Adaptive Neural Network System Research Articles

Complex adaptive learning cortical neural network systems for solving time-fractional difference equations with bursting and mixed-mode oscillation behaviours

Complex networks have been programmed to mimic the input and output functions in multiple biophysical algorithms of cortical neurons at spiking resolution. Prior research has demonstrated that the ineffectual features of membranes can be taken into account by discrete fractional commensurate, non-commensurate and variable-order patterns, which may generate multiple kinds of memory-dependent behaviour. Firing structures involving regular resonator chattering, fast, chaotic spiking and chaotic bursts play important roles in cortical nerve cell insights and execution. Yet, it is unclear how extensively the behaviour of discrete fractional-order excited mechanisms can modify firing cell attributes. It is illustrated that the discrete fractional behaviour of the Izhikevich neuron framework can generate an assortment of resonances for cortical activity via the aforesaid scheme. We analyze the bifurcation using fragmenting periodic solutions to demonstrate the evolution of periods in the framework’s behaviour. We investigate various bursting trends both conceptually and computationally with the fractional difference equation. Additionally, the consequences of an excitable and inhibited Izhikevich neuron network (INN) utilizing a regulated factor set exhibit distinctive dynamic actions depending on fractional exponents regulating over extended exchanges. Ultimately, dynamic controllers for stabilizing and synchronizing the suggested framework are shown. This special spiking activity and other properties of the fractional-order model are caused by the memory trace that emerges from the fractional-order dynamics and integrates all the past activities of the neuron. Our results suggest that the complex dynamics of spiking and bursting can be the result of the long-term dependence and interaction of intracellular and extracellular ionic currents.

Reduction in Residential Electricity Bill and Carbon Dioxide Emission through Renewable Energy Integration Using an Adaptive Feed-Forward Neural Network System and MPPT Technique

Increasing electricity demand and the emergence of smart grids have given home energy management systems new potential. This research investigates the use of an artificial neural network algorithm for a home energy management system. The system keeps track of and organizes the use of electrical appliances in a typical home with the objective of lowering consumer electricity bills. An artificial-neural-network-based maximum-power-point-tracking scheme is applied to maximize power generation from photovoltaic sources. The proposed neural network senses solar energy and calculates load requirements to switch between solar and grid sources effectively. The implementation of improved source utility does not require numerical calculations. Traditional relational operator techniques and fuzzy logic controllers are compared with the suggested neural network. The model is simulated in MATLAB, and the results show that the artificial neural network performs better in terms of source switching following load demand, with an operating time of less than 2 s and a reduced error of 0.05%. The suggested strategy reduces electricity costs without affecting consumer satisfaction and contributes to environmental friendliness by reducing CO2 emissions.

Adaptive Neural Model Matching Control for Uncertain Immune Systems via H∞ Approaches

The problem of the robust neural network-based model matching control is considered for a large class of uncertain immune systems. In order to achieve the purpose of therapeutic enhancement, it is essential to deal simultaneously with the effects of plant uncertainties, time-varying perturbations, and continuing environmental pathogens. Neural network control algorithm, robust <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$H_{\infty } $ </tex-math></inline-formula> control theory and VSC technique are combined to construct the hybrid adaptive/robust tracking control scheme such that the controlled immune system achieves a satisfactory model matching control performance. An adaptive neural network system is constructed to learn the behavior of the immune system dynamics. Moreover, an algebraic Riccati-like inequality must be solved to achieve a desired <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$H_{\infty } $ </tex-math></inline-formula> control performance. Consequently, the robust control scheme developed here can be analytically computed and easily implemented. Simulation results are presented to demonstrate the effectiveness of the proposed control scheme.

Hierarchical Fuzzy Signature and Neuro-Fuzzy Hybrid System for QMS Control

Recently, the Quality Management System (QMS) control supports organizations managers identifying the best practices to upgrade the efficiency and effectiveness. This control has become a successful tool to improve the organization decision-making process. However, QMS encloses several performance indicators inputs that necessitate to be managed. In fact, it may include as inputs, customers requirements, quality policies, standard procedures and many other criteria. Hence, to provide the control of QMS problem insurance, two approaches are investigated which are Hierarchical Fuzzy Signature (HFS) and NeuroFuzzy Hierarchical Hybrid system (NFHH), respectively. These approaches are applied in the case of an industrial company operating in the electromechanical sector. This company has to be creative, agile, responsive and especially ready for fierce competition. Then, the obtained results are compared to Adaptive Neural Fuzzy Inference and Neural Network Systems, regarding the learning phase. Consequently, all of them help the company to assess its overall performance. However, NFHH reaches the best accuracy, reduces the number of neurons and uses the parameters that keep the universal approximated property of neural networks and fuzzy systems. Keywords—Quality management system, Fuzzy logic, Neuro-Fuzzy, hierarchical structure, fuzzy signature

Model of a neural network adaptive system for a digital control loop of an electric drive

The problem of increasing the efficiency of power units of autonomous electric transport vehicles is considered. The task of creating a promising power system control device has been singled out. It is determined that in creating such devices, significant results can be obtained by using an intelligent module in the control loop of the electric drive. Goal. It is necessary to develop a power plant model with intelligent control, allowing to obtain data sets about currents, voltages and engine speeds in different modes of operation. The architecture of an intelligent control device, a PID controller based on a neural network, has been proposed; it has been proposed to exclude rotor angular velocity sensors from the classical feedback loop. The type and architecture of the neural network is defined. In the software environment MatLab the model of neuroemulator of the engine for formation of a training sample of a neural network by a method of Levenberg – Marquardt is developed. The trained neural network is implemented in the developed model of the electric motor control loop. The results of simulation of the intelligent control device showed a good convergence of the output influences generated by the neuroemulator with the actual parameters of the electric motor.

Multivariate data-based optimization of membrane adsorption process for wastewater treatment: Multi-layer perceptron adaptive neural network versus adaptive neural fuzzy inference system

Application of machine-learning methods to assess the batch adsorption of malachite green (MG) dye on chitosan/polyvinyl alcohol/zeolite imidazolate frameworks membrane adsorbents (CPZ) was investigated in this study. Our previous research results proved the suitability of the CPZ membranes for wastewater decoloring. In the current work, the residence time was combined with the other operational variables i.e., pH, initial dye concentration, and adsorbent dose (AD), to obtain the possible interactions involved in nonequilibrium adsorption. Two well-known soft-computing approaches, multi-layer perceptron adaptive neural network (MLP-ANN) and adaptive neural fuzzy inference system (ANFIS), were selected among different machine learning alternatives and then, comprehensively compared with each other considering reliability and accuracy for a 60 number of runs. The ANFIS structure with nine centers of clusters could predict the adsorption performance better than the ANN approach. Root mean square error (RMSE) and R-square were obtained 0.01822 and 0.9958 for the test data, respectively. The interpretability test resulted a linear trend predicted by the model and disclosed that the maximum value of the removal efficiency (99.5%) could be obtained when the amount of the inputs set to the upper limit. Lastly, the sensitivity analysis uncovered that the residence time has a decisive effect (relevancy factor > 80%) on the removal efficiency. According to the results, ANFIS is an effective and reliable tool to optimize and intensify the membrane adsorption process.

Hierarchical Adaptive Control System of a Manipulator Based on the Synthesis of a Neural Network of Fuzzy Inference and an Iterative Refinement Algorithm

The main aim of this work is to develop an adaptive control system for a kinematically redundant multilink industrial manipulator. Proposed solution allows to construct a unified real-time control system with the ability to control the accuracy of calculations. In order to achieve the required accuracy of the calculations and the performance of the control system, we propose an algorithm that is based on the so-called hybrid method for finding the solution of the inverse kinematics (IK) problem, including the adaptive neural network and fuzzy inference system with subsequent iterative refinement of numerical solution by the Newton --- Raphson method. The influence of the training sample size on the quality of the obtained initial approximation for the neural network part of the algorithm is described in the paper. The results of experimental studies of the developed hybrid algorithm are presented in comparison with the iterative and neural network methods for three-, five- and eight-link manipulator structures. Paper presents the main steps of the control system synthesis for kinematically redundant industrial manipulator, including the description for developed algorithms for finding an IK solution of multilink structures. The structure of a multi-level hierarchical manipulator control system, based on a programmable logic controller and electric stepping motors with the possibility of integration into the production system at various levels, is presented.

Adaptive Fuzzy Sliding Mode Guidance Law considering Available Acceleration and Autopilot Dynamics

Terminal guidance law for missiles intercepting high maneuvering targets considering the limited available acceleration and autopilot dynamics of interceptor is investigated. Conventional guidance laws based on adaptive sliding mode control theory were designed to intercept a maneuvering target. However, they demand a large acceleration for interceptor at the end of the terminal guidance, which may have acceleration saturation especially when the target acceleration is close to the available acceleration of interceptor. In this paper, a terminal guidance law considering the available acceleration and autopilot dynamics of interceptor is proposed. Then, a fuzzy system is utilized to approximate and replace the variable structure term, which can handle the unknown target acceleration. And an adaptive neural network system is adopted to compensate the effects caused by the designed overlarge acceleration of interceptor such that the interceptor with small available acceleration can intercept the high maneuvering target. Simulation results show that the guidance law with available acceleration and autopilot dynamics (AAADG) is highly effective for reducing the acceleration command and achieving a small final miss distance.

Prediction of Skid Resistance Value of Glass Fiber-Reinforced Tiling Materials

This research focuses on the use of adaptive artificial neural network system for evaluating the skid resistance value (British Pendulum Number; BPN) of the glass fiber-reinforced tiling materials. During the creation of the neural model, four main factors were considered: fiber, calcium carbonate content, sand blasting, and polishing properties of the specimens. The model was trained, tested, and compared with the on-site test results. As per the comparison of the outcomes of the study, the analysis and on-site test results showed that there is a great potential for the prediction of BPN of glass fiber-reinforced tiling materials by using developed neural system.

Design of an intelligent optimal neural network-based tracking controller for nonholonomic mobile robot systems

This paper addresses the trajectory-tracking control problem of mobile robot systems with nonholonomic constraints, in the presence of time-varying parametric uncertainties and external disturbances. This necessitates an accurate controller to meet as much control objectives, as possible. Therefore, this paper deals with an artificial intelligence control technique to design a robust controller to meet the control objectives. The design of the intelligent controller is based on the optimal control theory, the adaptive neural network system, and the robust control technique. The trajectory-tracking problem is solved using the optimal control methodology. Since the nonholonomic wheeled mobile robot is strongly nonlinear, the neural network system is applied to approximate the nonlinear function in the optimal control law. The robust controller, for his part, is then applied to adaptively estimate an unknown upper bound of the time-varying parametric uncertainties, external disturbances and approximation error of the neural network system. The stability of the closed-loop robot system is proven using the optimal control theory and Lyapunov stability analysis. The results of the simulation studies on three typical nonholonomic mobile robots are provided to demonstrate the effectiveness of the proposed controller. In addition, a comparative study with a recent robust adaptive controller shows that our proposed intelligent controller gives better results, in the sense that the output trajectory converges to the steady state faster with smaller tracking error.

Adaptive H2/H∞ tracking control for a class of uncertain robotic systems

ABSTRACTThis paper addresses the problem of designing mixed H2/H∞ tracking control for a large class of uncertain robotic systems. Nonlinear H∞ control theory, H2 control theory and intelligent adaptive control algorithm are combined to construct a hybrid adaptive/robust H2/H∞ tracking control scheme. One adaptive neural network system is constructed to approximate the behaviour of uncertain robot dynamics, and the other adaptive control algorithm is designed to estimate the behaviour of the modelled disturbance. Moreover, a robust H∞ control algorithm is designed to attenuate the effects of the unmodelled disturbance. Only a set of algebraic matrix Riccati-like equations is required to implement the proposed mixed H2/H∞ tracking controller, and so an explicit and closed-form solution is obtained. Consequently, the mixed H2/H∞ adaptive/robust tracking controller developed here can be analytically computed and easily implemented. Finally, simulations are presented to illustrate the effectiveness of the proposed control algorithm.

Robust H∞ Tracking Control of Uncertain Robotic Systems with Periodic Disturbances

AbstractThis paper addresses the problem of designing robust tracking control for a large class of uncertain robotic systems. A more general model of the external disturbance is employed in the sense that the external disturbance can be expressed as the sum of a modeled disturbance and an unmodeled disturbance, for example, any periodic disturbance can be expressed in this general form. An adaptive neural network system is constructed to approximate the behavior of unknown robot dynamics. An adaptive control algorithm is designed to estimate the behavior of the modeled disturbance, and in turn the robust H∞ control algorithm is required to attenuate the effects of the unmodeled disturbance only. Consequently, an intelligent adaptive/robust tracking control scheme is constructed such that an H∞ tracking control is achieved in the sense that all the states and signals of the closed‐loop system are bounded and the effect due to the unmodeled disturbance on the tracking error can be attenuated to any preassigned level. Finally, simulations are provided to demonstrate the effectiveness and performance of the proposed control algorithm.

Output feedback tracking control for nonlinear time-delay systems with tracking errors and input constraints

The adaptive tracking control problem is considered for a class of nonlinear time-delay systems in the presence of input and tracking error constraint. A reduced-order observer is designed to estimate the unmeasured state variables at first. Then, a constraint variable is utilized to ensure that the tracking error is within the prescribed boundaries. An auxiliary state is introduced to deal with the input saturation constraint. With the time-delay functions unavailable, we employ adaptive RBF neural network systems to approximate unknown functions. It is proved that the resulting closed-loop system is stable in the sense of semiglobal uniformly ultimately boundedness. The simulations are performed and the results demonstrate the effectiveness of the proposed approach.

Finite Time Control Design for Bilateral Teleoperation System With Position Synchronization Error Constrained.

Due to the cognitive limitations of the human operator and lack of complete information about the remote environment, the work performance of such teleoperation systems cannot be guaranteed in most cases. However, some practical tasks conducted by the teleoperation system require high performances, such as tele-surgery needs satisfactory high speed and more precision control results to guarantee patient' health status. To obtain some satisfactory performances, the error constrained control is employed by applying the barrier Lyapunov function (BLF). With the constrained synchronization errors, some high performances, such as, high convergence speed, small overshoot, and an arbitrarily predefined small residual constrained synchronization error can be achieved simultaneously. Nevertheless, like many classical control schemes only the asymptotic/exponential convergence, i.e., the synchronization errors converge to zero as time goes infinity can be achieved with the error constrained control. It is clear that finite time convergence is more desirable. To obtain a finite-time synchronization performance, the terminal sliding mode (TSM)-based finite time control method is developed for teleoperation system with position error constrained in this paper. First, a new nonsingular fast terminal sliding mode (NFTSM) surface with new transformed synchronization errors is proposed. Second, adaptive neural network system is applied for dealing with the system uncertainties and the external disturbances. Third, the BLF is applied to prove the stability and the nonviolation of the synchronization errors constraints. Finally, some comparisons are conducted in simulation and experiment results are also presented to show the effectiveness of the proposed method.

Polynomial Based Functional Link Artificial Recurrent Neural Network adaptive System for predicting Indian Stocks

AbstractA low complexity Polynomial Functional link Artificial Recurrent Neural Network (PFLARNN) has been proposed for the prediction of financial time series data. Although different types of polynomial functions have been used for low complexity neural network architectures earlier for stock market prediction, a comparative study is needed to choose the optimal combinations of the nonlinear functions for a reasonably accurate forecast. Further a recurrent version of the Functional link neural network is used to model more accurately a chaotic time series like stock market indices with a lesser number of nonlinear basis functions. The proposed PFLARNN model when trained with the well known gradient descent algorithm produces reasonable accuracy with a choice of range of weight parameters of the network. However, to improve the accuracy of the forecast further, the weight parameters of the recurrent functional neural network are optimized using an evolutionary learning algorithm like the differential evo...

Study of Data Mining Algorithms for Prediction and Diagnosis of Diabetes Mellitus

mellitus or simply diabetes is a disease caused due to the increase level of blood glucose. Various available traditional methods for diagnosing diabetes are based on physical and chemical tests. These methods can have errors due to different uncertainties. A number of Data mining algorithms were designed to overcome these uncertainties. Among these algorithms, amalgam KNN and ANFIS provides higher classification accuracy than the existing approaches. The main data mining algorithms discussed in this paper are EM algorithm, KNN algorithm, K-means algorithm, amalgam KNN algorithm and ANFIS algorithm. EM algorithm is the expectation-maximization algorithm used for sampling, to determine and maximize the expectation in successive iteration cycles. KNN algorithm is used for classifying the objects and used to predict the labels based on some closest training examples in the feature space. K means algorithm follows partitioning methods based on some input parameters on the datasets of n objects. Amalgam combines both the features of KNN and K means with some additional processing. ANFIS is the Adaptive Neuro Fuzzy Inference System which combines the features of adaptive neural network and Fuzzy Inference System. The data set chosen for classification and experimental simulation is based on Pima Indian Diabetic Set from University of California, Irvine (UCI) Repository of Machine Learning databases. Keywordsmining, Diabetes, EM algorithm, KNN algorithm, K- means algorithm, amalgam KNN algorithm, ANFIS algorithm

Metallurgical Secondary Cooling of Continuous Casting Based on Neural Network Adaptive System Design of Water Distribution

Efficient secondary cooling water control level slab continuous casting process and quality are closely related. Casting solidification heat transfer model is the basis of process control and optimization, heat transfer model based on determining the secondary cooling system is the most widely used method for casting production process can be simulated. However, when considering the many factors affecting the production and input conditions change significantly, real-time and strain of this method is not guaranteed. Therefore, the artificial intelligence optimization algorithms such as genetic algorithms, neural networks, fuzzy controllers, introducing continuous casting secondary cooling water distribution and dynamics of optimal control methods, the rational allocation of caster secondary cooling water and dynamic control is important.

Adaptive Neural Network Finite-Time Control for Uncertain Robotic Manipulators

An adaptive neural network finite-time controller (NNFTC) for a class of uncertain nonlinear systems is proposed by using the backstepping method, which employs an adaptive neural network (NN) system to approximate the structure uncertainties and uses a variable structure term to compensate the approximation errors, thus improving the robustness of the system to external disturbances. The controller is then applied to uncertain robotic manipulators, with a control objective of driving the system state to the original equilibrium point. It is proved that the closed-loop system is finite-time stable. Moreover, simulated and experimental results indicate that the proposed NNFTC is effective and robust.

Learning from Adaptive Neural Control of Electrically-Driven Mechanical Systems

This study presents deterministic learning from adaptive neural control of unknown electrically-driven mechanical systems. An adaptive neural network system and a high-gain observer are employed to derive the controller. The stable adaptive tuning laws of network weights are derived in the sense of the Lyapunov stability theory. It is rigorously shown that the convergence of partial network weights to their optimal values and locally accurate NN approximation of the unknown closed-loop system dynamics can be achieved in a stable control process because partial Persistent Excitation (PE) condition of some internal signals in the closed-loop system is satisfied. The learned knowledge stored as a set of constant neural weights can be used to improve the control performance and can also be reused in the same or similar control task. Numerical simulation is presented to show the effectiveness of the proposed control scheme.

Adaptive Neural Network Control with Backstepping for Surface Ships with Input Dead-Zone

This paper addresses the problem of adaptive neural network controller with backstepping technique for fully actuated surface vessels with input dead-zone. The combination of approximation-based adaptive technique and neural network system is used for approximating the nonlinear function of the ship plant. Through backstepping and Lyapunov theory synthesis, an indirect adaptive network controller is derived for dynamic positioning ships without dead-zone property. In order to improve the control effect, a dead-zone compensator is derived using fuzzy logic technique to handle the dead-zone nonlinearity. The main advantage of the proposed controller is that it can be designed without explicit knowledge about the ship motion model, and dead-zone nonlinearity is well compensated. A set of simulations is carried out to verify the performance of the proposed controller.