Neural Network Based Control Scheme Research Articles

Reducing Gait Asymmetry in Transfemoral Amputees — A Neural Network-Based Approach

Gait asymmetry between intact and prosthetic limbs is one of the factors that affect the long-term health of an amputee. Existing prosthetic devices have fixed controllers that cannot adapt to changes in user gait, thereby contributing to this asymmetry. In this paper, we use the stance period of both limbs to measure the asymmetry in gait. We propose a neural network-based control strategy that can adapt to changing gait in real time and help achieve near-natural gait.

Neural Network-Based Region Tracking Control for a Flexible-Joint Robot Manipulator

Abstract The present paper proposes a neural network-based adaptive region-tracking control strategy for a flexible-joint robot manipulator subjected to region constraints. The developed neural network-based control strategy can globally stabilize the robot manipulator and cope with model uncertainties and the external unknown bounded disturbances. Different from the existing literature, by using the sliding mode technology and the singular perturbation theory, the developed control strategy does not require the high-order derivatives of the link states such as jerk and acceleration since the high-order derivative information is not always available in practical applications. By using Lyapunov stability theory, it is proved that the proposed neural network-based control strategy can guarantee that all the parameter variables in the closed-loop system are bounded, and the flexible-joint robot manipulator with unknown dynamics can reach inside the dynamic region and also maintain the velocity matching with the desired moving region. Since the assumption of linearization of the unknown dynamic parameters is removed, the proposed control strategy does not require the calculation of the complex regression matrix. Therefore, the proposed method has great robustness and the ability of model generalization. Simulations are given to demonstrate the validity of the proposed control strategy.

Adaptive Artificial Neural Network-Based Control Through Attracting-Manifold Design and Lipschitz Constant Projection

Abstract This article presents an adaptive artificial neural network-based control scheme that strategically integrates the attracting-manifold design and a smooth Lipschitz-constant projection operator. The essence of the scheme is elaborated through the design of the reference-command-tracking control law of an nth-order single-input uncertain nonlinear system that can be transformed into the Brunovsky form. The method offers two major advantages: First, by employing the attracting-manifold design, it is possible to achieve asymptotic recovery of the ideal (deterministic) closed-loop dynamics. In other words, the perturbation caused by parametric uncertainties will be driven to zero asymptotically as a result of such a design, fostering a superior control performance. Second, the established smooth projection operator ensures the Lipschitz constant of the adaptive artificial neural network is bounded from above, thereby providing a certain degree of robustness against adversarial perturbations. The proposed method is validated through a numerical simulation example and compared with a standard certainty-equivalent neural-adaptive control method to demonstrate its superior performance.

Appointed-time prescribed performance control for 6-DOF spacecraft rendezvous and docking operations under input saturation

This study investigates the adaptive appointed-time control problem of chaser spacecraft rendezvous and docking with unknown target spacecraft information, parametric uncertainties, external disturbances, input saturation and performance constraints. Two kinds of novel prescribed performance function are first proposed to relax the limitation of the initial values of state signals and the contradiction between input saturation and performance constraints. Using dynamic surface control theory and a novel compensation system, an adaptive anti-saturation appointed-time prescribed performance controller and an adaptive anti-saturation appointed-time neural network-based prescribed performance controller are presented. An auxiliary variable is used to compensate for input saturation. Considering the strong approximation of the radial basis function neural network, an appointed-time neural network-based control scheme is proposed to compensate for external disturbances and parametric uncertainties. All closed-loop states converge to a neighborhood of zero in appointed-time, and the performance constraints of the relative position as well as the relative attitude are satisfied. A number of simulation scenarios are considered to verify the effectiveness of the proposed control strategies.

Robust adaptive neural control for a class of perturbed nonlinear systems with unmodeled dynamics and output disturbances

AbstractFor a nonlinear system with unmodeled dynamics and output disturbances, an adaptive neural network based control strategy is designed. For the unknown output interference, this article introduces a new module between the system output and the tracking signal for the first time. By constructing a new state, system becomes a new system without unknown output interference. In order to alleviate the major influence caused by the system dynamic disturbance and unknown time delay, appropriate Lyapunov–Krasovskii functions, and backstepping technology are employed. In designing some adaptive methods, the unmodeled dynamics, disturbances, and unknown time‐varying virtual coefficients are processed. Unknown function is approximated by neural network. All signals are semi‐global uniform ultimate boundedness. Simulation effectively verifies the control strategy.

Optimization and Power Management of Solar PV-based Integrated Energy System for Distributed Green Hydrogen Production

Photovoltaic-based integrated energy systems act as a possible modern technological solution for clean and affordable green hydrogen production. However, research attempts are required from the scientific community to develop power management, control, optimization algorithms, and new techniques for these integrated energy systems effective and economical operation. The integrated energy system considered in this work consists of a solar photovoltaic, battery, grid and proton exchange membrane (PEM) electrolyzer. PEM electrolyzer with a power rating of 100 kW is modelled as a controlled current sink to interfere with the DC bus directly. This work proposes a power management and control algorithm for the photovoltaic-based integrated energy system considering different parameters like availability of solar photovoltaic power, DC link voltage, battery state of charge (SOC), tariff and availability of grid power. Effective power-sharing among the different power sources increases system reliability and stability. Optimization is performed for optimal sizing and costing to achieve economical operation such that these photovoltaic-based integrated energy systems become affordable and are encouraged for wide use. The photovoltaic system operates at the maximum power point through a neural network-based control strategy; in addition to this, stacking in neural network topologies is also discussed for improving system accuracy in tracking the reference voltage for MPPT operation. The DC link voltage is controlled by interfacing the battery with the DC link via a bidirectional buck-boost power converter. The contribution of the work is to highlight the role of renewable energy sources for continuous green hydrogen production. Green hydrogen will be a critical driving force for the future transportation system, fuel cell-based power generation, and industries utilizing hydrogen in direct or indirect forms-that will help transition towards a carbon-free society. Optimization and time-domain simulation results indicated the economic and technical feasibility of the proposed photovoltaic-based integrated energy system for green hydrogen production with the best optimal configuration producing hydrogen at the cost of $ 4.806/kg and total net present cost of $749904.

Adaptive pid regulation method of uninterruptible power supply battery charge current based on artificial neural network

The paper considers the issue of improving the methods of uninterruptible power supply intelligent control based on adaptive algorithms with the use of artificial neural network technologies. An adaptive PID regulation method of the UPS battery charge current is proposed. A neural network based control scheme with adjustment of PID regulation coefficients has been developed. The simulation modeling was used to search for the artificial neural network topology, which would be the most effective according to the criteria of the battery charge current regulation accuracy. The use of artificial neural networks in the uninterruptible power supply control system made it possible to obtain more effective results for maintaining a stable battery charge current in the transient.

Machine learning and its impact on control systems: A review

Control systems play a vital role in our day to day lives, from mobile phones to autopilot systems that precisely navigate airplanes, they can be found anywhere. The primary role of any control system is to achieve a set point defined by the user and provide stability to the system while doing so. There are various approaches that are used for the design of control systems such as PID (Proportional, Integral and Derivative control) algorithms, Fuzzy logic controller, Neural Network controllers etc. Machine learning (ML) is a key tool in analysing time series data and can be used to predict the future states of any dynamic system, however, sufficient historic data is required. This key feature of ML based predictive algorithms can be employed in modelling a wide range of non-linear dynamic systems for ensuring system stability as well as seeking continuous improvement in system response. This paper gives a brief introduction of conventional methods used in the design of control systems with focus on Neural Network based control strategies which employ Recurrent Neural Network (RNN) and Long Short Term Memory (LSTM) networks for system identification and predictive control of dynamic processes.

Accurate power-sharing, voltage regulation, and SOC regulation for LVDC microgrid with hybrid energy storage system using artificial neural network

ABSTRACT In this paper, an artificial neural network-based control strategy is proposed for low voltage DC microgrid (LVDC microgrid) with a hybrid energy storage system (HESS) to improve power-sharing between battery and supercapacitor (SC) to suit the demand-generation imbalance, maintain state-of-charge (SOC) within boundaries and thereby to regulate the dc bus voltage. The conventional controller cannot track the SCs current rapidly with the high-frequency component that will place dynamic stress on the battery, further resulting in shorter battery life. The significant advantage is that in the proposed control strategy, redirections of unwaged battery currents to SCs for fast compensations enhance battery life span. The proposed control strategy effectiveness was investigated by simulations, including a comparison of overshoot/undershoot and settling time in dc bus voltage with a conventional control strategy. The results have been experimentally verified by hardware-in-loop (HIL) on a field-programmable gate array (FPGA)-based real-time simulator.

Analysis of a Function Fitting Neural Network based Control Strategy for a Hybrid Energy Storage System applied to Electric-Vehicles

Analysis of a Function Fitting Neural Network based Control Strategy for a Hybrid Energy Storage System applied to Electric-Vehicles

Online sequential extreme learning machine based adaptive control for wastewater treatment plant

Wastewater Treatment Plant (WWTP) is challenging to regulate for its complex chemical and biological characteristics, and its precise mathematical model is usually not accessible due to the limitation of available measurement. Traditional methods highly rely on human intervention and cannot adapt to the varying environment. Meanwhile, adaptive neural network based control strategies are encountered with dilemmas of local minima, slow convergence, huge time consumption, and low efficiency. To overcome such challenges, in this paper, a novel Online Sequential Extreme Learning Machine (OS-ELM) based adaptive control is discussed. In contrast, no a prior human experience or off-line training phase is required. It also has fast training speed thanks to randomly generated parameters of the hidden layer and Moore–Penrose pseudo-inverse. Moreover, the performance of the system is guaranteed even in the presence of time-varying dynamics and uncertainties through a learning mechanism in an online manner. The stability of the closed-loop system is shown strictly and extensive comparison case studies substantiate the feasibility of the scheme.

Fully Adaptive-Gain-Based Intelligent Failure-Tolerant Control for Spacecraft Attitude Stabilization Under Actuator Saturation

This article investigates the attitude stabilization problem of a rigid spacecraft with actuator saturation and failures. Two neural network-based control schemes are proposed using anti-saturation adaptive strategies. To satisfy the input constraint, we design two controllers in a saturation function structure. Taking into account the modeling uncertainties, external disturbances, and adverse effects from actuator faults and failures, the first anti-saturation adaptive controller is implemented based on radial basis function neural networks (RBFNNs) with a fixed-time terminal sliding mode (FTTSM) containing a tunable parameter. Then, we upgrade the proposed controller to a fully adaptive-gain anti-saturation version, in order to strengthen the robustness and adaptivity with respect to actuator faults and failures, unknown mass properties, and external disturbances. In the two schemes, all of the designed adaptive parameters are scalars, thus they only require light computational load and can avoid the redesign process of the controller during spacecraft operation. Finally, the feasibility of the proposed methods is illustrated via two numerical examples.

Motion/force control scheme for electrically driven cooperative multiple mobile manipulators

This paper presents the motion and force control problem of rigid-link electrically driven cooperative mobile manipulators handling a rigid object. Although, the motion/force control problem of cooperative mobile manipulators has been enthusiastically studied. But there is little research on the motion/force control of electrically driven cooperative mobile manipulators. Due to the inclusion of the actuator dynamics with the manipulator’s dynamics, the controller exhibits some important characteristics. For the electromechanical system, we have designed a novel controller at the dynamic level as well as at the actuator level. In the proposed control scheme, at the dynamic level, uncertain non-linear mechanical dynamics is approximated with a hybrid controller containing model-based control scheme combined with model-free neural network based control scheme together with an adaptive bound. The adaptive bound is used to suppress the effects of external disturbances, friction terms, and reconstruction error of the neural network. At the actuator level, for the approximation of the unknown electrical dynamics, the model-free neural network is utilized. The developed control scheme provides that the position tracking errors, as well as the internal force, converge to the desired levels. Additionally, direct current motors are also controlled in such a way that the desired currents and torques can be attained. In order to make the overall system to be asymptotically stable, online learning of the weights and the parameter adaptation of the parameters is utilized in the Lyapunov function. The superiority of the developed control method is carried out with the numerical simulation results and its superior robustness is shown in a comparative manner.

Design and robustness analysis of autonomous vehicles in intersections

Abstract The paper proposes the design of a neural-network-based control strategy of autonomous vehicles in intersections. The motivation of the neural network approach is to reduce the numerically-intensive computation of the optimization problem in which the motions of autonomous vehicles are formed. In the method the neural network is trained through a preliminary optimal off-line solution. Moreover, a robustness analysis on the neural network based control strategy is proposed. The focus of the analysis is to consider the impact of the position and speed estimation errors on the motion of vehicles. The design and the analysis are illustrated through CarSim simulation examples.

Artificial neural network-based control strategies for PMSG-based grid connected wind energy conversion system

This paper presents a novel control strategy employing the artificial neural network-based direct torque control (DTC) strategy and direct power control (DPC) strategy for permanent magnet synchronous generator (PMSG)-based wind energy conversion system (WECS). The Aeolos 3 kW wind system ratings are considered in this paper for real time system validation. The proposed control strategy combines artificial neural network (ANN)-based DTC as machine side control (MSC) strategy and ANN-based DPC as grid side control (GSC) strategy in order to track the maximum power and to regulate the active power, respectively. To generate the switching pulse for the converter in generator side and grid side, a neural network-based technique is employed. Furthermore, the DC link voltage is optimised using the GSC topology. The proposed controller provides a stable voltage and power which is an effective solution for grid integration and constant power flow from the generator system to the grid system. The proposed coordinate control strategy is validated using MATLAB/Simulink software to analysis the working condition. The experimental test bench is implemented using dSPACE to study the real time analysis of the proposed control strategy.

Artificial neural network-based control strategies for PMSG-based grid connected wind energy conversion system

This paper presents a novel control strategy employing the artificial neural network-based direct torque control (DTC) strategy and direct power control (DPC) strategy for permanent magnet synchronous generator (PMSG)-based wind energy conversion system (WECS). The Aeolos 3 kW wind system ratings are considered in this paper for real time system validation. The proposed control strategy combines artificial neural network (ANN)-based DTC as machine side control (MSC) strategy and ANN-based DPC as grid side control (GSC) strategy in order to track the maximum power and to regulate the active power, respectively. To generate the switching pulse for the converter in generator side and grid side, a neural network-based technique is employed. Furthermore, the DC link voltage is optimised using the GSC topology. The proposed controller provides a stable voltage and power which is an effective solution for grid integration and constant power flow from the generator system to the grid system. The proposed coordinate control strategy is validated using MATLAB/Simulink software to analysis the working condition. The experimental test bench is implemented using dSPACE to study the real time analysis of the proposed control strategy.

Robust neural network–based tracking control and stabilization of a wheeled mobile robot with input saturation

SummaryThis paper presents a robust neural network–based control scheme to deal with the problem of tracking and stabilization simultaneously for a wheeled mobile robot subject to parametric uncertainties, external disturbances, and input saturation. At first, a new error‐state transformation scheme is designed by introducing some auxiliary variables as an additional virtual control signals to reduce the adverse effect caused by the underactuation. These variables can change their structures for different desired trajectories to be tracked. Then, a robust control law is proposed combining with a kinematic controller and a dynamic controller, while a three‐layer neural network system is applied to approximate model uncertainties. Stability analysis via the Lyapunov theory shows that the proposed controller can make tracking errors converge to bounded neighborhoods of the origin. Finally, some simulation results are illustrated to verify the effectiveness of the proposed control strategy.

Neural Network Based ILST Control Strategy for a DSTATCOM with Solar Photovoltaic System for Power Quality Improvement

The power quality compensator chosen in this paper is a DSTATCOM which integrates a three phase four leg Voltage Source Converter (VSC) with a DC capacitor (Cdc). A single stage Solar Photovoltaic (SPV) system is implemented for maintaining the DC link voltage (Vdc) of the Shunt Active Power Filter (SAPF). The system proposed, improves the Power Quality (PQ) of the utility system as well as feeds the solar energy extracted into the system. Maximum Power Point Tracking (MPPT) is implemented to excerpt solar energy. The purpose of the proposed system is to mitigate harmonics and neutral current (Isn),to compensate reactive power and to maintain power factor near unity. The SAPF improves the PQ of the proposed system. The control algorithm proposed for SAPF is Neural Network (NN) based Improved Linear Sinusoidal Tracer (ILST) algorithm. An instantaneous compensation technique controls the variations in PV power of the SPV system with fast dynamic response. The proposed system is also compared with battery operated Boost Converter (BC) SPV fed SAPF system. The battery operated SPV system boost the DC link voltage and thus improves PQ throughout the day. The NN based control strategy and the SPV system are modeled, developed and validated in MATLAB SIMULINK. The simulation results justify the effectiveness of the propounded system.

Power Management Comparison for a Dual-Motor-Propulsion System Used in a Battery Electric Bus

The efficiency performance of multi-motor-driven system highly depends on the power management. Three aspects of contribution have been made in this study. 1) A predictive power management for a DMPS is developed. To improve the performance of the predictive power management, an adaptive velocity predictor is proposed and the coefficients of proposed predictor can update its parameters according to identified driving patterns. Simulation results show that the new velocity predictor have best prediction performance compared with traditional predictors. 2) A neural network based power management is proposed. According to the optimization results of dynamic programming, radial-basis-function neural network is trained. The input dimensions and the number of hidden layer neurons of the neural network are optimized. 3) The performance of proposed control strategies are compared with three different drive cycles including MANHATTAN cycle, Japanese 1015 cycle, and UDDSHDV cycle. Simulation results indicate that compared with original control strategy, the predictive control strategy and neural network based control strategy show better efficiency performance. The neural network based strategy is verified by hardware-in-loop experiment and experiment results indicate that the control performance in real hardware shows similar property with simulation results.

Adaptive Neural Network Control for the Trajectory Tracking of the Furuta Pendulum.

The purpose of this paper is to introduce a novel adaptive neural network-based control scheme for the Furuta pendulum, which is a two degree-of-freedom underactuated system. Adaptation laws for the input and output weights are also provided. The proposed controller is able to guarantee tracking of a reference signal for the arm while the pendulum remains in the upright position. The key aspect of the derivation of the controller is the definition of an output function that depends on the position and velocity errors. The internal and external dynamics are rigorously analyzed, thereby proving the uniform ultimate boundedness of the error trajectories. By using real-time experiments, the new scheme is compared with other control methodologies, therein demonstrating the improved performance of the proposed adaptive algorithm.