Neural Compensation Research Articles

Artificial neural network-based power quality compensator

A reliable and efficient adaptive neural network-based active power filter to estimate and compensate harmonic distortion from supply mains is presented in this paper. Nowadays, there is drastic rise of current and voltage harmonics in power systems, caused by nonlinear loads. Active power filters (APF) are used to mitigate harmonics and thereby, improve power quality. This paper deals with application of artificial neural network in shunt active power filter which provides ease in implementation and fast dynamic response compared to conventional active power filters. In depth analysis of neural network applications in the intelligent control and estimation for power quality compensation is presented in this paper. Here, both reference compensating current generation scheme as well as current controller for the active power filter are developed using artificial neural network technique. Effective compensation provided by the proposed artificial neural network-based shunt active power filter is proved through simulation results. Experimental analysis carried out using dSPACE DS1104 also validates power quality improvement by the proposed APF scheme.

Implementation of an Adaptive Neural Terminal Sliding Mode for Tracking Control of Magnetic Levitation Systems

In this article, an adaptive neural terminal sliding mode is implemented for tracking control of magnetic levitation systems with the presence of dynamical uncertainty and exterior perturbation. By proposing a novel fast terminal sliding manifold function with the dynamic coefficients, the system state variables quickly converge the equilibrium point on the manifold function. Besides, an adaptive, robust reaching control law combined with radial basis function neural network compensator drives the system fast approaching the sliding manifold function regardless of whether the initial value is near or far from the sliding manifold and reduces the chattering of the conventional terminal sliding mode control. With a design approach based on the combination of the proposed sliding manifold and the combined control law, the implemented control method provides a control performance with significant improvement in the terms of chattering reduction, high tracking accuracy, fast convergence along with simple design for real applications. The experimental work is implemented for a real magnetic levitation system to demonstrate the superior efficiency of the proposed terminal sliding mode control. The stable evidence of the proposed method is also completely verified by Lyapunov-based method.

Classification of Depression with Brain Network Characteristics Based on Multiphase Map Deep Neural Network Equilibrium Compensation

Classification of Depression with Brain Network Characteristics Based on Multiphase Map Deep Neural Network Equilibrium Compensation

Three-dimensional mandibular motion trajectory-tracking system based on BP neural network.

The aim of this study was to develop a prototype three-dimensional optical motion capture system based on binocular stereo vision, Back-propagation (BP) Neural Network and 3D compen-sation method for accurate and real-time recording of mandibular movement. A specialized 3D method of compensation to eliminate the involuntary vibration motions by human heart beating and respiration. A kind of binocular visual 3D measurement method based on projection line and a calibration method based on BP neural network is proposed to solve the problem of the high complexity of camera calibration process and the low accuracy of 3D measurement. The accuracy of the proposed system is systematically evaluated by means of electric platform and clinical trials, and the root-mean-square is 0.0773 mm. Finally, comparisons with state-of-the-art methods demonstrate that our system has higher reliability and accuracy. Meanwhile, the motion trajectory-tracking system is expected to be used in the diagnosis of clinical oral diseases and digital design of restoration.

Deep GRU Neural-Network Prediction and Feedforward Compensation for Precision Multi-Axis Motion Control Systems

This article proposes a gated recurrent unit (GRU) neural network prediction and compensation (NNC) strategy for precision multiaxis motion control systems with contouring performance orientation. First, some characteristic contouring tasks are carried out on a multiaxis linear-motor-driven motion system, and the true contouring error values obtained by the Newton numerical calculation method are used as the training data of a developed artificial GRU neural network. Essentially, the proposed GRU neural network structure can be viewed as a data-based black-box error model, which can capture the dynamic characteristics of contouring motion rather accurately. The well-trained GRU network can predict the contouring error precisely even under the tasks those have not been conducted during the training session. Moreover, the predicted contouring error is compensated into the reference contour as feedforward compensation to improve the final contouring performance. Comparison between the predicted contouring error and the actual contouring error practically proves the effective prediction ability of the proposed GRU neural network. Furthermore, comparative experiments among proportional–integral–differential, iterative learning control (ILC), and the proposed NNC controller are conducted. The results consistently validate that NNC can basically achieve excellent contouring motion performance as ILC, significantly without need of motion repetition and iteration. Due to the implementation convenience and excellent prediction/compensation ability, the proposed NNC would have good potential in industrial mechatronic applications.

Artificial neural network-based power quality compensator

Artificial neural network-based power quality compensator

Neural anomalies during vigilance in schizophrenia: Diagnostic specificity and genetic associations

Impaired vigilance is a core cognitive deficit in schizophrenia and may serve as an endophenotype (i.e., mark genetic liability). We used a continuous performance task with perceptually degraded stimuli in schizophrenia patients (N = 48), bipolar disorder patients (N = 26), first-degree biological relatives of schizophrenia patients (N = 55) and bipolar disorder patients (N = 28), as well as healthy controls (N = 68) to clarify whether previously reported vigilance deficits and abnormal neural functions were indicative of genetic liability for schizophrenia as opposed to a generalized liability for severe psychopathology. We also examined variation in the Catechol-O-methyltransferase gene to evaluate whether brain responses were related to genetic variation associated with higher-order cognition. Relatives of schizophrenia patients had an increased rate of misidentification of nontarget stimuli as targets when they were perceptually similar, suggestive of difficulties with contour perception. Larger early visual responses (i.e., N1) were associated with better task performance in patients with schizophrenia consistent with enhanced N1 responses reflecting beneficial neural compensation. Additionally, reduced N2 augmentation to target stimuli was specific to schizophrenia. Both patients with schizophrenia and first-degree relatives displayed reduced late cognitive responses (P3b) that predicted worse performance. First-degree relatives of bipolar patients exhibited performance deficits, and displayed aberrant neural responses that were milder than individuals with liability for schizophrenia and dependent on sex. Variation in the Catechol-O-methyltransferase gene was differentially associated with P3b in schizophrenia and bipolar groups. Poor vigilance in schizophrenia is specifically predicted by a failure to enhance early visual responses, weak augmentation of mid-latency brain responses to targets, and limited engagement of late cognitive responses that may be tied to genetic variation associated with prefrontal dopaminergic availability. Experimental results illustrate specific neural functions that distinguish schizophrenia from bipolar disorder and provides evidence for a putative endophenotype that differentiates genetic liability for schizophrenia from severe mental illness more broadly.

Increased Activation of Default Mode Network in Early Parkinson's With Excessive Daytime Sleepiness.

Background and ObjectivesThe underlying neuropathology of excessive daytime sleepiness (EDS) remains elusive in Parkinson’s disease (PD). We aim to investigate neural network changes that underlie EDS in PD.MethodsEarly PD patients comprising eighty-one patients without EDS (EDS−) and seventeen patients with EDS (EDS+) received a resting state functional MRI scan and the Epworth Sleepiness Scale (ESS). Connectivities within the default mode network (DMN), motor and basal ganglia networks were compared between the EDS+ and EDS− groups. Correlations between network connectivity and the severity of EDS were investigated through linear regression.ResultsEDS+ patients displayed a trend of increased network connectivity of the posterior DMN (pDMN). A significant positive correlation was found between connectivity of the ventromedial prefrontal cortex in the pDMN and ESS.ConclusionEDS+ patients are likely to display increased activation in the DMN, suggesting neural compensation in early PD or impaired attentiveness due to mechanisms such as mind-wandering.

Adaptive control of micro-electro-mechanical system gyroscope using neural network compensator

This article proposes an adaptive control scheme with a neural network compensator for controlling a micro-electro-mechanical system gyroscope with disturbance and model errors. The adaptive neural network compensator is used to compensate the nonlinearities in the system based on its universal approximation and improve tracking performance of the gyroscope. The neural compensator, which is trained online, is combined with adaptive control of the Lyapunov framework system to approach the unknown system disturbance and model errors. The system stability is deduced by the Lyapunov stability theory, and the simulation of the micro-electro-mechanical system gyroscope is carried out on Matlab/Simulink, verifying the superior performance of the neural control compensation method.

Assessing Neural Compensation With Visuospatial Working Memory Load Using Near-Infrared Imaging.

Alzheimer's disease is characterized by the progressive deterioration of cognitive abilities particularly working memory while mild cognitive impairment (MCI) represents its prodrome. It is generally believed that neural compensation is intact in MCI but absent in Alzheimer's disease. This study investigated the effects of increasing task load as a means to induce neural compensation through a novel visual working memory (VSWM) task using functional near-infrared spectroscopy (fNIRS). The bilateral prefrontal cortex (PFC) was explored due to its relevance in VSWM and neural compensation. A total of 31 healthy controls (HC), 12 patients with MCI and 18 patients with mild Alzheimer's disease (mAD) were recruited. Although all groups showed sensitivity in terms of behavioral performance (i.e. score) towards increasing task load (level 1 to 3), only in MCI load effect on cortical response (as measured by fNIRS) was significant. At lower task load, bilateral PFC activation did not differ between MCI and HC. Neural compensation in the form of hyperactivation was only noticeable in MCI with a moderate task load. Lack of hyperactivation in mAD, coupled with significantly poorer task performance across task loads, suggested the inability to compensate due to a greater degree of neurodegeneration. Our findings provided an insight into the interaction of cognitive load theory and neural compensatory mechanisms. The experiment results demonstrated the feasibility of inducing neural compensation with the proposed VSWM task at the right amount of cognitive load. This may provide a promising avenue to develop an effective cognitive training and rehabilitation for dementia population.

Adaptive Neural Output Feedback Compensation Control for Intermittent Actuator Faults Using Command-Filtered Backstepping.

Effectively compensating unknown intermittent actuator faults in uncertain decentralized nonlinear systems is a very difficult problem, and very few results have been obtained. In this article, to address this issue, an adaptive neural output feedback compensation control scheme based on command-filtered backstepping is developed. First, we design a bank of observers to estimate the system states and utilize neural networks with random hidden nodes to approximate the unknown functions of these observers. Second, a smooth projection algorithm is used to online update estimated parameters in the controllers such that the possible ceaseless increase in the estimated parameters caused by intermittent actuator faults can be eliminated. Due to the presence of intermittent jumps of unknown parameters, a modified Lyapunov function is developed to analyze the system stability. It is proved that the boundedness of all closed-loop system signals is ensured and the ultimate bound of the tracking error depends on design parameters, adjustable jumping amplitude of Lyapunov function, and minimum fault time interval. Third, by analyzing the system transient performance, the peaking phenomenon at the starting instant of the system operation can be removed, and a root mean square type of bound is established to illustrate that the transient tracking error performance is tunable by design parameters. Finally, simulations studies are done to illustrate the effectiveness of the theoretical results.

Design of Contour Error Coupling Controller Based on Neural Network Friction Compensation

In two-axis servo contour motion control, friction and various uncertainties unavoidably exist, reducing the contour control accuracy. This paper proposes a neural network contour error coupling control method based on LuGre friction compensation, which includes a contour error calculation model, single-axis computed torque controller (CTC), and neural network friction compensation controller. The LuGre friction model can describe servo system’s complicated static and dynamic friction characteristics, and the RBF neural network has a universal approximation property to realize compensation control of friction. Simulation results indicate that the proposed contour error control method can effectively compensate for the effect of friction and improve contour control accuracy.

Adaptive block compensation trajectory tracking control based on LuGre friction model

Aiming at the problems of modeling error and uncertain external disturbance in the multi-joint robot control model, an adaptive block compensation trajectory tracking controller based on LuGre friction model is proposed. Firstly, the algorithm divides the interference term of LuGre friction model into three parts with different physical quantities. Secondly, an adaptive neural network compensator is designed to assess the three parts of the LuGre friction model. Thirdly, a robust sliding mode controller is developed to reduce the influence of these estimation errors of neural network compensator and other uncertain disturbances and ensure that the system converges in a finite time at the same time. Finally, numerical simulations under different input and disturbance signals for the planar multi-joint robot and the inverted pendulum are conducted to validate the effectiveness of the proposed controller, and the performance of the proposed controller is compared with conventional sliding mode controller to illustrate the usefulness and efficiency of the proposed controller.

Time-delay estimation based computed torque control with robust adaptive RBF neural network compensator for a rehabilitation exoskeleton

A new approach to gait rehabilitation task of a 12 DOF lower limb exoskeleton is proposed combining time-delay estimation (TDE) based computed torque control (CTC) and robust adaptive RBF neural networks. In addition to the conventional advantages of the CTC, TDE technique is integrated to estimate unmodeled dynamics and external disturbance. To realize more accurate tracking, a robust adaptive RBF neural networks compensator is designed to approximate and compensate TDE error. The final asymptotic stability is guaranteed with Lyapunov criteria. To validate the proposed approach, co-simulation experiments are realized using SolidWorks, SimMechanics and MATLAB/Robotics Toolbox. Compared to CTC, sliding mode based CTC and TDE based CTC, the higher performances of the proposed controller are demonstrated by co-simulation.

B-03 Neurocognitive Outcome from Anaplastic Astrocytoma: A Case Study of Neural Compensation

Abstract Objective Anaplastic astrocytoma is a rare malignant brain tumor occurring in adults aged 30 to 50, with an incidence of 5 to 8 people per 100,000. Cognitive dysfunction is a complication after tumor resection, resulting from neurotrauma of surgery, chemotherapy, and radiation treatment. Neural compensation and neuroplasticity can allow intact cognitive functioning after brain resection in brain areas expected to be impacted by the neurotrauma. Method The current case study is a 45-year-old female post -resection for dendritic cell vaccine clinical trial followed by adjuvant radiation of an anaplastic astrocytoma of the frontal lobe. MRI of the brain post right frontal craniotomy was indicative of extra-axial fluid communicating with the anterior aspect of the right lateral ventricle, hemosiderin along the medial resection margin, and T2/FLAIR signal abnormality in the left medial and right inferior frontal lobes. Results Test data revealed intact global cognitive functioning (WAIS-IV; FSIQ = 101 SS), with below expected performance on verbal comprehension (WAIS-IV; VCI = 98 SS). Consistent with the area of resection, she had impairments in visuoconstruction and executive functioning specific to mental generativity. However, performances with regard to math achievement were intact (WIAT; Math Problem Solving = 110 SS; Numerical Operations = 104 SS), despite significant neurotrauma to the prefrontal cortex. She also produced left lateralized fine motor weakness. Conclusions Accommodations for work, outpatient therapies, and psychotherapy were recommended. This case enhances the understanding of neurocognitive performance after neurotrauma involving tumor resection and radiation treatment in rare malignant brain tumor, and highlights the impact of neural compensation after resection.

Neural Compensatory Response During Complex Cognitive Function Tasks in Mild Cognitive Impairment: A Near-Infrared Spectroscopy Study

The present pilot study was aimed at conducting a comparative analysis of the level of activation in the prefrontal cortex among a normal elderly group and amnestic and nonamnestic mild cognitive impairment (MCI) groups and investigating the presence of neural compensatory mechanisms according to types of MCI and different cognitive tasks. We performed functional near-infrared spectroscopy (fNIRS) along with cognitive tasks, including two-back test, Korean color word Stroop test, and semantic verbal fluency task (SVFT), to investigate hemodynamic response and the presence of neural compensation and neuroplasticity in the prefrontal cortex of patients with amnestic and nonamnestic MCI compared with a healthy elderly group. During the two-back test, there was no significant difference in the bilateral region-of-interest (ROI) analysis in the three groups. During the Stroop test, right-sided hyperactivation compared to the left side during the task was shown in the nonamnestic MCI and normal groups with statistical significance. Mean acc∆HbO2 on the right side was highest in the nonamnestic MCI group (0.30 μM) followed by the normal group (0.07 μM) and the amnestic MCI group (-0.10 μM). Otherwise, intergroup ROI analysis of acc∆HbO2 in these activated right sides showed no significant difference. During the VFT test, there was no significant difference in the bilateral region-of-interest analysis in the three groups. The highest mean acc∆HbO2 was shown in the normal group (0.79 μM) followed by the nonamnestic MCI group (0.52 μM) and the amnestic MCI group (0.21 μM). Otherwise, there was no significant difference between groups. The hemodynamic response during fNIRS showed different findings according to MCI types and cognitive tasks. Among the three tasks, the Stroop test showed results that were suggestive of neural compensatory mechanisms in the prefrontal cortex in nonamnestic MCI.

Adaptive Backstepping Current Control of Active Power Filter Using Neural Compensator

A backstepping-based adaptive controller with neural compensator is designed for harmonic suppression in a three-phase active power filter (APF). The fundamental rule of backstepping method is to take some state variables as “virtual controls” and then design intermediate controller. An adaptive neural controller using radial basis function (RBF) is derived to estimate the APF system nonlinearity and strengthen the current’s tracking property and power grid quality. Simulations studies indicate the proposed backstepping-based adaptive neural controller has good current tracking behavior and increased power quality.

T86. Clinical Trajectories Over More Than 15 Years Are Associated With Reward Activity Identifying Potential Regions of Neural Compensation in Bipolar Adults

Childhood onset of bipolar disorder shows severe clinical, social, and developmental outcomes that may be explained by neural abnormalities that develop over the course of the illness. Identifying these abnormalities may provide neural targets for remediation during the course of the illness to improve functioning in this debilitating disorder.

Event-triggered neural adaptive failure compensation control for stochastic systems with dead-zone output

This paper investigates event-triggered adaptive compensation control in the face of uncertain stochastic nonlinear system with actuator failure and output dead-zone. It is still an arduous task and challenge to design a compensation controller for uncertain stochastic nonlinear system. In order to avoid damaging output caused by the nonlinearity of the system, blended neural network integration with Nussbaum-type function is proposed. It is established to ensure the provision of the tracking error constraints, which is based on backstepping Lyapunov function technique. Additionally, system transmission resource constraints and actuator failure problems exist simultaneously in the system, which is extremely challenging for control design. More transmission resources are demanded when the system is suffered with actuator failures, while the system transmission resource is limited. Thus, the requirements cannot be achieved. It is difficult and challenged to ensure the system tracking performance. Using the proposed event-triggered controller and combining the Lyapunov synthesis, a novel optimization algorithm is deduced to guarantee the closed-loop system stability and the convergence of the tracking error. The simulation results illustrate the effectiveness of the proposed neural networks adaptive control approach.

Adaptive Neural Compensator for Robotic Systems Control

In the area of robotics systems, there are numerous applications where robots are expected to move rapidly from one place to another, or follow desired trajectories while maintaining good dynamic behavior. However, certain non-linearities, uncertainties in dynamics and external perturbations make the design of ideal controllers a complicated task in many situations. In this paper, we propose a control scheme that combines a nominal feedback controller with a classical PD and a robust adaptive compensator based on artificial neural networks. Using this control scheme, it is possible to obtain a fully tuned compensation parameters and a strong robustness with respect to uncertain dynamics and different non-linearities, as well as to keep the output tracking error bounded to values close to zero. In order to show the performance of the proposed technique, a SCARA (Selective Compliant Articulated Robot Arm) type robot with two degrees of freedom is considered in this case; but this control proposal can be applied to different systems with dynamic variations. The efficiency and performance of the control law is demonstrated through simulation results and the stability analysis is carried out using Lyapunov's theory.