Fuzzy Basis Function Network Research Articles

Research on Fault Diagnosis Algorithm of Power Cable Based on Deep Learning

The self-learning method is used to realize the online monitoring of optical cable after the fault occurs. According to the fuzzy characteristic of cable fault, a fuzzy basis function network is constructed which is similar to nonlinear function. Thus, the dynamic characteristics of the cable system are obtained. Then, this paper iterates each additional sample to realize the learning of the center point and radial of the network. A normal multivariable fuzzy distribution is used to determine the expected output of the new sample, and recursive operations are performed on the matrix and covariance matrix of the original sample. In this way, all failure probability distributions at the sampling points are obtained. Finally, the hardware module of the system is designed from two aspects: the distributed short-circuit fault collecting and processing equipment and the on-line temperature measuring equipment for grounding fault. In this way, the functions of software module reset, storage and communication are realized.

A Data-Driven Approach of Takagi-Sugeno Fuzzy Control of Unknown Nonlinear Systems

A novel approach to build a Takagi-Sugeno (T-S) fuzzy model of an unknown nonlinear system from experimental data is presented in the paper. The neuro-fuzzy models or, more specifically, fuzzy basis function networks (FBFNs) are trained from input–output data to approximate the nonlinear systems for which analytical mathematical models are not available. Then, the T-S fuzzy models are derived from the direct linearization of the neuro-fuzzy models. The operating points for linearization are chosen using the evolutionary strategy to minimize the global approximation error so that the T-S fuzzy models can closely approximate the original unknown nonlinear system with a reduced number of linearizations. Based on T-S fuzzy models, optimal controllers are designed and implemented for a nonlinear two-link flexible joint robot, which demonstrates the possibility of implementing the well-established model-based optimal control method onto unknown nonlinear dynamic systems.

Real-time estimation of hospital discharge using fuzzy radial basis function network and electronic health record data

Hospital resources are scarce and should be properly distributed and justified. Information about how long patients stays in critical intensive care units can provide significant benefits to hospital management resources and optimal admission planning. In this paper, we propose an approach for estimating intensive care unit length of stay using fuzzy radial basis function neural network model. The predictive performance of the model is compared to others using data collected over 13,587 admissions and 54 predictive factors from five critical units with discharges between 2001 and 2012. The proposed model compared to others demonstrated higher accuracy and better estimations. The three most influential factors in predicting length of stay at the early stage of pre-admission were demographic characteristics, admission type, and the first location within the hospital prior to critical unit admission. We have found about 63% of patients with multiple chronic conditions, stayed significantly longer in hospital. Enabling the proposed prediction model in clinical decision support system may serve as reference tools for communicating with patients and hospital managers.

Robust Wheel Wear Monitoring System for Cylindrical Traverse Grinding

Although many advanced signal processing techniques and novel machine learning algorithms have been applied to the monitoring of grinding processes in the literature, most of these techniques and algorithms are only effective under specific conditions and are unusable under other grinding conditions, such as varying wheel types or workpiece materials. This article proposes a robust grinding wheel wear monitoring system to eliminate these restrictions. Physical information generated during the grinding process is collected by a power sensor, accelerometers, and acoustic emission sensors. After the signals are preprocessed, features are extracted via different signal processing techniques, and a novel normalization scheme is applied to make these features independent of the wheel type, workpiece material, and grinding parameters. The features that are most related to wheel wear are selected according to the statistical criterion. An interval type-2 fuzzy basis function network is adopted to develop a wheel wear monitoring model, which is capable of predicting wheel wear under various grinding conditions and generating upper and lower prediction bounds according to the fluctuation of features. Based on the wheel wear model, a robust monitoring scheme to schedule timely wheel dressing and ensure workpiece surface finish could be established.

Adaptive Fuzzy Output Feedback Simultaneous Posture Stabilization and Tracking Control of Wheeled Mobile Robots With Kinematic and Dynamic Disturbances

This article proposes a simultaneous posture (i.e., position and heading direction angle) stabilization and tracking control for differential-drive wheeled mobile robots (WMRs) using an adaptive fuzzy output feedback tracking control (AFOFTC) method and a fuzzy basis function network (FBFN). Existing studies on simultaneous posture stabilization and tracking control of WMRs do not compensate for the uncertainties and disturbances caused by kinematics and dynamics (i.e., kinematic and dynamic disturbances). To inhibit the performance degradation caused by these disturbances, an FBFN-based adaptive fuzzy output feedback posture stabilization and tracking control of WMRs is proposed considering the unavailable velocities of the WMR. In this approach, i) kinematic error equations are derived in polar coordinates, ii) a nonlinear velocity observer for WMRs containing the kinematic and dynamic uncertainties is designed, iii) a backstepping-like feedback linearization controller is designed based on the kinematics in polar coordinates to generate a reference trajectory, and iv) an AFOFTC law is proposed based on the dynamics to ensure that the WMR follows the generated reference trajectory. The proposed method can maintain satisfactory posture stabilization and tracking control performance despite the kinematic and dynamic disturbances. A stability analysis as well as the simulation and experimental results are provided to validate the proposed scheme.

A multimodal intelligent monitoring system for turning processes

Online process condition monitoring is an essential component of closed-loop process-level automation of machining operations. This paper describes the development of an intelligent monitoring system for turning processes, which consists of three units: a tool wear predictor, a chatter detector and a tool chipping detector. Features are extracted from the signals of multiple low-cost and low-intrusive sensors, and then normalized using a novel scheme to eliminate their dependence on cutting conditions, workpiece materials and cutting tools. A systematic feature selection procedure, coupled with automated signal preprocessing parameter selection, is presented to select the optimal feature set for each unit. The tool wear unit is built with type-2 fuzzy basis function networks to predict tool wear with uncertainty bounds, while the chatter unit and tool chipping unit are built with support vector machines to maximize the classification fidelity. Experimental results show that the monitoring system achieved high accuracy, generalized applicability and satisfactory robustness for all the three process conditions, by using two affordable sensors: a power meter and an accelerometer. The three monitoring schemes are integrated into a monitoring software so that they can be implemented in different environments with minimal calibration efforts.

Robust Tool Wear Monitoring Using Systematic Feature Selection in Turning Processes With Consideration of Uncertainties

This paper describes a robust tool wear monitoring scheme for turning processes using low-cost sensors. A feature normalization scheme is proposed to eliminate the dependence of signal features on cutting conditions, cutting tools, and workpiece materials. In addition, a systematic feature selection procedure in conjunction with automated signal preprocessing parameter selection is presented to select the feature set that maximizes the performance of the predictive tool wear model. The tool wear model is built using a type-2 fuzzy basis function network (FBFN), which is capable of estimating the uncertainty bounds associated with tool wear measurement. Experimental results show that the tool wear model built with the selected features exhibits high accuracy, generalized applicability, and exemplary robustness: The model trained using 4140 steel turning test data could predict the tool wear for Inconel 718 turning with a root-mean-square error (RMSE) of 7.80 μm and requests tool changes with a 6% margin on average. Furthermore, the developed method was successfully applied to tool wear monitoring of Ti–6Al–4V alloy despite different mechanisms of tool wear, i.e., crater wear instead of flank wear.

Modeling and robust controlling of laser welding process on high strength titanium alloy using fuzzy basis function networks and robust Takagi-Sugeno fuzzy controller

The purpose of this paper is to present and implement a novel approach to data-driven modeling and robust control of a laser keyhole welding process. The objective is to maintain the penetration depth of the laser welding process at a desired value under system uncertainties. A framework was proposed to estimate the keyhole diameter and the keyhole penetration depth in real time by type-1 and type-2 fuzzy basis function networks, and an adaptive divided difference filter. A robust Takagi-Sugeno fuzzy controller was implemented to adjust the laser power of the welding process. Experimental results demonstrated that the fuzzy models provided an accurate estimation of both the welding geometry and its variations due to uncertainties, and the robust Takagi-Sugeno fuzzy controller successfully reduced the penetration depth variation and improved the quality of the welding process.

Modeling of unstructured uncertainties and robust controlling of nonlinear dynamic systems based on type-2 fuzzy basis function networks

This paper proposes new methods for modeling unstructured uncertainties and robust controlling of unknown nonlinear dynamic systems by using a novel robust Takagi Sugeno fuzzy controller (RTSFC). First, a new training algorithm for an interval type-2 fuzzy basis function network (FBFN) is proposed. Next, a novel technique is presented to convert the interval type-2 FBFN to an interval type-2 Takagi Sugeno (TS) fuzzy model. Based on the interval type-2 TS and type-2 FBFN models, a robust controller is presented with an adjustable convergence rate. Since the type-2 fuzzy model with its new training technique can effectively capture the unstructured uncertainties and accurately estimate the upper and lower bounds of unknown nonlinear dynamic systems, the stability condition of the proposed control system is much less conservative than other robust control methods that are based on norm bounded uncertainties. Simulation results on an electrohydraulic actuator show that the RTSFC can reduce steady state error under different conditions while maintaining better responses than the other robust sliding mode controllers.

One-step models for soft computing techniques. Industrial application to image processing in quality assurance process

The authors present an approach based on a one-step method using soft computing techniques for a quality assurance process in the form of dimensional checking parameters via industrial image processing. This method offers a high grade of precision in processes to solve the evaluation problems on-line applications. As well known to the researchers, the approaches in a single iteration of these techniques for artificial neural networks (ANN) such as adaptive neuro fuzzy inference systems (ANFIS) and radial basis function network (RBFN) are not documented in the literature. This work provides the simplification to one-step that provides the chance of creation and implementation of these models for online applications without loss of time in the iterations needed to adjust the model (training) to generate a fast response. The main objective of this paper is to provide a model capable of approximating the solution of a function that represents the system, that function is based on historical data of the process but the operators that compound the function are unknown. The relations between the inputs and outputs are known, but the interactions between variables are unknown. Based on the literature, the soft computing techniques are trained by trial and error because they do not have a stop criterion; also, the functions that provide an approximation are unknown in most cases. To solve the problem mentioned above, this paper proposes the one-step method without training. It is necessary to approximate the solution avoiding the overshoot and damping produced by classic approaches. The deviation generated is in the order of one standard deviation whose magnitude is in the order of common approaches for image processing as it is documented in literature for the best case RBFN.

Gain estimation of nonlinear dynamic systems modeled by an FBFN and the maximum output scaling factor of a self-tuning PI fuzzy controller

This paper proposes new techniques to calculate the dynamic gains of nonlinear systems represented by fuzzy basis function network (FBFN) models. The dynamic gain of an FBFN can be approximated by finding the maximum of norm values of the locally linearized systems or by solving a non-smooth optimal control problem. From the proposed gain calculation techniques, a novel adaptive multilevel fuzzy controller (AMLFC) with a maximum output scaling factor is presented. To guarantee the system stability, a stability condition is derived, which only requires that the output scaling factor of the AMLFC be bounded. Therefore, this paper provides a systematic and simple design practice for controlling nonlinear systems by using an AMLFC. The AMLFC is simulated in a tower crane control system. Simulation results show that AMLFC is not only robust but also provides improved transient performances compared with the robust adaptive fuzzy controller.

Intelligent Adaptive Motion Control Using Fuzzy Basis Function Networks for Electric Unicycle

AbstractThis paper presents two intelligent adaptive controllers, called self‐balancing and speed controllers, for self‐balancing and motion control, respectively, of an electric unicycle using fuzzy basis function networks (FBFN), which are employed to approximate model uncertainties and unknown friction between the wheel and the terrain surface. Both controllers are established based on the linearized model of the vehicle whose model uncertainties and parameter variations are caused by different riders and terrain. An adaptive backstepping controller together with online learning FBFN and sensing information of the rider's body inclination then is presented to achieve self‐balancing motion control. By adding an electronic throttle as the input device of speed commands, a decoupling sliding‐mode controller with online learning FBFN is proposed to accomplish self‐balancing and speed control. The performance and merit of the two proposed control methods are exemplified by conducting four simulations and three experiments on a laboratory‐built electric unicycle.

Intelligent Adaptive Trajectory Tracking Control for an Autonomous Small‐Scale Helicopter Using Fuzzy Basis Function Networks

AbstractThis paper presents an intelligent adaptive trajectory tracking controller using fuzzy basis function networks (FBFN) for an autonomous small‐scale helicopter. The FNFN is used online to learn the vehicle mass and the coupling effect between the force and the moments, and the intelligent adaptive controller then is synthesized systematically using a backstepping technique. This controller is designed to accomplish the ultimate boundedness of the closed‐loop helicopter dynamics and accommodate agile flight maneuvers. Two nonlinear simulations on hovering and trajectory tracking are conducted to show the effectiveness and merit of the proposed controller, which is also shown to be superior by performance comparison with a well‐known neural‐network controller.

FBFN-based adaptive repetitive control of nonlinearly parameterized systems

An adaptive repetitive control scheme is presented for a class of nonlinearly parameterized systems based on the fuzzy basis function network (FBFN). The parameters of the fuzzy rules are tuned with adaptive schemes. To attenuate chattering effectively, the discontinuous control term is approximated by an adaptive PI control structure. The bound of the discontinuous control term is assumed to be unknown and estimated by an adaptive mechanism. Based on the Lyapunov stability theory, an adaptive repetitive control law is proposed to guarantee the closed-loop stability and the tracking performance. By means of FBFNs, which avoid the nonlinear parameterization from entering into the adaptive repetitive control, the controller singularity problem is solved. The proposed approach does not require an exact structure of the system dynamics, and the proposed controller is utilized to control a model of permanent-magnet linear synchronous motor subject to significant disturbances and parameter uncertainties. The simulation results demonstrate the effectiveness of the proposed method.

Efficient environmental noise reduction system for speech enhancement via fuzzy-based extreme learning machine and fuzzy adaptive Wiener filter

This paper proposes the application of fuzzy-based extreme learning machine (FELM) and fuzzy-based adaptive Wiener filter for background noise reduction system to improve the signal to noise ratio (SNR) and to reduce the minimum mean square error (MMSE). The Wiener filter minimizes the mean square error and it provides better performance than the conventional filters. Though the background noise is uncertain, fuzzy inference systems (FIS) are proposed in ELM to classify the background noises and Wiener filters are proposed to update Wiener filter coefficients that will increase the SNR of the filtered speech signal. Fuzzy adaptive Wiener filter depends on the adaptation of the filter transfer function from sample to sample based on the signal statistics and it is implemented in time domain rather than in frequency domain to accommodate for the varying nature of the speech signal. The proposed FELM is compared with ELM and fuzzy radial basis function network (FRBFN) for noise classification and fuzzy adaptive Wiener filter is compared with Wiener and adaptive Wiener filters for noise cancellation. Simulation results show that FELM improves the percentage of classification by 9% than FRBFN, 4% than ELM and fuzzy adaptive Wiener filter improves the SNR by 8 dB than the conventional Wiener filter. The real-time implementation of the system is done using TMS320C6713 DSK starter kit. The real-time practical setup using DSK shows an improved SNR of 5 dB.

Output feedback control for free-floating space robotic manipulators base on adaptive fuzzy neural network

Free-floating space robotic manipulators with uncertainty of model and external disturbances are considered by the paper. Joints motors of space manipulators usually work at low-velocity operating conditions, tachometers are difficult to satisfy signal request. The paper brings forward a model-free fuzzy basis function network output feedback control strategy of space robot base on velocity reconstructed. Firstly, space robot dynamic model is established, and then velocity signals are reconstructed by observer, the fuzzy basis function network is used to approximate nonlinear function of observer and controller and compensated the effect of system deviations to the velocity reconstruction. The network weights and the parameters can be tuned online by hybrid back-propagation algorithm and e-correction algorithm, with no off-line learning phase required. The proposed observer and controller do not require any space robot dynamic model (including the inertia matrix inverse). Based on Lyapunov theory, the paper proves that the whole closed-loop system is uniformly ultimately bounded (UUB). Comparison results further show the proposed control method of superiority comparison. Simulation results verify that the observer-based output feedback control strategy can achieve higher precision, and has important engineering value to uncertainty space robots without velocity information.

Intelligent Motion Compensation for Improving the Tracking Performance of Shipborne Phased Array Radar

The shipborne phased array radar must be able to compensate the ship’s motion and track the maneuvering targets automatically. In this paper, the real-time beam pointing error compensation mechanism of a planar array antenna for ship’s motion is designed to combine with the Kalman filtering. The effect of beam pointing error on the tracking performance of shipborne phased array radar is examined. A compensation mechanism, which can automatically correct the beam pointing error of the planar antenna array, is proposed for shipborne phased array radar in order to achieve the required tracking accuracy over the long dwell time. The automatic beam pointing error compensation mechanism employs the parallel fuzzy basis function network (FBFN) architecture to estimate the beam pointing error caused by roll and pitch of the ship. In the simulation, the models of roll and pitch are used to evaluate the performance of beam pointing error estimation mechanism based on the proposed parallel FBFN architecture. In addition, the effect of automatic beam pointing error compensation mechanism on the tracking performance of adaptive extended Kalman filter (AEKF) implemented in ship borne phased array radar is also investigated. Simulations find out that the proposed algorithms are stable and accurate.

Strategies for intelligent drill wear prediction using multiple sensor signals

In this work, strategies for artificial intelligence (AI)-based drill flank wear prediction systems have been presented. Four AI-based models, namely, back propagation neural network (BPNN), radial basis function network (RBF), normalised radial basis function network (NRBF) and fuzzy radial basis function network (FRBF) have been used. Signals have been acquired from dynamometer, current sensor and accelerometer during drilling under different cutting conditions. Wear sensitive features in time-domain and frequency domain have been extracted. Different strategies, i.e., various combinations of the selected features and process parameters used as inputs to the AI models, are formulated. For most of the strategies, BPNN model gives better wear prediction results followed by NRBF model. But for noisy features such as features from vibration signal can be better related to drill wear using FRBF model.

불확실성을 가지는 전기 구동 논홀로노믹 이동 로봇의 궤적 추종을 위한 강인 적응 퍼지 백스테핑 제어

This paper proposes a robust adaptive fuzzy backstepping control scheme for trajectory tracking of an electrically driven nonholonomic mobile robot with uncertainties and actuator dynamics. A complete model of an electrically driven nonholonomic mobile robot described in this work includes all models of the uncertain robot kinematics with a nonholonomic constraint, the uncertain robot body dynamics with uncertain frictions and unmodeled disturbances, and the uncertain actuator dynamics with disturbances. The proposed control scheme uses the backstepping control approach through a kinematic controller and a robust adaptive fuzzy velocity tracking controller. The presented control scheme has a voltage control input with an auxiliary current control input rather than a torque control input. It has two FBFNs(Fuzzy Basis Function Networks) to approximate two unknown nonlinear robot dynamic functions and a robust adaptive control input with the proposed adaptive laws to overcome the uncertainties such as parameter uncertainties and external disturbances. The proposed control scheme does not a priori require the accurate knowledge of all parameters in the robot kinematics, robot dynamics and actuator dynamics. It can also alleviate the chattering of the control input. Using the Lyapunov stability theory, the stability of the closed-loop robot control system is guaranteed. Simulation results show the validity and robustness of the proposed control scheme.

Adaptive Repetitive Control of Nonlinearly Parameterized Systems Based on Fuzzy Basis Function Networks

In this note, an adaptive repetitive control scheme is presented for a class of nonlinearly parameterized systems based on fuzzy basis function networks (FBFNs). The parameters of the fuzzy rules are tuned with adaptive schemes. The adaptive repetitive control law is derived by using Lyapunov synthesis method to guarantee the closed-loop stability and the tracking performance. By means of FBFNs, the controller singularity problem is solved as it avoids the nonlinear parameterization from entering into the adaptive control and repetitive control. The proposed approach has the added advantage that it does not require an exact structure of the system dynamics. The proposed controller is applied to control a model of permanent-magnet linear synchronous motor (PMLSM) subject to significant disturbances and parameter uncertainties. The simulation results demonstrate the effectiveness of the proposed method in terms of significant reduction in chattering while maintaining asymptotic convergence.