Recurrent Radial Basis Function Research Articles

Neurocomputing in Civil Infrastructure

This article presents a review of the recent applications of artificial neural networks (ANN) for civil infrastructure including structural system identification, structural health monitoring, structural vibration control, structural design and optimization, prediction applications, construction engineering, and geotechnical engineering. The most common ANN used in structural engineering is the backpropagation neural network followed by recurrent neural networks and radial basis function neural networks. In recent years, newer hybrid techniques have been used in structural engineering by a number of researchers such as the neuro-fuzzy inference system, time-delayed neuro-fuzzy inference system, and wavelet neural networks. Deep machine learning techniques are among the newest techniques to find applications in civil infrastructure systems.

Nonlinear system identification based on a self-organizing type-2 fuzzy RBFN

This paper presents a new self-evolving recurrent Type-2 Fuzzy Radial Basis Function Network (T2FRBFN) in which the weights are considered Gaussian type-2 fuzzy sets and uncertain mean in each RBF neuron. The capability of the proposed T2FRBFN for function approximation and dynamical system identification perform better than the conventional RBFN. A novel type-2 fuzzy clustering is presented to add or remove the hidden RBF neurons. For parameter learning, back-propagation with adaptive learning rate is used. Finally the proposed T2FRBFN is applied to identification of three nonlinear systems as case studies. A comparison between T2FRBFN and the conventional RBFN as well as the method of Rubio-Solis and Panoutsos (2015) is presented. Simulation results and their statistical description show that the proposed T2FRBFN perform better than the conventional RBFN.

Nonlinear Model Predictive Control Based on a Self-Organizing Recurrent Neural Network.

A nonlinear model predictive control (NMPC) scheme is developed in this paper based on a self-organizing recurrent radial basis function (SR-RBF) neural network, whose structure and parameters are adjusted concurrently in the training process. The proposed SR-RBF neural network is represented in a general nonlinear form for predicting the future dynamic behaviors of nonlinear systems. To improve the modeling accuracy, a spiking-based growing and pruning algorithm and an adaptive learning algorithm are developed to tune the structure and parameters of the SR-RBF neural network, respectively. Meanwhile, for the control problem, an improved gradient method is utilized for the solution of the optimization problem in NMPC. The stability of the resulting control system is proved based on the Lyapunov stability theory. Finally, the proposed SR-RBF neural network-based NMPC (SR-RBF-NMPC) is used to control the dissolved oxygen (DO) concentration in a wastewater treatment process (WWTP). Comparisons with other existing methods demonstrate that the SR-RBF-NMPC can achieve a considerably better model fitting for WWTP and a better control performance for DO concentration.

Long-range prediction of retail sales using recurrent radial basis function neural network

The literature survey on sales forecasting reveals that few works have been reported on long-range forecasting of sale volumes. On the other hand, there is a need of such long-range forecasting of sales data to devise suitable organisational strategy. The existing soft computing-based forecasting models provide poor prediction performance. Keeping this in view a new soft computing model is developed and utilised for prediction of seasonally adjusted (SA) and non-seasonally adjusted (NSA) sales volumes up to 24 months. The simulation results of real-life data show an excellent prediction performance compared to that of four other contemporary soft computing models.

The Proposal of Two New Recurrent Radial Basis Function Neural Networks

Two types of new recurrent RBF neural networks are introduced here and are applied on four test problems that are used for identification. The proposed recurrent RBF neural networks use both the power of the recurrent neural networks together with the abilities of the RBF neural networks so it can achieve good results.

Adaptive hybrid control system using a recurrent RBFN-based self-evolving fuzzy-neural-network for PMSM servo drives

In this paper, an adaptive hybrid control system (AHCS) based on the computed torque control for permanent-magnet synchronous motor (PMSM) servo drive is proposed. The proposed AHCS incorporating an auxiliary controller based on the sliding-mode, a recurrent radial basis function network (RBFN)-based self-evolving fuzzy-neural-network (RRSEFNN) controller and a robust controller. The RRSEFNN combines the merits of the self-evolving fuzzy-neural-network, recurrent-neural-network and RBFN. Moreover, it performs the structure and parameter-learning concurrently. Furthermore, to relax the requirement of the lumped uncertainty, an adaptive RRSEFNN uncertainty estimator is used to adaptively estimate the non-linear uncertainties online, yielding a controller that tolerate a wider range of uncertainties. Additionally, a robust controller is proposed to confront the uncertainties including approximation error, optimal parameter vector and higher order term in Taylor series. The online adaptive control laws are derived based on the Lyapunov stability analysis, so that the stability of the AHCS can be guaranteed. A computer simulation and an experimental system are developed to validate the effectiveness of the proposed AHCS. All control algorithms are implemented in a TMS320C31 DSP-based control computer. The simulation and experimental results confirm that the AHCS grants robust performance and precise dynamic response regardless of load disturbances and PMSM uncertainties.

Adaptive Dynamic Sliding-Mode Control System Using Recurrent RBFN for High-Performance Induction Motor Servo Drive

In this paper, an adaptive dynamic sliding-mode control system (ADSMCS) with recurrent radial basis function network (RRBFN) for indirect field-orientation control induction motor (IM) drive is proposed. The ADSMCS comprises a dynamic sliding-mode controller (DSMC), an RRBFN uncertainty observer and a robust controller. The DSMC is proposed to reduce the chattering phenomenon. However, due to the uncertainty bound being unknown of the switching function for the DSMC, an ADSMCS is proposed to increase the robustness and improve the control performance of IM drive. In the ADSMCS, an RRBFN uncertainty observer is used to estimate an unknown nonlinear time-varying function of lumped parameter uncertainty online. Moreover, the adaptive learning algorithms for the RRBFN are derived using the Lyapunov stability theorem to train the parameters of the RRBFN online. Furthermore, a robust controller is proposed to confront the uncertainties including approximation error, optimal parameter vector and higher order term in Taylor series. A computer simulation and an experimental system are developed to validate the effectiveness of the proposed ADSMCS. All control algorithms are implemented in a TMS320C31 DSP-based control computer. The simulation and experimental results confirm that the ADSMCS grants robust performance and precise response regardless of load disturbances and IM uncertainties.

Improved Un-Gauged Streamflow Prediction Using Three Artificial Neural Networks Methods for Leeser Zab

Artificial neural networks have flexible structures that allow multi-input and multi-output modeling's. This is particularly important in streamflow forecasting where inflows at multiple locations are considered. Three different artificial neural networks methods which are feed forward neural networks FFNN, layered recurrent neural network LRNN and radial basis function networks RBFNN were investigated in this research to predict the un gauged Leeser Zab data north Iraq at two locations. Comparisons were made between the performance of the three methods to cope the missing data at two gauging stations for this stream. In order to accomplish this task , historical inflow series is employed for training, validation and test with different proportions, many structures of ANNs were used by selecting different combinations of inputs and using different number of neurons at the hidden layer. Also a different selections for the outputs by using single and multiple outputs were tried. The ANN prediction models has been found to be workable tool for prediction Leeser Zab stream flow correctly but the layer recurrent neural network kind and radial basis function neural networks were most suitable to reflect the behavior of the Leeser Zab flow especially the RBFNN type since the predicted flow results using this method were quite cohesive to the exact data flows. It was concluded from the experiment that lower proportion of training, validation and test ratio gives better result with higher accuracy

Neural network-based adaptive multiuser detection schemes in SDMA–OFDM system for wireless application

Neural network applications in adaptive multiuser detection (MUD) schemes are suggested here in the context of space division multiple access–orthogonal frequency division multiplexing system. In this paper, various neural network (NN) models like feed forward network (FFN), recurrent neural network (RNN) and radial basis function network (RBFN) are adopted for MUD. MUD using NN models outperforms other existing schemes like genetic algorithm--assisted minimum bit error rate (MBER) and minimum mean square error MUDs in terms of BER performance and convergence speed. Among these NN models, the FNN MUD performs efficiently as RNN in full load scenario, where the number of users is equal to number of receiving antennas. In overload scenario, where the number of users is more than the number of receiving antennas, the FNN MUD performs better than RNN MUD. Further, the RBFN MUD provides a significant enhancement in performance over FNN and RNN MUDs under both overload and full load scenarios because of its better classification ability due to Gaussian nonlinearity. Extensive simulation analysis considering Stanford University Interim channel models applied for fixed wireless applications shows improvement in convergence speed and BER performance of the proposed NN-based MUD algorithms.

A NEW RBF NEURAL NETWORK FOR PREDICTION IN INDUSTRIAL CONTROL

A novel neural architecture for prediction in industrial control: the 'Double Recurrent Radial Basis Function network' (R2RBF) is introduced for dynamic monitoring and prognosis of industrial processes. Three applications of the R2RBF network on the prediction values confirmed that the proposed architecture minimizes the prediction error. The proposed R2RBF is excited by the recurrence of the output looped neurons on the input layer which produces a dynamic memory on both the input and output layers. Given the learning complexity of neural networks with the use of the back-propagation training method, a simple architecture is proposed consisting of two simple Recurrent Radial Basis Function networks (RRBF). Each RRBF only has the input layer with looped neurons using the sigmoid activation function. The output of the first RRBF also presents an additional input for the second RRBF. An unsupervised learning algorithm is proposed to determine the parameters of the Radial Basis Function (RBF) nodes. The K-means unsupervised learning algorithm used for the hidden layer is enhanced by the initialization of these input parameters by the output parameters of the RCE algorithm.

Dynamic behavioral modeling for strongly nonlinear doherty pas using real-valued time-delay recurrent RBF model

This paper proposes a Real-Valued Time-Delay Recurrent Radial Basis Function (RVTDRRBF) model suitable for dynamic modeling of the strongly nonlinear behaviors of the Doherty Power Amplifiers (DPAs). This model has four Tapped Delay Lines (TDLs), which account for the memory effect of the DPA. The structure of the RVTDRRBF model is simpler than the traditional FeedForward Neural Networks (FFNNs) model. Weights and centers of the proposed model can be resolved by the Orthogonal Least Square (OLS) and Singular Value De-composition (SVD) algorithm. A three-carrier Wideband Code Division Multiple Access (WCDMA) signal is taken as the test signal. The simulation results in frequency-domain and time-domain for a DPA with 51 dBm output illustrate a good agreement between the RVTDRRBF model and measurement data. Moreover, comparing the Normalized Mean Square Error (NMSE) of RVTDRRBF model, memory polynomial model and RVTDRBF model, it can be noticed that the proposed RVTDRRBF model is more accurate than the RVTDRBF model and the memory polynomial model in modeling the strong dynamic nonlinearity of the DPAs.

Autonomous and adaptive procedure for cumulative failure prediction

An autonomous adaptive reliability prediction model using evolutionary connectionist approach based on Recurrent Radial Basis Function architecture is proposed. Based on the currently available failure time data, Fuzzy Min–Max algorithm is used to globally optimize the number of the k Gaussian nodes. This technique allows determining and initializing the k-centers of the neural network architecture in an iterative way. The user does not have to define arbitrary some parameters. The optimized neural network architecture is then iteratively and dynamically reconfigured as new failure occurs. The performance of the proposed approach has been tested using sixteen real-time software failure data.

Static and Dynamic Neural Network Approach for Short Term Flood Forecasting A Comparative Study

artificial neural networks (ANNs) have been applied to various hydrologic problems recently. This research demonstrates static and dynamic neural approach by applying Time lagged recurrent neural network and Radial basis function neural network to rainfall-runoff modeling for the upper area of Wardha River in India. The model is developed by processing online data over time using static and dynamic connections. Methodologies and techniques of the two models are presented in this paper and a comparison of the short term runoff prediction results between them is also conducted. The prediction results of the Time lagged recurrent neural network indicate a satisfactory performance in the three hours ahead of time prediction. The conclusions also indicate that Time lagged recurrent neural network is more versatile than Radial basis function neural network and can be considered as an alternate and practical tool for predicting short term flood flow. General Terms:- Prediction, Radial basis function neural network, Time lagged recurrent neural network.

Modeling piezoelectric actuators with Hysteretic Recurrent Neural Networks

This paper describes the application of Hysteretic Recurrent Neural Networks (HRNNs) to the modeling of polycrystalline piezoelectric actuators. Because piezoelectric materials exhibit voltage/strain relationships that are hysteretic and rate-dependent, the HRNN is composed of neurons with activation functions that incorporate these characteristics. Individual neurons are shown to agree with existing models of ideal single-crystal piezoelectric behavior. The combination of many such neurons into a network allows prediction of the heterogeneous behavior of polycrystalline materials. This model is shown to approximate the strain and polarization of an unloaded commercial stack actuator at multiple loading rates. A comparison is made to a recurrent Radial Basis Function Network model, and the HRNN is demonstrated to more accurately generalize across data sets. The model is further shown to execute on a PC platform at rates over 100 Hz, fast enough to support its application to real-time control.

WITHDRAWN: Recurrent NARX time series model prediction: A feedforward and recurrent radial basis function network

This article has been withdrawn at the request of the author(s) and/or editor. The Publisher apologizes for any inconvenience this may cause. The full Elsevier Policy on Article Withdrawal can be found at http://www.elsevier.com/locate/withdrawalpolicy.

Towards Accurate and Reproducible Predictions for Prognostic: an Approach Combining a RRBF Network and an AutoRegressive Model

In prognostic's field, the lack of knowledge on the behavior of equipments can impede the development of classical dependability analysis, or the building of effective physic-based models. Following that, artificial neural networks (ANNs) appear to be well suited since they can learn from data gathered from equipments. In this paper, an approach combining a Recurrent Radial Basis Function network (RRBF) and an AutoRegressive with eXogenous inputs model (ARX) is proposed in order to perform the prediction step of prognostics: the ARX attempts to correct the error of predictions of the RRBF. Moreover, since performances of an ANN can be closely related to initial parameterization of the network, a criterion is defined to quantify the reproducibility of predictions and thereby a priori estimate the usefulness of neural network structure. The whole aims at improving the prediction step of prognostics, which is critical with respects to real applicative conditions.

Réseau de neurones doublement récurrent à base de fonctions radiales : RR2FR Application au pronostic prévisionnel

This paper introduces a new neural architecture for prediction: Double Recurrent Radial Basis Function network (R 2 RBF). The proposed R 2 RFR is excited by the recurrence of output looped neurons on the input layer. An unsupervised learning algorithm is proposed to determine the parameters of the Radial Basis Function (RBF) nodes. An application of the R 2 RBF network on the values prediction of the time series confirmed that the proposed architecture minimizes the prediction error.

Radial Basis Function Nets for Time Series Prediction

This paper introduces a novel ensemble learning approach based on recurrent radial basis function networks (RRBFN) for time series prediction with the aim of increasing the prediction accuracy. Standing for the base learner in this ensemble, the adaptive recurrent network proposed is based on the nonlinear autoregressive with exogenous input model (NARX) and works according to a multi-step (MS) prediction regime. The ensemble learning technique combines various MSNARX-based RRBFNs which differ in the set of controlling parameters. The evaluation of the approach includes a discussion on the performance of the individual predictors and their combination

Nonlinear Fading Channel Equalization of BPSK Signals Using Multiplicative Neuron Model

A high order feed forward neural network architecture with optimum number of nodes is used for adaptive channel equalization in this paper.The replacement of summation at each node by multiplicatio n results in more powerful mapping because of its capability of processing higher-order information from training data. The equalizer is tested on Rayleigh fading channel with BPSK signals. Performance comparison with recurrent radial basis function (RRBF) neural network show that the proposed equalizer provides compact architecture and satisfactory results in terms of bit error rate performance at various levels of signal to noise ratios for a Rayleigh fading channel.

Neural Network Based Traffic Prediction for Wireless Data Networks

In a wireless network environment accurate and timely estimation or prediction of network traffic has gained much importance in the recent past. The network applications use traffic prediction results to maintain its performance by adopting its behaviors. Network Service provider will use the prediction values in ensuring the better Quality of Service(QoS) to the network users by admission control and load balancing by inter or intra network handovers. This paper presents modeling and prediction of wireless network traffic. Here traffic is modeled as nonlinear and non-stationary time series. The nonlinear and non-stationary time series traffic is predicted using neural network and statistical methods. The results of both the methods are compared on different time scales or time granularity. The Neural Network (NN) architectures used in this study are Recurrent Radial Basis Function Network (RRBFN) and Echo state network (ESN).The statistical model used here in this work is Fractional Auto Regressive Integrated Moving Average (FARIMA) model. The traffic prediction accuracy of neural network and statistical models are in the range of 96.4% to 98.3% and 78.5% to 80.2% respectively.