Search

This article deals with the development of four modified radial basis function neural network (RBFNN) models. The corresponding learning algorithms associated with the updating of internal parameters of the models are derived. The conventional inputs are used in the first and second modified RBFNN models (models 3 and 4) whereas exponential nonlinear inputs are used in the fifth and sixth RBFNN models to provide additional nonlinearity for achieving a better solution of nonlinear classification, and direct and inverse modeling problems. To assess and compare the performance potentiality of the proposed four new RBFNN models, one classification problem, one direct modeling problem, and one inverse modeling problem are solved through computer simulation-based experiments. For comparison and to assign the performance rank of each of the four modified RBFNN models, two conventional and commonly used RBFNN models (models 1 and 2) are also simulated. To access the performance of different models during the training phase of Examples 1 and 2, the root mean-square error (RMSE) value, mean absolute deviation (MAD), and the number of iterations required to achieve convergence are obtained. For the third example, only the first two performance measures are found. During the testing or validation phase, the output responses of the different models of Example 2 are compared with the desired response analysis. For Example 3, the bit-error rate (BER) plots are compared. The observation of all the results demonstrates consistent ranks of all models in the case of all three examples. It is, in general, found that the ranks of the models 1–6 are 6, 4, 3, 2, 5, and 1, respectively. In essence, in terms of all performance measures, model M-6 with an exponential version of inputs with weights on both layers occupies the first position whereas model M-4 with conventional inputs as the second position. [ABSTRACT FROM AUTHOR]

Eutrophication is the leading cause of freshwater impairment. Of China’s major cities, Wuhan has the largest coverage of the lake area, while lake eutrophication is increasing. Thus, the objective of this study was to monitor and assess lake eutrophication in Wuhan. For this purpose, the lake eutrophication was quantified using the Trophic Level Index (TLI). The TLI was modeled by building a radial basis function neural network (RBFNN) model with a set of indices derived from Sentinel-2 Multispectral Instrument (MSI) imagery and a parameter of lake environmental characteristics. Applied MSI indices are the Maximum Chlorophyll Index (MCI) and the spectral band ratio of b5/b4 and b3/b4. In addition, the applied lake environmental parameter is the chlorophyll-a dependent month. Based on 54 MSI images from April 2018 to March 2020, the lake eutrophication was successfully determined. The main findings are as follows: 1) The higher percentage of eutrophic area was registered in the autumn (from September to November) and the highest was registered on September 12th, 2019; 2) Compared to the same period in 2018, the eutrophic area in the autumn of 2019 was significantly higher. In the autumn of 2018, from 74.94% to 86.38% of the lakes were eutrophic, while from 83.54% to 93.24% of the lakes were eutrophic in the autumn of 2019; 3) The TLI values were between 30 and 70 for the majority of lakes in Wuhan, suggesting that most lakes were mesotrophic, light eutrophic, or medium eutrophic, with few oligotrophic and heavy eutrophic lakes. Most oligotrophic and mesotrophic lakes were located in northern Wuhan, while eutrophic lakes were located in central and southern Wuhan. Our result indicates that the developed algorithm from this study could be applied for rapid and effective monitoring of eutrophication in lakes.

In this paper, optimized radial basis function neural networks (RBFNNs) are employed to construct a sliding mode control (SMC) strategy for quadrotors with unknown disturbances. At first, the dynamics model of the controlled quadrotor is built, where some unknown external disturbances are considered explicitly. Then SMC is carried out for the position and the attitude control of the quadrotor. However, there are unknown disturbances in the obtained controllers, so RBFNNs are employed to approximate the unknown parts of the controllers. Furtherly, Particle Swarm optimization algorithm (PSO) based on minimizing the absolute approximation errors is used to improve the performance of the controllers. Besides, the convergence of the state tracking errors of the quadrotor is proved. In order to exposit the superiority of the proposed control strategy, some comparisons are made between the RBFNN based SMC with and without PSO. The results show that the strategy with PSO achieves quicker and smoother trajectory tracking, which verifies the effectiveness of the proposed control strategy.

• This work first time deploys the radial basis function neural network (RBFNN)-based control with norm estimation in a new way to facilitate the control development. • Inequality constraint of using norm estimation is relaxed but still retaining the same properties of the original system. • A nonlinear disturbance state observer (NDSO)-based RBFNN is employed to not only estimate unmeasured states but also lumped disturbances and uncertainties. • An observer-based NN finite time control is established to achieve the system output tracking regulation. Due to the difficulty encountered in dealing with unstructured system dynamics with unmeasured system state variables, this paper presents a novel observer-based neural network finite-time output control strategy for general high-order nonlinear systems (HNSs). The suggested technique is performed based on the backstepping-like control (BSC) scheme with a hybrid nonlinear disturbance-state observer and norm estimation-based radial basis function neural network (RBFNN). This helps not only reduce the number of estimated parameters but also relax the restriction of using inequality when exploiting the norm estimation concept in a conventional way; thus, retaining the same properties of the original system. Therefore, an observer-based finite-time output feedback control is established to deal with the unstructured dynamical behaviors and satisfying the output tracking regulation with the semi-global practically finite-time stability (SGPFS) guaranteed for the closed-loop system. The effectiveness and workability of the proposed algorithm is verified by a numerical simulation on a specific practical application. [ABSTRACT FROM AUTHOR]

Direct Torque Control (DTC) is widely used in motion control of motors. Classical DTC utilizes the Park transformation(dq-model) or observer to estimate the stator fluxlinkage and electromagnetic torque. However, the dq-model and observer lack descriptions of the cogging torque and magnetic saturation in permanent magnet synchronous motors (PMSM). This paper proposes a novel method based on the surrogate model to estimate stator fluxlinkage and torque for PMSM DTC. First, a finite element model (FEM) was established based on the parameters of PMSM. Using the FEM and Latin hypercube sampling (LHS), a surrogate model for estimating the stator fluxlinkage and torque of PMSM was constructed. The stator fluxlinkage and torque surrogate model, active flux observer, and dq current model under different control methods are constructed, respectively. The servo control parameters were tuned using the extended symmetric optimum algorithm, and simulation and experiments were conducted. Finally, electromechanical coupling models with a ball screw feed system were built based on various control and feedback methods. Simulation and experiment results indicate that the surrogate model reduces torque ripple, stator fluxlinkage offset, and feed system tracking error. [ABSTRACT FROM AUTHOR]

Space robots are widely used for on-orbit capturing. After capturing a rotational non-cooperative target, its momentums produce unpredictable movements for the resultant space robot-target compound system. In order to address this problem, a collision-free detumbling strategy (CFDS) is proposed for space robots with a single 7-degree-of-freedom (DOF) manipulator. First, an acceleration potential field (AccPF) is presented for planning detumbling velocity trajectories, which guide the compound system to terminal rest while guaranteeing collision avoidance in dynamic environments. Energy conservation and acceleration limitations are considered in the trajectory generation. Afterwards, with a radial basis function neural network (RBFNN) approximating the target's uncertain dynamics, an adaptive control scheme is established to precisely track the planned detumbling velocity trajectory. Furthermore, system stability is certified by a Lyapunov function. The effectiveness of the CFDS is eventually validated by simulations. [ABSTRACT FROM AUTHOR]

In this study, an improved Adaptive Fast Terminal Synergetic Controller (AFTSC) using Dual Radial Basis Function (RBF) Neural Networks (NNs) for output voltage control of an uncertain DC/DC step-down converters is proposed. Using the considered AFTSC, the with new manifold proposed here enables the DC/DC step-down converter's state variables to track the preferred reference voltage in presence of disturbances from any initial condition with proper precision and limited time. To rendering the design more robust, a sort of dual RBFNNs are utilized to approximate in real-time unknown converter non-linear dynamics and reduce the modeling error without calling upon usual model linearization and simplifications. The stability of the closed-loop system is assured by means of the Lyapunov method. Considering the PWM DC–DC step-down converter as an example, the considered adaptive RBFNN-FTSC law is studied in detail and implemented on a dSPACE ds1103 card. All the simulation and experimental results illustrate the efficiency and feasibility of the suggested controller. [ABSTRACT FROM AUTHOR]

This paper considers an adaptive neural event-triggered control problem of the molten steel level for twin roll strip casting systems with the inclined angle. Firstly, the model is improved and simplified into an affine nonlinear system by exploiting input compensation technique. Then, an adaptive observer is established based on radial basis function neural network (RBFNN). Furthermore, an adaptive event-triggered control scheme and corresponding adaptive updated laws are designed. It is proved that the proposed control method can guarantee the system output can follow reference signal and all closed-loop signals are bounded. Finally, the validity of the control scheme is verified through semi-experimental system dynamic model.

Aiming at the attitude control problem of fixed wing UAV, this paper introduces S-plane control, which has good control effect in the field of underwater UAV, into the attitude control of UAV. At the same time, aiming at the problem that the coefficient setting of parameters in S-plane control completely depends on experience and cannot be adjusted adaptively, the radial basis function neural network (RBFNN) is introduced, and a neural network S-plane control model which can realize on-line adaptive adjustment of the coefficient of parameters in S-plane control is proposed. The simulation results based on the data of a certain UAV show that compared with the S-plane control, the proposed neural network S-plane control model has the characteristics of fast response speed, strong anti-interference ability, and strong robustness. In addition, it also has the function of adaptive adjustment, which shows good control performance.

In this article, a cooperative adaptive iterative learning fault-tolerant control (CAILFTC) algorithm with the radial basis function neural network (RBFNN) is proposed for multiple subway trains subject to the time-iteration-dependent actuator faults by using the multiple-point-mass dynamics model. First, an RBFNN is utilized to cope with the unknown nonlinearity of the subway train system. Next, a composite energy function (CEF) technique is applied to obtain the convergence property of the presented CAILFTC, which can guarantee that all train speed tracking errors are asymptotic convergence along the iteration axis; meanwhile, the headway distances of neighboring subway trains are kept in a safety range. Finally, the effectiveness of theoretical studies is verified through a subway train simulation. [ABSTRACT FROM AUTHOR]

This study evaluated different methods for geoid undulation prediction and included two types of artificial neural networks (ANNs) - the radial basis function neural network (RBFNN) and the generalized regression neural network (GRNN) - as well as conventional methods including multiple linear regression (MLR) and ten different interpolation techniques. In this work, k-fold cross-validation was used to evaluate the model and its behavior on the independent dataset. With this validation method, each of a k number of groups has the chance to be divided into training and testing data. The performances of the methods were evaluated in terms of the root mean square error (RMSE), mean absolute error (MAE), Nash-Sutcliffe efficiency coefficient (NSE), correlation coefficient (R²), and using graphical indicators. The evaluation of the performance of the datasets obtained using cross-validation was performed in two ways. When the method having the minimum error result was accepted as the most appropriate method, the natural neighbor (NN) gave better results than the other methods (RMSE = 0.142 m, MAE = 0.097 m, NSE = 0.98986, and R² = 0.99011). On the other hand, it was observed that on average, the GRNN exhibited the best performance (RMSE = 0.185 m, MAE = 0.137 m, NSE = 0.98229, and R² = 0.98249). [ABSTRACT FROM AUTHOR]

Significant anomalous geoelectric potential variations have been observed in a section of the NEUBA II gas pipeline along its route in the district of Saavedra, near the area of Goyena, province of Buenos Aires (Argentine), where it goes through major lithological, edaphological and hydrological variations. Detailed research was conducted, showing that these disturbances may be intensified with variations of the Earth's magnetic field, during a magnetic storm, as the pipe-to-soil potential (PSP) values remained constant for weeks and then fluctuations from 0.1 V to 0.15 V were recorded in various parts of the pipeline. On the other hand, to provide another analysis of these variations, models based on the distributed source transmission line (DSTL) theory were used, proposing a uniform geoelectric field along the pipeline route. A design was proposed that would allow modeling the response of the pipeline to variations of induced geoelectric fields, taking into account their intensification based on points of discontinuity due to subsoil characteristics or differences in its structure. Good consistency was achieved between the observed and modeled PSPs. The analysis and monitoring of these PSPs is a useful tool to identify the potential risks caused by geomagnetically induced currents in the pipes that would increase the effects due to the structure or the environment in which it is buried. [ABSTRACT FROM AUTHOR]

To overcome the uncertainties of the nonlinear model of a morphing aircraft, this paper presents a high-precision adaptive back-stepping control method based on the radial basis function neural network (RBFNN). Firstly, based on the analysis of static and dynamic aerodynamic parameters of the morphing aircraft, its nonlinear control law is designed by using the conventional back-stepping method. The RBFNN is introduced to approximate online the uncertain terms of the nonlinear control law so as to improve its robustness. The robust term is designed to eliminate the approximation error caused by the RBFNN. Secondly, the tracking differentiator is designed through solving the virtual control variables, thus solving the "differential expansion" problem existing in the traditional back-stepping method. The Lyapunov stability analysis proves that our method can ensure that the tracking error of a closed-loop system converges finally and that its signals are uniformly bounded. Finally, the digital simulation model of the morphing aircraft is established with the MATLAB/Simulink; our method is compared with the conventional back-stepping control method. The simulation results show that our method has a higher control precision and stronger robustness.

Box transformer substation (BTS) is an important power distribution environment. To ensure the safe and stable operation of the power distribution system, it is critical to monitor the BTS operation and diagnose its faults in a reliable manner. In the Internet of Things (IoT) environment, this paper aims to develop a real-time and accurate online strategy for BTS monitoring and fault diagnosis. The framework of our strategy was constructed based on the IoT technique, including a sensing layer, a network layer and an application layer. On this basis, a BTS fault diagnosis method was established with variable precision rough set (VPRS) as the pre-network and the radial basis function neural network (RBFNN) as the back-fed network. The VPRS and the RBFNN were selected, because the BTS faults have many characteristic parameters, with complex nonlinear relationship with fault modes. Finally, a prototype of our strategy was developed and applied to the fault diagnosis of an actual BTS. The results fully demonstrate the effectiveness and feasibility of our strategy.

Phase-only nulling with low sidelobe level is a problem of interest in array synthesis which is a tedious problem without an analytical solution. In this communication, a novel framework for the phase-only nulling based on radial basis function neural network (RBFNN) is proposed to predict the phase adjustment for the array pattern nulling with sidelobe control. In the process of network training, the parameters of the RBFNN are optimized simultaneously based on the self-adaptive differential evolution (SADE) algorithm, which aims to improve the approximation ability and reduce the complexity of the network. Simulation results of the optimized RBFNN models show compact network structure and form the array pattern under desired performance with good generalization capability. [ABSTRACT FROM AUTHOR]

In this article, a radial basis function neural network-based adaptive iterative learning fault-tolerant control (RBFNN-AILFTC) algorithm is developed for subway trains subject to the time-iteration-dependent actuator faults and speed constraint by using the multiple-point-mass dynamic model. First, the RBFNN is utilized to approximate the time-iteration-dependent unknown nonlinearity of the subway train system; then, the iterative learning mechanism is used to tackle the outstanding repetitive operational pattern of a subway train which runs from one station to the next strictly according to the operation timetable schedule every day within the finite time interval, and the adaptive mechanism is designed for dealing with the time and the iteration-varying factors of the subway train. Second, a barrier composite energy function technique is exploited to obtain the convergence property of the proposed RBFNN-AILFTC scheme for subway train system, which can guarantee that the tracking error is asymptotic convergence along the iteration axis, meanwhile keep the speed profile of the subway train system satisfies the constraint. Finally, a subway train simulation is shown to verify the effectiveness of the theoretical studies. [ABSTRACT FROM AUTHOR]

This paper presents a novel facial image feature extraction technique using radial basis function neural network (RBFNN) and adaptive SIFT-SURF algorithm specifically for facial image feature extraction. The RBFNN adaptive SIFT-SURF algorithm is formulated as a derivative-based optimization problem which is trained using adaptive recursive least squares (ARLS) algorithm. The MATLAB interface for the implementation and evaluation of the proposed RBFNN adaptive SIFT-SURF algorithm for facial image feature extraction is also presented. A total of 1500 facial image samples were taken under different light intensities and postures to investigate the efficiency of the proposed algorithm. The performance of the proposed RBFNN adaptive SIFT-SURF algorithm has been compared with the standard and well established SIFT algorithm on the 1500 facial image samples. The simulation results show the superior performance of the proposed RBFNN adaptive SIFT-SURF algorithm when compared to the standard SIFT algorithm in terms of several evaluation parameters especially with the lowest computation time and a large number of descriptors. The efficiency of the proposed RBFNN adaptive SIFT-SURF algorithm indicates that it can be deployed for the design of real-time feature extract systems for facial images, signatures, thumbprints with so on. [ABSTRACT FROM AUTHOR]

This paper considers an adaptive neural event-triggered control problem of the molten steel level for twin roll strip casting systems with the inclined angle. Firstly, the model is improved and simplified into an affine nonlinear system by exploiting input compensation technique. Then, an adaptive observer is established based on radial basis function neural network (RBFNN). Furthermore, an adaptive event-triggered control scheme and corresponding adaptive updated laws are designed. It is proved that the proposed control method can guarantee the system output can follow reference signal and all closed-loop signals are bounded. Finally, the validity of the control scheme is verified through semi-experimental system dynamic model. [ABSTRACT FROM AUTHOR]

The dynamic properties of bridges can be affected not only through damage but also from ambient uncertainty. False-positive or negative alarms may be raised if environmental effects are not considered in the detection algorithm. This article presents a two-step data-driven approach that can incorporate temperature effects in vibration-based damage detection and localization, provided the temperature is also measured. To detect the occurrence of damage, prediction errors of an autoassociative neural network (AANN) framework are first employed as a temperature-invariant novelty index (NI). Further, for damage localization, NIs associated with each of the damage cases, are classified using a radial basis function neural network (RBFNN). The proposed algorithm is numerically tested on a multispan bridge structure involving measurement uncertainties. The performance of the RBFNN classifier is further assessed using different classifier performance metrics. The possibility of positive and negative false alarms raised by the proposed algorithm and its sensitivity to measurement noise contamination is also investigated. It is observed that the proposed data-based approach can efficiently isolate a fault, even in the presence of temperature variation. [ABSTRACT FROM AUTHOR]

• Predicting CO 2 equilibrium solubility using machine learning methods. • First application of XGBoost to CO 2 equilibrium solubility of mixed amines. • XGBoost is more accurate than RBFNN and SVR. • Verification of the ductility of XGBoost in mixed amine solutions. In this work, three machine learning methods were employed to predict carbon dioxide (CO 2) equilibrium solubility in blended amine solutions consisting of ethanolamine (MEA), N,N-diethylethanolamine (DEEA) and methyldiethanolamine (MDEA). Three machine learning algorithms, Radial Basis Function Neural Network (RBFNN), Support Vector Machine Regression (SVR) and Extreme Gradient Boosting (XGBoost), were used to fit the experimental results. We found that the predicted values of the three models developed for the CO 2 equilibrium solubility were in good agreement with the experimental results, and the comparison of the mean absolute percentage error (MAPE) and root mean square error (RMSE) results showed that XGBoost had the best prediction accuracy with the MAPE of less than 1%. Four different amine blends were then used to evaluate the expandability of the XGBoost model, namely MEA and 1-dimethylamino-2-propanol (1DMA2P), MDEA and piperazine (PZ), diethylenetriamine (DETA) and PZ, and MEA and triethanolamine (TEA); the MAPE between the experimental and predicted results were 0.30%, 0.91%, 0.86% and 1.63%, respectively. The results obtained showed that the XGBoost model has enormous potential for application in dealing with the CO 2 equilibrium solubility in blended amine solutions. [ABSTRACT FROM AUTHOR]

The railway relay plays an important role in railway systems. Its reliability has a significant effect on the safety of passengers and train operation, which can be reflected using degradation parameters. In this paper, a novel hybrid method based on the multi-layer decomposition and the radial basis function neural network (RBFNN) is proposed for life prediction of railway relays. As the degradation parameter series are usually nonlinear and non-stationary, it is vital to develop an essential method to preprocess the degradation series. In order to improve the prediction accuracy, a multi-layer decomposition method is developed first for data pre-processing, which blends complete ensemble empirical mode decomposition (CEEMD) and an improved variational mode decomposition (IVMD) with a stopping criterion for determining the decomposition modes number. It is noted that IVMD is then used to decompose the high-frequency intrinsic mode functions (IMFs) obtained using CEEMD to improve the prediction accuracy. Furthermore, RBFNN is applied to all the components for prediction. And the prediction results of all the components are reconstructed as the predicted degradation series. Finally, the effectiveness and robustness of the proposed novel hybrid prediction method are verified on one-step prediction and multi-step prediction by comparing other commonly used prediction methods. The experimental results indicate that the proposed hybrid prediction method performs best on the complex degradation parameters of safety relays.

This paper presents a robust adaptive backstepping control (RABC) for high-speed permanent-magnet synchronous motor (HSPMSM) drive system. The proposed RABC achieves high performance operation by incorporating an ideal backstepping controller (IBC), a recurrent radial basis function neural network (RRBFNN) uncertainty observer, and a robust controller. The Lyapunov stability theorem is utilized to design the IBC as a position controller of the HSPMSM servo drive system. To enhance the disturbance rejection capability during parameter changes, certain information is needed within the backstepping control law so that the system performance would not sorely be affected. To mitigate the need for the lumped parameter uncertainties within the backstepping controller, an online adaptive observer based on RRBFNN is designed to estimate the nonlinear parameter uncertainties. Moreover, the robust controller is intended to retrieve the remaining of the RRBFNN approximation errors. To assure the stability of the proposed RABC, the Lyapunov stability analysis is used to derive the online adaptive control laws. The performance of the proposed RABC is verified by simulation and experimental analysis under different operating conditions and parameter uncertainties. The test results validate the effectiveness of the proposed RABC scheme to achieve preferable tracking performance regardless of external disturbances and parameter uncertainties.

When a cross-domain robot (CDR) flies on the water surface, the large pitch angle and roll angle may lead to water flooding into the robot cabin or even overturning. In addition, the CDR is influenced by some uncertain parameters and external disturbances, such as the water resistance and current. To constrain the robot attitude angle and improve the robustness of the controller, a non-singular terminal sliding mode asymmetric barrier control (NTSMABC) algorithm is proposed. All the uncertain disturbances are regarded as a lump disturbance, and a radial basis function neural network (RBFNN) is designed to compensate for the output of the controllers. Unlike the traditional quadrotors, the robot controls the yaw angle by paddles when the robot flies on the water surface. To prevent the actuator saturation and the robot from rolling over due to excessive yaw angular velocity, an adaptive integral sliding mode barrier control (AISMBC) algorithm is proposed to constrain the yaw angular velocity directly. This algorithm adaptively adjusts the gain of the sliding surface to suppress the influence of the lump disturbance on the robot. Another RBFNN is designed to compensate for the output of the controller. Simulation results demonstrate the effectiveness of the proposed control methods.

The paper proposes the method to deal with control problems of unmodeled components of the four-wheeled Omni-directional mobile robot. It is commonly challenging to design a model-based control scheme to achieve smooth movement in the tracking process due to the unknown elements in the mathematical model of the robot or external disturbances. Our main contribution focuses on designing an adaptive controller based on neural networks with online weight updating laws and Fuzzy logic to guarantee the high accuracy of the robot's movement when the unknown factors adversely affect the robot control. At the initial step, a Dynamic Surface Control plays a role as a core of the controller for the robot system. Then, with the ability to estimate the appropriate value for uncertain nonlinear parts, a Radial Basis Function Neural Network is designed. Finally, a Fuzzy law is to utilize control parameters in each period to increase the adaptive behavior of the system. The stability and convergence of the system are proven by the Lyapunov's stability theory. The simulation results illustrate the validity and the efficiency of the proposed control algorithm when the system is lack of robot model's information. [ABSTRACT FROM AUTHOR]

Electroencephalography (EEG) is a major clinical tool to diagnose, monitor and manage neurological disorders which is mostly affected by artifacts. Given the importance and the need for an automated method to remove artifacts, in this paper some intelligent automated methods are proposed which are composed of three main parts as extraction of effective input, filtering and filter optimization. Wavelet transform is utilized to extract the effective input, and the wavelet approximation coefficients are used as an effective input signal. In addition, Radial Basis Function Neural Network (RBFNN) has been used for filtering. The appropriate number of RBFs has been selected using extensive simulations, and the optimal value of spread parameter has been achieved by Bees algorithm (BA). Finally, the proposed artifact removal schemes have been evaluated on some real contaminated EEG signals in Mashad Ghaem hospital database. The results show that the proposed artifact removal schemes are able to effectively remove artifacts from EEG signals with little underlying brain signal distortion. [ABSTRACT FROM AUTHOR]

The primary aim of this study is to investigate suitable Statistical Neural Network (SNN) models and their hybrid version for COVID-19 mortality prediction in Indian populations and is to estimate the future COVID-19 death cases for India. SNN models such as Probabilistic Neural Network (PNN), Radial Basis Function Neural Network (RBFNN), and Generalized Regression Neural Network (GRNN) are applied to develop the COVID-19 Mortality Rate Prediction (MRP) model for India. For this purpose, we have used two datasets as D1 and D2. The performances of these models are evaluated using Root Mean Square Error (RMSE) and “R,” a correlation value between actual and predicted value. To improve prediction accuracy, the new hybrid models have been constructed by combining SNN models and the Non-linear Autoregressive Neural Network (NAR-NN). This is to predict the future error of the SNN models, which adds to the predicted value of these models for getting better MRP value. The results showed that the PNN and RBFNN-based MRP model performed better than the other models for COVID-19 datasets D2 and D1, respectively.

Sedimentation management is one of the primary factors in achieving sustainable development of water resources. However, due to difficulties in conducting in-situ tests, and the complex nature of fine sediments, it remains a challenging task when dealing with issues related to settling velocity. Hence, the machine learning model appears as a suitable tool to predict the settling velocity of fine sediments in water bodies. In this study, three different machine learning-based models, namely, the radial basis function neural network (RBFNN), back propagation neural network (BPNN), and self-organizing feature map (SOFM), were developed with four hydraulic parameters, including the inlet depth, particle size, and the relative x and y particle positions. The five distinct statistical measures, consisting of the root mean square error (RMSE), Nash–Sutcliffe efficiency (NSE), mean absolute error (MAE), mean value accounted for (MVAF), and total variance explained (TVE), were used to assess the performance of the models. The SOFM with the 25 × 25 Kohonen map had shown superior results with RMSE of 0.001307, NSE of 0.7170, MAE of 0.000647, MVAF of 101.25%, and TVE of 71.71%.

Radial basis function neural networks are a widely used type of artificial neural network. The number and centers of basis functions directly affect the accuracy and speed of radial basis function neural networks. Many studies use supervised learning algorithms to obtain these parameters, but this leads to more parameters that need to be determined, thereby making the system more complex. This study proposes a modified nearest neighbor-based clustering algorithm for training radial basis function neural networks. The calculation of this clustering algorithm is not large, and it can adapt to varying densities. Furthermore, it does not require researchers to set parameters based on experience. Simulation proves that the clustering algorithm can effectively cluster samples and optimize the abnormal samples. The radial basis function neural network based on modified nearest neighbor-based clustering has higher accuracy in curve fitting than the conventional radial basis function neural network. Finally, the path tracking control based on a radial basis function neural network of a magnetic microrobot is investigated, and its effectiveness is verified through simulation. The test accuracy and training accuracy of the radial basis function neural network was improved by 23.5% and 7.5%, respectively.

In this paper, the position/velocity tracking control problem of high speed train(HST) is investigated with considering some inevitable factors such as the input nonlinearity due to different notches of traction/braking forces, aerodynamic resistance, in-train force, external disturbance and unknown actuator failures, which lead to the uncertainty and nonlinearity of HST. Aiming at the system characteristics of HST, a set of integer-order control methods based on the excellent approximation ability of Radial Basis Function Neural Network(RBFNN) are established firstly, and motivated by them, a kind of RBFNN-based fractional-order control methods are proposed by utilizing the genetic attenuation properties of fractional calculus(FC) in order to improve the control performance in the work. It should be pointed out that all the developed methods are able to deal with uncertainties and nonlinearities as well as actuator failures without the need for any “trail and error” process. The feasibility and effectiveness of the proposed control methods are verified by Lyapunov theoretical analysis and numerical simulation studies. Besides, the control performance of integer-order control system and fractional-order control system is compared and analyzed, and the results show that the fractional-order control system is superior. [ABSTRACT FROM AUTHOR]

Crosswell electromagnetic (EM) method has fundamentally improved the horizontal detection ability of well logging and will become an increasingly promising approach for the secondary exploration of hydrocarbon reservoir. We applied orthogonal least squares (OLS) radial basis function neural network (RBFNN) based on improved Gram–Schmidt (G–S) procedure to three-dimensional (3D) crosswell EM inversion problems. In the inversion process of the simplified crosswell model with single-grid conductivity anomalies and normal oil reservoir, compared the inversion results of other five neural networks, OLS-RBFNN was proved to have the best global optimization ability and the fastest sample learning speed and the average inversion error of low conductivity anomalies model (4%) and oil reservoir model (9%) can meet the inversion requirements of crosswell EM method. Only the OLS-RBFNN could achieve ideal inversion results in the most concerned central area of crosswell model, and the inversion accuracy of this algorithm will be more outstanding when the model becomes more complex. Merely using the three-component time-domain crosswell EM data of two wells, the inversion of 3D medium conductivity in the crosswell dominant exploration area can be effectively realized through the nonlinear approximation of the OLS-RBFNN. [ABSTRACT FROM AUTHOR]

To harness electricity from solar photo-voltaic (SPV) cell at maximum power point is the challenge for researcher. Artificial Intelligence (AI) plays a major role in control and estimation, hence, if trained properly the AI system can provide accurate solutions. Since the fast maximum power point tracking (MPPT) of SPV system is desirable, the RBFNN MPPT algorithm that has the advantage of universal approximation and fast learning is designed using Gaussian activation function and compared with conventional perturb and observe (P&O), back propagation neural network (BPNN) in this paper. The system is tested for MPPT operation on different values of irradiances and a comparative evaluation of power tracking efficiency, settling time, ripple content, average power loss is presented for the SPV system to validate the proposed algorithm. [ABSTRACT FROM AUTHOR]

This paper develops an effective identification and compensation mechanism for the disturbance‐like parametric friction of a typical underactuated tractor‐trailer vehicle system. To begin with, a parametric friction model is proposed to describe various friction effects associated with the system velocity, and then a disturbance‐like parametric friction concept is introduced by considering the motion characteristics of tractor‐trailer vehicle. Next, the radial basis function neural network (RBFNN) is employed to identify the friction due to its high convergence rate, superior approximation precision and local‐minima avoidance ability. Afterwards, a sliding mode control (SMC) is utilized to compensate the identified friction due to its numerous merits, such as strong robustness and fast convergence. On the basis of the effective combination of identification and compensation mechanism, a favorable transient performance can be achieved during the desired velocity tracking process. Lastly, the simulation results confirm that the RBFNN‐based disturbance‐like parametric friction identification and compensation mechanism can effectively improve the trajectory tracking performance of tractor‐trailer vehicle. [ABSTRACT FROM AUTHOR]

To reduce pesticide waste caused by low spray droplet coverage on target surface, the spraying system parameters of a six-rotor UAV sprayer are optimized. Computational Fluid Dynamics (CFD) is employed to simulate the downwash airflow distribution and droplet deposition for the UAV sprayer. Latin hypercube sampling method is used to obtain 15 sample points to train the surrogate model with CFD results. The validated CFD model is coupled with radial basis function neural networks (RBFNN) and genetic algorithm (GA) to optimize the spraying system parameters, including numbers of nozzles, horizontal position of nozzles and vertical position of nozzles. The overall correlation coefficient (R) of this RBFNN model is about 0.949, indicating good fitting performance. The optimal nozzles number of a six-rotor UAV sprayer is 4. The spraying droplet coverage is increased by 16.27%, 23.25% and 18.24% at a rotor height of 1 m, 2 m and 3 m above from the target surface, respectively, compared with the original six-rotor UAV spraying system structure. The spraying droplet coverage and pesticide utilization of the six-rotor UAV sprayer are improved through optimizing the mounting number and mounting positions of the nozzles in the spraying system. The research results provide a theoretical basis for further design and optimization of the multi-rotor UAV sprayers. • The validated CFD model coupled with RBFNN and GA is employed to optimize the UAV spraying system parameters. • The overall correlation coefficient (R) of this RBFNN model is about 0.949, indicating good fitting performance. • The spraying droplet coverage is increased by 16.27%, 23.25% and 18.24% with the optimized spraying system parameters. • The coupled method of CFD, RBFNN and GA can provide a theoretical basis for further design of the UAV sprayers. [ABSTRACT FROM AUTHOR]

For urban waterlogging alleviation, green infrastructures have been widely concerned. How to carry out scientific green infrastructure planning becomes an important issue in flood control and disaster relief. Based on historical media records of urban waterlogging from 2017 to 2020 and combined with variables about topography, land cover and socioeconomics, we used the Radial Basis Function Neural Network (RBFNN) to conduct urban waterlogging susceptibility assessment and simulate the risk of waterlogging in different scenarios of green land configuration in Shenzhen. The results showed that: (1) high proportions of impervious surface and population could increase the risks in Luohu and Futian districts, followed by Nanshan and Baoan districts, while high proportions of green space could effectively reduce the risks in southeastern Shenzhen; (2) urban waterlogging in Luohu and Futian districts can be alleviated by strengthening green infrastructure construction while Longgang and Longhua districts should make comprehensive use of other flood prevention methods; (3) turning existing urban green space into impervious surfaces would increase the risks of waterlogging, which is more evident in places with high proportions of green space such as Dapeng and Yantian districts. The effectiveness of green infrastructures varies in different spatial locations. Therefore, more attention should be paid to protecting existing green spaces than cultivating more green infrastructures in urban waterlogging alleviation.

High parking accuracy, comfort and stability, and fast response speed are important indicators to measure the control performance of a fully automatic operation system. In this paper, aiming at the problem of low accuracy of the fully automatic operation control of urban rail trains, a radial basis function neural network position output-constrained robust adaptive control algorithm based on train operation curve tracking is proposed. Firstly, on the basis of the mechanism of motion mechanics, the nonlinear dynamic model of train motion is established. Then, RBFNN is used to adaptively approximate and compensate for the additional resistance and unknown interference of the train model, and the basic resistance parameter adaptive mechanism is introduced to enhance the anti-interference ability and adaptability of the control system. Lastly, on the basis of the RBFNN position output-constrained robust adaptive control technology, the train can track the desired operation curve, thereby achieving the smooth operation between stations and accurate stopping. The simulation results show that the position output-constrained robust adaptive control algorithm based on RBFNN has good robustness and adaptability. In the case of system parameter uncertainty and external disturbance, the control system can ensure high-precision control and improve the ride comfort.

Aiming at the problem of the large tracking error of the desired curve for the automatic train operation (ATO) control strategy, an ATO control algorithm based on RBF neural network adaptive terminal sliding mode fault-tolerant control (ATSM-FTC-RBFNN) is proposed to realize the accurate tracking control of train operation curve. On the one hand, considering the state delay of trains in operation, a nonlinear dynamic model is established based on the mechanism of motion mechanics. Then, the terminal sliding mode control principle is used to design the ATO control algorithm, and the adaptive mechanism is introduced to enhance the adaptability of the system. On the other hand, RBFNN is used to adaptively approximate and compensate the additional resistance disturbance to the model so that ATO control with larger disturbance can be realized with smaller switching gain, and the tracking performance and anti-interference ability of the system can be enhanced. Finally, considering the actuator failure and the control input limitation, the fault-tolerant mechanism is introduced to further enhance the fault-tolerant performance of the system. The simulation results show that the control can compensate and process the nonlinear effects of control input saturation, delay, and actuator faults synchronously under the condition of uncertain parameters, external disturbances of the system model and can achieve a small error tracking the desired curve.

The consistent and efficient method for the identification of biometrics is the iris recognition in view of the fact that it has richness in texture information. A good number of features performed in the past are built on handcrafted features. The proposed method is based on the feed-forward architecture and uses k-means clustering algorithm for the iris patterns classification. In this paper, segmentation of iris is performed using the circular Hough transform that realizes the iris boundaries in the eye and isolates the region of iris with no eyelashes and other constrictions. Moreover, Daugman's rubber sheet model is used to transform the resultant iris portion into polar coordinates in the process of normalization. A unique iris code is generated by log-Gabor filter to extract the features. The classification is achieved using neural network structures, the feed-forward neural network and the radial basis function neural network. The experiments have been conducted using the Chinese Academy of Sciences Institute of Automation (CASIA) iris database. The proposed system decreases computation time, size of the database and increases the recognition accuracy as compared to the existing algorithms. [ABSTRACT FROM AUTHOR]

Summary: In this paper, a novel adaptive control scheme is proposed based on radial basis function neural network (RBFNN). The considered system is deduced by the structure of RBFNN with nonzero time‐varying parameter that installed in the fore‐end and terminal of RBFNN. With this structure and the Taylor expansion of any smooth continuous nonlinear function, a universal approximation of RBFNN is addressed according to the analysis of the character of continuous homogenous function and the Euler's theorem. The approximation accuracies can be adjusted online by the nonzero time‐varying parameter in the device with the degree of continuous homogenous function, which expand the semiglobally stability to global stability over conventional neural controller design approaches. Based on the theory analysis of barrier Lyapunov function, the violation of time‐varying constraints can be subjugated without wrecked. Finally, simulation results are carried out to verify the effectiveness by the design methods. [ABSTRACT FROM AUTHOR]

A contactless method uses infrared image features and radial basis function neural network (RBFNN) to detect contamination levels for porcelain insulators is proposed in this paper. First, theory evidence for contamination levels detection by infrared images is inferred. Then, the denoising and image segmentation is implemented to suppress the image noise and eliminate the affection of the background. Nine color moment features related to the contamination levels are extracted from the insulator images. Finally, an RBFNN is constructed to identify the contamination levels. The images features and ambient relative humidity are taken as the inputs of the RBFNN. For improve the precision of the detection, a new method based on the statistical probability of the values of each contamination feature component is proposed to select the initial hidden center parameters for RBFNN hidden nodes. An improved learning algorithm combined with the gradient descent algorithm and a random number control factor are proposed to modify the hidden center parameters and the weights vectors. Testing results show that the selected color moment features are effective on the contamination levels representation and the constructed RBFNN performs better than back propagation neural network (BPNN) and generalized regression neural network (GRNN) on contamination levels identification. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel approach for tracking control with switching formation in nonomniscient constrained space for multiagent system (MAS). The introduction of switching formation results from the situation where MAS is maneuvering in restricted path which is often the case for real world application. The preplanned trajectory may be inaccurate due to the lack of sufficient environmental information. In this case, agents may have to rapidly avoid collisions with unexpected obstacles or even switch the formation to guarantee the passability. A concept of avoidless disturbance is proposed. To solve the undesirable chattering on resulting trajectory caused by avoidless disturbance and the existing adaptation algorithm for neural network (NN) weights, a local path replanning approach is designed such that the potential force generated by avoidless disturbance is acted on the original desired trajectory outputting the locally replanned path for an agent. An NN-based controller is designed and the performance is validated using Lyapunov functions. Simulations are carried out to illustrate the effectiveness of proposed strategies. [ABSTRACT FROM AUTHOR]