Neural Network Controller Research Articles

Dynamics and Stabilization of Chaotic Monetary System Using Radial Basis Function Neural Network Control

In this paper, we investigated a three-dimensional chaotic system that models key aspects of a monetary system, including interest rates, investment demand, and price levels. The proposed system is described by a set of autonomous quadratic ordinary differential equations. We analyze the dynamic behavior of this system through equilibrium points and their stability, Lyapunov exponents (LEs), and bifurcation diagrams. The system demonstrates a variety of behaviors, including chaotic, periodic, and equilibrium states depending on parameter values. Additionally, we explore the multistability of the system and present a radial basis function neural network (RBFNN) controller design to stabilize the chaotic behavior. The effectiveness of the controller is validated through numerical simulations, highlighting its potential applications in economic and financial modeling.

Dynamic event-triggered attitude synchronization of multi-spacecraft formation via a learning Chebyshev neural network control approach

Purpose This paper aims to investigate the attitude synchronization issue of multi-spacecraft formation flying systems under the limited communication resources. Design/methodology/approach The authors propose a distributed learning Chebyshev neural network controller (LCNNC) combining a dynamic event-triggered (DET) mechanism and a learning CNN model to achieve accurate multi-spacecraft attitude synchronization under communication constraints. Findings The proposed method can significantly reduce the internal communication frequency and improve the attitude synchronization accuracy. Practical implications This method requires the low communication resources, has a high control accuracy and is thus suitable for engineering applications. Originality/value A novel DET mechanism-based LCNNC is proposed to achieve the accurate multi-spacecraft attitude synchronization under communication constraints.

Digital Twin-Based Smart Feeding System Design for Machine Tools

With the continuous development of intelligent manufacturing technology, the application of intelligent feed systems in modern machine tools is becoming increasingly widespread. Digital twin technology achieves the monitoring and optimization of the entire life cycle of a physical system by constructing a virtual image of the system, while neural network controllers, with their powerful nonlinear fitting ability, can accurately capture and simulate the dynamic behavior of complex systems, providing strong support for the optimization control of intelligent feed systems. This article discusses the design and implementation of an intelligent feed system based on digital twins and neural network controllers. Firstly, this article establishes a mathematical model based on the traditional ball screw structure and analyzes the dynamic characteristics and operating mechanism of the system. Subsequently, the mathematical model is fitted using a neural network controller to improve control accuracy and system response speed. The experimental results demonstrate that the neural network controller shows good consistency in fitting traditional mathematical models, not only effectively capturing the nonlinear characteristics of the system but also maintaining stable control performance under complex operating conditions.

Variable universe fuzzy inference decided neural network feed-forward compensation for pid control of motor position

To improve the compensation accuracy of neural network decided by traditional fuzzy inference engine (FIE), a variable universe fuzzy inference (VUFI) decided neural network feed-forward compensation method for PID control of motor position is proposed. The method contains four control blocks: fuzzy inference (FI) block, variable universe (VU) block, neural network (NN) control block, and basic control block. The FI block adaptively decides the feed-forward control quantity of NN based on the error and the error change rate of control system to keep the dynamic performance of NN at the beginning of the control or the transient jump signal. The VU block constructs the contraction expansion (CE) factors from the absolute change quantity of NN weights and adjusts the universes of membership functions based on the learning situation of NN in real time so that high-precision inference can be achieved. The NN control block consists of a neural network identifier (NNI) and a neural network controller (NNC) that has the same structure with the NNI, and the NNI dynamically transfers network parameters to the NNC and outputs feed-forward compensation control quantity after learning the dynamic inverse model of DC servo motor online. The basic control block adopts a PID controller, which provides the learning samples for NN online. Experimental results proved that the proposed method can improve the inference accuracy of FIE without sacrificing steady-state accuracy, significantly reduce overshoot, and have better stability and dynamic performance.

Implementation and Performance Evaluation of Intelligent Techniques for Controlling a Pressurized Water Reactor

Pressurized water reactors (PWRs) are the most common and widely used type of reactor, and ensuring the stability of the reactor is of utmost importance. The challenges lie in effectively managing power fluctuations and sudden changes in reactivity that could result in unsafe situations. Reactor power fluctuations can cause changes in behavior. At the same time, the transfer of heat from the fuel to the coolant and reactivity changes resulting from differences in fuel and coolant temperatures can also make the systemunpredictable. The primary goal of a power controller used in a nuclear reactor is to sustain the specified power level, which guarantees the security of the power plant. To address these challenges, this paper presents a dynamic model of a PWR and applies several control techniques to the system for power level control. Specifically, a traditional PID controller, a Neural Network controller, a Fuzzy self-tuned PID controller, and a Neuro-Fuzzy self-tuned PID controller were individually designed and evaluated to enhance the performance of the reactor power control system under constant and variable reference power. In addition, the robustness of each controller was appraised against time delays and external disturbances. The system was tested with various initial power values to evaluate its performance under different conditions. The results demonstrate that the Neuro-Fuzzy self-tuned PID controller has the best performance, as well as the fastest response time compared to the other controllers.Furthermore, the intelligent controllers were found to exhibit good robustness against time delays and external disturbances. The system's stability was not significantly affected by changes in the initial power value, although it had a minor effect on the transient response. Overall,the findings of this study can inform the design and optimization of control systems for PWRs, with the ultimate goal of improving their safety, reliability, and performance.

Neural network controller for hybrid energy management system applied to electric vehicles

Hybrid energy storage systems based on batteries and supercapacitors can mitigate the aging of electric vehicle batteries aging by avoiding high currents and rapid discharge cycles. This system requires energy management systems that efficiently split the power during real driving cycles. This paper outlines a design methodology for creating a Multilayer Perceptron neural controller that governs the power distribution between the storage system components. In parallel, a Proportional-Integral-Derivative controller was implemented to ensure voltage regulation in the primary DC bus, ensuring stable operation of the entire system and providing flexibility to the neural design. The controller was tuned using particle swarm optimization, hippopotamus optimization, and differential evolution algorithms, designed to minimize the battery root-mean-square current in a comprehensive vehicle simulation. The training was structured into layers to approach the physical problem and controller optimization facets. The controller’s primary goal is to minimize the battery strain, mitigating stress events and prolonging its lifespan. The energy management neural controller was designed using a simple drive cycle and later validated with a realistic cycle. The findings demonstrate the feasibility of achieving a significant reduction in current, both in peak value and on average, especially when compared to basic energy management strategies. The process uncovered a strategy that prioritizes the use of the supercapacitor in power balance, with this effect being more pronounced during critical load events. The main bus voltage remained stable, with no deviations exceeding 5%, owing to the voltage regulator’s stability. Additionally, the particle swarm optimization and the hippopotamus optimization techniques exhibited notable performance compared to the differential evolution algorithm for this problem. Subsequently, the design was evaluated in a more complex cycle, revealing a slight decrease in average battery current performance. However, it still maintained a significant reduction in peak battery current compared to traditional controllers. Nevertheless, this methodology demonstrated power management optimization efficacy and can be extended to more complex scenarios.

IoT enabled microgrid framework using a novel dispersal diffusion artificial neural network controller for PV systems and wind energy to minimize electrical faults

IoT enabled microgrid framework using a novel dispersal diffusion artificial neural network controller for PV systems and wind energy to minimize electrical faults

Adaptive NN Control for a Flexible Manipulator With Input Backlash and Output Constraint

Adaptive NN Control for a Flexible Manipulator With Input Backlash and Output Constraint

Hybrid energy management strategy based on neural networks robust to fluctuations in operational load

As global regulations on carbon neutrality tighten, the maritime industry, led by the IMO, has set a goal of achieving net-zero emissions by 2050. The electrification of ship propulsion systems is consequently accelerating. Hybrid electric propulsion ships, which combine internal combustion engines with auxiliary power sources, offer a practical solution by improving fuel efficiency and reducing emissions. However, heuristic or pre-trained energy management strategies designed for fixed conditions often struggle under fluctuating loads. This study addresses these limitations by proposing a hybrid energy management strategy that generates probabilistic operational profiles based on representative load data. A neural network controller, trained on nine global control solutions for newly generated operation cycles using Markov chain methods, maintains approximately 37% engine thermal efficiency even under variable loads. A performance validation showed that the proposed strategy resulted in 0.56% higher fuel consumption compared to the global solution, while still supporting implementation with each time step calculation taking approximately 0.072 ms, achieving near real-time performance. This small deviation highlights the trade-off between time-efficient applicability and near-optimal efficiency. By overcoming the limitations of fixed-condition strategies, this research offers a robust and adaptive approach for hybrid propulsion systems, making it suitable for immediate use in eco-friendly ships.

Power Quality Improvement of PV fed Grid Connected System using ANN Controlled Shunt Active Power Filter

Harmonics are common in integrated power systems, especially with the increasing use of nonlinear loads (NLL), such as those found in photo voltaic (PV) systems connected to the grid. Traditional LC filters Shunt active power filters (SAPF) have been developed to effectively correct harmonics and improve power quality performance. This study presents a three-phase voltage-fed SAPF implementation to mitigate harmonics using an artificial neural network (ANN) controller. The SAPF control system focuses on generating reference source currents to counterbalance the harmonic effects caused by NLL. The model's effectiveness is validated using experimental data gathered from a nonlinear load through MATLAB/Simulink simulations.

Data-driven control of echo state-based recurrent neural networks with robust stability guarantees

In this work we propose a new data-based approach for robust controller design for a rather general class of recurrent neural networks affected by bounded measurement noise. We first identify the model set compatible with available data in a selected model class via set membership (SM). Then, incremental input-to-state stability and desired performances for the closed loop system are enforced robustly to all models in the identified model set via a linear matrix inequality (LMI) optimization problem. Numerical results show the effectiveness of the comprehensive method.

Performance Analysis of DC Motors With Integrated Proportional-Integral and Artificial Neural Network Control

Direct current (DC) motors are frequently utilized in various applications, and the motor's pace is affected by applied loads as it fluctuates. A power converter must be employed to control the velocity of the motor by varying the armature voltage. One of the options for the power converter is the one-quadrant DC chopper. In this case, the investigation will turn the one-quadrant chopper into a system by merging velocity and current control into the DC motor. The speed is regulated by controlling the armature voltage. This may be accomplished using a controlled rectifier. The contribution of the research is to test the effectiveness of Artificial Neural Network Control (ANN) and Proportional-Integral (PI) controllers to control the speed of a DC motor using a one-quadrant DC chopper. Therefore, due to technological advancements, the authors will utilize the training data of the artificial neural network of Proportional- Integral controllers in MATLAB's Simulink. Test results demonstrate the artificial neural network (ANN's) superior ability to regulate system response, showing enhancements in delay time, rise time, overshoot, and steady-state error compared to the PI controller. These findings underscore the potential of ANN as a more sophisticated choice for DC motor control, although further research is required to finetune its performance through rigorous training.

Analysis and control of hybrid convection-radiation drying systems toward energy saving strategy.

Food drying is a vital technology in agricultural preservation, where hybrid drying provides efficient moisture removal. However, the transient behavior of the moisture ratio ( ) and associated drying rate ( ) could increase energy consumption if not managed with an effective energy-saving strategy. This study investigates the impacts of hybrid convection-radiation drying on , , drying time, and energy consumption. Unlike previous studies, which have not fully addressed the connections between operating conditions and drying kinetics, this work provides new insights into these relationships. Additionally, an artificial neural network (ANN) control model is applied to optimize energy consumption, offering a new approach to improving the efficiency of the drying process. During the experiment, the airflow velocity was varied from 0.7m/s to 1.5m/s, the airflow temperature was varied from 40°C to 60°C, and the radiation intensity was varied from 1500W/m2 to 3000W/m2. The results showed that the transient behavior exhibited four data groups with consistent and through the mean and statistical analysis. Increasing radiation intensity and air temperature has decreased the drying time, while higher airflow has increased the drying time. The energy indices were enhanced by increasing radiation intensity and temperature while reducing airflow velocity. The measured of all groups exhibited similar kinetics behavior, while the associated exhibited similar clustering through the self-organizing map. Those findings were further controlled using an ANN model with 99% predicting accuracy. With airborne heating at 60°C and airflow at 0.7m/s, the radiation intensity is transiently controlled, showing a 6.5% drying time reduction and a 36% energy saving. Thus, controlling the radiation intensity and its impact on the slice properties is highly desirable in future work. This work could help designers improve the processing efficiency and energy conservation of garlic slices drying.

Improvement of the Transient Levitation Response of a Magnetic Levitation System Using Hybrid Fuzzy and Artificial Neural Network Control

Improvement of the Transient Levitation Response of a Magnetic Levitation System Using Hybrid Fuzzy and Artificial Neural Network Control

Adaptive NN Force Loading Control of Electro-Hydraulic Load Simulator

To address the issues of derivative explosion in traditional backstepping control and the strong nonlinearity of hydraulic systems, this paper develops an adaptive neural network control method tailored for electro-hydraulic load simulators. Neural networks are employed to handle external disturbances, modeling uncertainties, and the derivatives of virtual control inputs. First, the precise state-space equations of the system are derived. Next, the approximation property of neural networks is used to design an adaptive backstepping controller, and the symmetric barrier Lyapunov function is used to prove the boundedness of the controller and control parameters. Finally, experiments are conducted to verify the effectiveness and reliability of the control algorithm. The results demonstrate that the proposed control algorithm exhibits excellent tracking performance and effectively reduces control errors.

Analysis of Three Phase Induction Motor Settings To Load Changes Using Artificial Neural Networks

benefits of induction motors in the industrial world today are numerous. This is because this induction motor has many advantages such as strong and simple construction and is cheaper to maintain. However, there are weaknesses in regulating speed and torque. One of the 3-phase induction motor speed regulation is vector control. This method is connected to a PID controller to control the input and output to reach the expected value. However, with the development of an increasingly modern era, it can now be done automatically, especially in regulating PID control, in this experiment, it wants to be done using an artificial neural network controller to be able to study the data that has been done by the PID controller to find out the input and output expected by the system so that the speed control on the induction motor runs stably. this method is able to control the speed to remain stable even though the load changes. In this study, loading tests will be carried out at 0 Nm, 80 Nm, 140 Nm, 290 Nm and 400 Nm and reference speeds of 40 rps, 75 rps and 120 rps. The simulation results of using the artificial neural network controller show that the system can reach the expected reference speed in less than 3 ms and the average steady state error is 0.5241%.

Adaptive force/position NN control of asymmetric teleopertion system based on fuzzy CMAC

Adaptive force/position NN control of asymmetric teleopertion system based on fuzzy CMAC

A hybrid algorithm for photovoltaic maximum power point tracking using Artificial Neural Network and Kinetic Gas Molecular Optimization

The quest for efficient photovoltaic (PV) energy conversion systems has led to the development of Maximum Power Point Tracking (MPPT) algorithms. In this paper, we propose a novel hybrid approach that combines the power of Artificial Neural Network (ANN) with the optimization capability of Kinetic Gas Molecular Optimization (KGMO) for MPPT. We present a high-level outline of the proposed algorithm along with example MATLAB code snippets for each step, highlighting its potential for improving PV system performance. This paper proposes a metaheuristic optimized multilayer feed‐forward artificial neural network (ANN) controller to extract the maximum power from available solar energy. Firstly, to improve the maximum power point (MPP) delivered by PV arrays and to overcome the drawbacks in the conventional MPPT method under irradiation variation, a hybrid MPPT controller is designed, in which the input parameters include the PV array voltage and current. The output parameter is the duty cycle of the DC/DC boost converter. The proposed approach abbreviated as ANN-KGMO MPPT controller is based on the Kinetic Gas Molecular Optimization (KGMO) Algorithm which is useful to train the developed ANN and to evolve the connection weights and biases to get the optimal values of duty cycle converter corresponding to the MPP of a PV array. Finally, the performance of the proposed control system is confirmed by simulation tests on a 2 kW PV system. In addition, the performance of an ANN-KGMO -based MPPT controller is also compared to the conventional perturb and observe (P&O) method. To analyze the results, simulations are performed by using MATLAB software.

Event‐Triggered Adaptive Neural Network Control for 4WS4WD Wheeled Mobile Robot

ABSTRACTIn this article, an event‐triggered adaptive neural network controller based on threshold band is designed for a four wheels independently steered and four wheels independently driven (4WS4WD) mobile robot. The 4WS4WD mobile robot is attracting attention for its excellent motion performance such as manipulation versatility and posture flexibility. However, its control difficulty is increased due to its characteristics of being controlled by eight motors for steering and driving respectively. Also, since the robot itself has limited computing and communication resources, real‐time control cannot be guaranteed. To copy with that, the kinematic and dynamics models are first introduced for the 4WS4WD mobile robot. Second, an adaptive neural network controller with low‐frequency learning rate is utilized to control the mobile robot since there are unknown perturbations in the model. It can maintain system stability while handing unidentified model perturbations. The stability of the controller is demonstrated by Lyapunov stability analysis. An event‐triggered based on threshold band is suggested to lessen the amount of computation in the control process. Finally, the simulation outcomes further demonstrate how the suggested approach can greatly lessen the computational and communication cost while maintaining control performance.

Direct adaptive neural control for precision piezoelectric actuator without hysteresis compensation

This paper introduces a novel Direct Adaptive Neural Control (DANC) approach to enhance the precision of piezoelectric actuator motion control significantly. By effectively addressing the complexities of hysteresis, DANC simplifies system design while delivering superior performance. A Radial Basis Function Neural Network (RBFNN) is employed to handle system uncertainties robustly, and a switching control mechanism guarantees stability, rigorously verified through Lyapunov analysis. Experimental results using a Thorlabs PZS001 actuator convincingly demonstrate substantial performance improvements over conventional PID control, the feedforward-based neural network and feedback control (FF-FB control).