to UAD, which is constituted of a fuzzy sliding mode controller (FSMC) and a fuzzy-based compensator (CO). Firstly, the FSMC and CO are designed independently, and then an adaptive fuzzy structure is discovered to combine them. Solutions for avoiding the singular cases of the fuzzy-based function approximation and reducing the calculating cost are proposed. Based on the solutions, fuzzy sliding mode technique, lumped disturbance observer and Lyapunov stability analysis, a closed-loop adaptive control law is formulated. Simulations along with a real application based on a semi-active train-car suspension are performed to fully evaluate the method. The obtained results reflected that vibration of the chassis mass is insensitive to UAD. Compared with the other fuzzy sliding mode control strategies, the CO-FSMC can provide the best control ability to reduce unwanted vibrations.

A novel minimum–maximum data-clustering algorithm for vibration control of a semi-active vehicle suspension system

In this paper, a new control system for a semi-active vehicle suspension is presented by proposing a novel minimum–maximum (min–max) pure data-clustering algorithm. The min–max pure data-clustering algorithm is used to create pure data clusters which are the basic structure for building fuzzy sets and membership functions of an adaptive neuro-fuzzy inference system. Based on the data clusters created by the min–max pure data-clustering algorithm, an appropriate structure of the adaptive neuro-fuzzy inference system is established and used to identify the dynamic behaviour of a semi-active vehicle suspension system featuring a magnetorheological-fluid-based damper. In this control system, both the measured data and an inverse dynamic model of the damper are used. To calculate the desired damping force value corresponding to the road profile at a specific time, a fuzzy inference system is built on the basis of a genetic algorithm. In this work, the fitness function of the genetic algorithm is satisfactorily considered to create the optimal fuzzy inference system structure, which expresses the ride-comfort-oriented tendency. Based on the desired force value, the desired current value is obtained by the inverse dynamic model. This is the optimal current value to be applied to the magnetorheological-fluid-based damper to reduce the acceleration of the vehicle. The effectiveness of the proposed control algorithm is demonstrated by vibration control performances such as reducing the vertical acceleration of the vehicle body and increasing the road-holding ability of the vehicle tyre. In addition, a comparison between the proposed work and the previous work is undertaken in order to show the superior vibration control performance of the proposed control strategy.

Novel fuzzy sliding controller for MRD suspensions subjected to uncertainty and disturbance

In this paper, the design of an adaptive optimal fuzzy sliding controller (AOFSC) for semi-active magnetorheological damper (MRD) vehicle suspension system subjected to uncertainty and disturbance (UAD) whose time-varying rate may be high but bounded is presented. This is a combination of an adaptive optimal fuzzy sliding mode controller (FSMCop), a nonlinear disturbance observer (NDO) and an inverse MRD model to create the desired control force, including the compensation part for UAD. An adaptive control law is proposed to establish a fuzzy-gain update law for the AOFSC and a parameter tie for the NDO such that the convergence of systems dynamic response is a Lyapunov asymptotically stable process. Surveys via simulations along with via an experimental apparatus showed that the high capability to exterminate vibration, robust stability, and the economic efficiency are the main advantages of the proposed AOFSC. Structure and operating principle of the controller AOFSC, in which (t1) and (t) respectively denote the previous and present loop.Display Omitted Designing the NDO for compensating for external disturbance.Designing the FSMCop.Establishing the parameter ties between the update laws of the FSMCop and NDO.An application of the proposed method to a real MRD suspension system.