Search

Remaining useful life (RUL) prediction has been increasingly considered in many industrial fields for the reliability and safety of their systems. As a data analysis tool of deep learning, deep convolutional neural network (CNN) shows great potential for RUL prediction. This paper proposes an intelligent RUL prediction method based on a double-CNN model architecture. Given the powerful feature extraction capability of CNN, the proposed method is fed with original vibration signals with no need to resort to any feature extractor, which can also retain the useful information in maximum. The prediction includes two stages: first, incipient fault point is identified by the first CNN model and a proposed “3/5” principle; then, the second CNN model is constructed for RUL prediction. In practice, RULs of identical components are different from each other, which poses a major challenge in RUL prediction. To overcome this problem, an intermediate reliability variable is first calculated in this paper, instead of directly predicting the RUL value. Then, a mapping algorithm is proposed to map reliability to RUL. To demonstrate the effectiveness of the proposed method, data of four tests of bearing degradation are utilized for RUL prediction. Compared with state-of-the-art methods, the proposed method shows higher prediction accuracy and robustness. The prediction results and evaluation indexes demonstrated the effectiveness and superiority of the proposed method. [ABSTRACT FROM AUTHOR]

This paper proposes and experimentally evaluates a method, which is able to measure the refrigeration capacity of compressors in extremely short times. Traditionally, the measurement of this parameter is done using specialized and expensive test rigs and can take up to 4.5 h for the measurement of a single compressor. To overcome this long measurement time, it is proposed to use a simpler measuring system, in which, instead of measuring the refrigeration capacity directly, the rate of pressure rise imposed by the compressor to a cylinder with fixed volume is measured. This paper shows that both variables have a high correlation factor when the test and compressor conditions are taken into account. As the correlation among the variables is nonlinear, it is not possible to use simple mathematical correlation functions, so neural networks are used for that purpose. By using the proposed method, the refrigeration capacity of a compressor can be determined in less than 7 s, which is approximately 0.05% of the time required by traditional methods. To guarantee the metrological reliability of the results presented by the proposed neural model, this paper proposes a novel approach based on the Monte Carlo and the bootstrap methods. [ABSTRACT FROM AUTHOR]

This paper proposes a novel intelligent fault diagnosis method to automatically identify different health conditions of wind turbine (WT) gearbox. Unlike traditional approaches, where feature extraction and classification are separately designed and performed, this paper aims to automatically learn effective fault features directly from raw vibration signals while classify the type of faults in a single framework, thus providing an end-to-end learning-based fault diagnosis system for WT gearbox without additional signal processing and diagnostic expertise. Considering the multiscale characteristics inherent in vibration signals of a gearbox, a new multiscale convolutional neural network (MSCNN) architecture is proposed to perform multiscale feature extraction and classification simultaneously. The proposed MSCNN incorporates multiscale learning into the traditional CNN architecture, which has two merits: 1) high-level fault features can be effectively learned by the hierarchical learning structure with multiple pairs of convolutional and pooling layers; and 2) multiscale learning scheme can capture complementary and rich diagnosis information at different scales. This greatly improves the feature learning ability and enables better diagnosis performance. The proposed MSCNN approach is evaluated through experiments on a WT gearbox test rig. Experimental results and comprehensive comparison analysis with respect to the traditional CNN and traditional multiscale feature extractors have demonstrated the superiority of the proposed method. [ABSTRACT FROM AUTHOR]

This paper presents a practical discrete-time fractional order terminal sliding mode (DFOTSM) control strategy for high-precision tracking tasks based on a linear motor. In particular, the practical parametric uncertainties involving sliding friction, uncertain payload, and disturbance in tracking tasks are considered in this paper. Combining Grünwald–Letnikov fractional order definition and terminal sliding mode technique, the proposed method synthesizes a novel DFOTSM control law to drive the sliding mode dynamics into the stable region in finite steps theoretically, even though the system is suffering from uncertainties and disturbances, and the motion on the surface can guarantee higher tracking precision than the conventional discrete-time terminal sliding surface by selecting suitable controller parameters. The theoretical analyses give out the guideline of parameter selection, and experiments are carried out on the linear-motor-based test platform to demonstrate that the proposed controller is easily implemented and can achieve high-precision tracking, fast response, and considerable robustness to uncertainties. [ABSTRACT FROM AUTHOR]

The general trend toward more intelligent energy-aware ac drives is driving the development of new motor topologies and advanced model-based control techniques. Among the candidates, pure reluctance and anisotropic permanent magnet motors are gaining popularity, despite their complex structure. The availability of accurate mathematical models that describe these motors is essential to the design of any model-based advanced control. This paper focuses on the relations between currents and flux linkages, which are obtained through innovative radial basis function neural networks. These special drive-oriented neural networks take as inputs the motor voltages and currents, returning as output the motor flux linkages, inclusive of any nonlinearity and cross-coupling effect. The theoretical foundations of the radial basis function networks, the design hints, and a commented series of experimental results on a real laboratory prototype are included in this paper. The simple structure of the neural network fits for implementation on standard drives. The online training and tracking will be the next steps in field programmable gate array based control systems. [ABSTRACT FROM PUBLISHER]

The Fuzzy ARTMAP is a supervised learning method, providing high accuracy in many classifications. In this paper, we explore the role of hardware accelerators in remote sensing classification missions. We focus on the designing and implementing a massively parallel hardware architecture on a field-programmable gate array (FPGA) of the performance phase's algorithm. The implementation is mapped on Xilinx Virtex 6 XC6VLX240T FPGA chip for an embedded system using Xilinx ISE 14.5 software. Embedded blocks dedicated to digital signal processing (DSP) and blocks memories are used. DSP48E1 is part of Virtex-6 FPGAs devices. It boasts of increased capability over previous generations, and is highly customizable, a key feature of this primitive that has motivated and enabled the work presented in this paper. We illustrate the application of this methodology to the valuation of various schemes involving embedded elements. This paper also presents a summary of the performance cost with regard to the speed, power, and required computational resources. [ABSTRACT FROM AUTHOR]

This paper presents a control strategy for the position and posture control of a planar four-link underactuated manipulator with a passive second link based on a back-propagation neural network (BPNN) model. First, using the differential evolution (DE) optimization algorithm, the target angles of all links are calculated. The third and fourth links are then controlled from their initial angles to zero. Next, the control process is divided into the following two steps: the control objective of the passive link is implemented in the first step, and the control objectives of the active links are implemented in the second step. In the first step, the system is reduced to a planar virtual Pendubot, a BPNN model based on the DE algorithm is trained to obtain the coupling relationship between the links of this Pendubot. This model gives the speed of the first link to move the passive link to approach the target angle. Then, a fuzzy controller is designed to indirectly control the passive link to its target angle. In the second step, the position mode of the servo controller is adopted to control all active links to their target angles. Experimental results verified the effectiveness of the proposed control strategy. [ABSTRACT FROM AUTHOR]

The classification of distinct problems of insulators in the distribution networks is a task that requires operator's experience. The applications of techniques to automate the inspection of electrical systems with the objective of detecting faults in insulators have shown to be reasonable alternatives to improve reliability in power grid. In this paper, based on the development of an experimental setup, signals are acquired considering three distinct faults in insulators. In this case, 13.8 kV (rms) is applied in drilled, contaminated, and good insulators considering an ultrasound detector connected to a computer. In the sequence, a multiclass classification method is proposed considering the ensemble of classifiers. The method considers the association of five distinct techniques, Bottom-Up segmentation, wavelet energy coefficient, principal component analysis, and particle swarm optimization associated with ensemble extreme learning machine (EN-ELM). Named optimized ensemble extreme learning machine, the present approach outperforms the original EN-ELM method. Finally, results show significant increase in robustness and faster training procedure when compared to classical approaches. [ABSTRACT FROM AUTHOR]

This paper presents an adaptive neural network backstepping control design for an experimental helicopter with three rotational degrees of freedom. The helicopter system is described by a state-space model with unmodeled dynamics and external disturbances. By resorting to radial basis function neural networks with online gradient descent training, such unmodeled dynamics and external disturbances can be compensated in real time. With the help of Lyapunov stability criteria, it is proved that the desired trajectories can be boundedly tracked by the corresponding attitude angles of the helicopter under the proposed adaptive neural network backstepping controller. Compared with robust adaptive integral backstepping technique, our approach has smaller root mean squared error. Finally, four sets of contrastive experiment results are, respectively, given to show the effectiveness and advantages of the proposed control scheme. [ABSTRACT FROM AUTHOR]

Owing to the increasing complexity of electrical systems, diagnostic techniques of cables used for connecting electrical elements are essential for system maintenance in order to prevent a failure that can cause significant impacts on the overall electrical systems. Multicore structures are typically used as control and instrumentation cables in nuclear power plants, and the failure of the control and instrumentation cables can result in a disaster such as a radiation leak. In this paper, a method for the diagnosis of multicore cables is proposed based on the reflectometry. The diagnosis relates to the classification of defective cores in joint, which is one of the weakest parts in cable systems. The reflected signals obtained through reflectometry are converted into images by an advanced image processing algorithm, and the images are classified using artificial neural networks. The proposed method is demonstrated by experimental data using a real-world multicore cable. In the experiment, the faults are emulated similar to real-world defects using a potentiometer. It is expected that the proposed technique will enhance the stability and reliability of multicore cable systems. [ABSTRACT FROM AUTHOR]

In this paper, a distributed fast fault diagnosis approach is proposed for multimachine power systems based on deterministic learning (DL) theory. First, a learning estimator is constructed to accumulate the knowledge of transient fault dynamics in power systems. Through DL, a partial persistent excitation condition of the local neurons is satisfied and convergence of the neural weights is achieved. Thus, a knowledge bank can be established and gradually updated. To deal with the high dimensional data, only the neurons centered in a neighborhood of the system trajectory are activated. In this manner, the computation load is mitigated. Second, by using the learnt knowledge, a distributed fault diagnosis scheme is designed to monitor the power system. The memory of the transient fault dynamics can be quickly recalled and a fast fault diagnosis decision is made. Finally, based on the concepts of mismatch interval and duty ratio, the diagnosis capabilities of the proposed scheme are investigated. The effectiveness of the proposed diagnosis scheme is demonstrated by computer simulation and the hardware-in-loop experimental test based on the RT-LAB realtime simulator. [ABSTRACT FROM AUTHOR]

In this paper, an adaptive trajectory tracking control algorithm for underactuated unmanned surface vessels (USVs) with guaranteed transient performance is proposed. To meet the realistic dynamical model of USVs, we consider that the mass and damping matrices are not diagonal and the input saturation problem. Neural networks (NNs) are employed to approximate the unknown external disturbances and uncertain hydrodynamics of USVs. Moreover, both full-state feedback control and output feedback control are presented, and the unmeasurable velocities of the output feedback controller are estimated via high-gain observer. Unlike the conventional control methods, we employ the error transformation function to guarantee the transient tracking performance. Both simulation and experimental results are carried out to validate the superior performance via comparing with traditional potential integral control approaches. [ABSTRACT FROM AUTHOR]

This paper develops an event-triggered decentralized tracking control (DTC) approach for modular reconfigurable robots (MRRs) by using adaptive dynamic programming. By establishing a decentralized neural network (NN) observer, which uses local input–output data and desired states of coupling subsystems, the local dynamics of MRR subsystem can be obtained. In order to obtain the DTC, the tracking error subsystem is augmented by the exosystem with the desired trajectory. Based on the event-triggered mechanism and a modified local cost function, the DTC is derived by solving the local Hamilton–Jacobi–Bellman equation via a local critic NN with asymptotically stable structure. The stability of the entire closed-loop MRR system is analyzed by Lyapunov's direct method. The simulation of a two-degree of freedom MRR system ensures that the developed event-triggered DTC scheme is effective. [ABSTRACT FROM AUTHOR]

This paper introduces a new application of deep belief network (DBN) as an emerging artificial neural network for recognizing and classifying different partial discharge (PD) patterns. Phase resolved PD (PRPD) technique with different window intervals is used to manipulate three PD types, including corona, surface, and internal discharges measured in a high-voltage lab. Four approaches are proposed for extracting features from the raw measured data. In the first approach, the DBN is used as both a feature extractor and a PD classifier. The other three approaches extract discriminatory features using statistical and vector-norm-based operators to train a DBN classifier. The impact of the various phase windows through the PRPD method on the performance of the trained classifier is evaluated to obtain the best window interval. It is shown that the deep architectures are capable of learning important distinguishable features from PD data without any data preprocessing. This eliminates time-consuming feature extraction processes that produce the handcrafted features. Based on a comparison analysis, when the input data are corrupted by noise levels or no feature extraction technique is used to preprocess the data, the proposed approach outperforms other techniques, such as artificial neural networks, fuzzy logic classifiers, and support vector machines. [ABSTRACT FROM AUTHOR]

A global and real-time control with embedded hardware artificial neural network (ANN) for imbalance current suppression of parallel connected insulated gate bipolar transistors (IGBTs) is first proposed in this paper. This method focuses on control strategy and control execution. The former one is realized by porting the ANN-based PID (ANN-PID) strategy in the control loop to yield the real-time and adaptive characteristics without IGBT quantity limitation. The latter one is realized by designing the IGBT gate quantity of charge regulator (GQR) to execute the command from ANN-PID controller. The evaluation of ANN-PID algorithm results 0.023% mean error in IGBT current control that reveals the feasibility of the proposed method. A full prototype with FPGA-based hardware accelerator for ANN-PID computing, including the designed GQR circuit, has been built for realization and qualification in a buck converter with parallel connected IGBTs. The experimental results show that the performance of the proposed method in imbalance current suppression is improved about 3.5–5.5 times as the load increase from low to high with the advantage of immunity to load change and the current imbalance can be suppressed within 4%. [ABSTRACT FROM AUTHOR]

In recent years, artificial intelligent techniques have been extensively explored in the field of health monitoring and fault diagnosis due to their powerful capabilities. In this paper, we propose a deep coupled dense convolutional network (CDCN) with complementary data to integrate information fusion, feature extraction, and fault classification together for intelligent diagnosis. In this framework, built-in and external sensor data are first developed to form the input of network in parallel. Then, a one-dimensional CDCN is proposed, which not only could naturally build deeper network with alleviating the loss of features and gradient vanishing, but also develops a double-level information fusion strategy, including self-information fusion and mutual-information fusion, to facilitate the transmission of fault information and capture more comprehensive features. Finally, the extracted joint features are used for fault recognition and classification. The proposed approach is evaluated on a planetary gearbox test-bed. The results demonstrate the validity and superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicles are an essential component in the realization of Industry 4.0. With drones helping to improve industrial safety and efficiency in utilities, construction, and communication, there is an urgent need for drone-based intelligent applications. In this paper, we develop a unified framework to simultaneously detect and count vehicles from drone images. We first explore why the state-of-the-art detectors fail in highly dense drone scenes, which provides more appropriate insights. Then, we propose an effective loss to push the anchors toward matching the ground-truth boxes as much as possible, specifically designed for scale-adaptive anchor generation. Inspired by attention mechanisms in the human visual system, we maximize the mutual information between object classes and features by combining bottom-up cues with top-down attention mechanisms specifically designed for feature extraction. Finally, we build a counting layer with a regularized constraint related to the number of vehicles. Extensive experiments demonstrate the effectiveness of our approach. For both tasks, our proposed method achieves state-of-the-art results on all four challenging datasets. In particular, our results reduce error by a larger factor than previous methods. [ABSTRACT FROM AUTHOR]

Online recognition of gestures is critical for intuitive human–robot interaction (HRI) and further push collaborative robotics into the market, making robots accessible to more people. The problem is that it is difficult to achieve accurate gesture recognition in real unstructured environments, often using distorted and incomplete multisensory data. This paper introduces an HRI framework to classify large vocabularies of interwoven static gestures (SGs) and dynamic gestures (DGs) captured with wearable sensors. DG features are obtained by applying data dimensionality reduction to raw data from sensors (resampling with cubic interpolation and principal component analysis). Experimental tests were conducted using the UC2017 hand gesture dataset with samples from eight different subjects. The classification models show an accuracy of 95.6% for a library of 24 SGs with a random forest and 99.3% for 10 DGs using artificial neural networks. These results compare equally or favorably with different commonly used classifiers. Long short-term memory deep networks achieved similar performance in online frame-by-frame classification using raw incomplete data, performing better in terms of accuracy than static models with specially crafted features, but worse in training and inference time. The recognized gestures are used to teleoperate a robot in a collaborative process that consists in preparing a breakfast meal. [ABSTRACT FROM AUTHOR]

This paper proposes a novel gradient descent least squares regression (GDLSR)-based neural network (NN) structure for the control of grid-integrated solar photovoltaic (PV) system with improved power quality. Here, a single-layer neuron structure is used for the extraction of fundamental component (FC) from the load current. During FC extraction, GDLSR-based NN structure attenuates harmonic components, noise, dc offsets, bias, notches and distortions from the nonlinear current, which improves the power quality under normal as well as under abnormal grid conditions. This single-layer GDLSR-based NN structure has a very simple architecture, which decreases the computational burden and algorithm complexity. Therefore, it is easy in implementation. In this work, the GDLSR-based control technique is tested on a single-phase single-stage grid-integrated PV topology with the nonlinear loads. The prime objective of the GDLSR-based NN structure is to provide reactive power compensation, power factor correction, harmonics filtering and mitigation of other power quality issues. Moreover, when solar irradiation is zero, the dc-link capacitor and voltage-source converter act as a distribution static compensator, which enhances the utilization factor of the system. The proposed system is modeled, and its performances are verified experimentally on a developed prototype in different grid disturbances as well as solar insolation variation conditions, which performances have satisfied the motive of the proposed technique as well as the IEEE-519 standard. [ABSTRACT FROM AUTHOR]

The demand for electricity for the fused magnesia smelting process is calculated by the moving average of the energy consumption of a given period, which gives the efficiency of the smelting and indicates if the power supply should be cut off. Therefore, demand forecasting is very important for the operation of the smelting process. However, it is difficult to forecast demand since the power changes depend on the smelting process and raw materials. To obtain an accurate model, the mechanism model of the smelting process is combined with the data-driven method using artificial intelligence technology in this paper. The mechanism model is described by a linear model with unknown parameters. The uncertainties and error of the mechanism model are modeled by neural networks with unknown order. The maximal information coefficient method and the rule reasoning are combined to identify the order. To effectively combine the mechanism model and the data-driven model, a new saturated alternating identification strategy is proposed. The results of simulations and industrial applications show that the effectiveness of the proposed intelligent method of demand forecasting has been validated. [ABSTRACT FROM AUTHOR]

Nowadays digital surveillance systems are universally installed for continuously collecting enormous amounts of data, thereby requiring human monitoring for the identification of different activities and events. Smarter surveillance is the need of this era through which normal and abnormal activities can be automatically identified using artificial intelligence and computer vision technology. In this paper, we propose a framework for activity recognition in surveillance videos captured over industrial systems. The continuous surveillance video stream is first divided into important shots, where shots are selected using the proposed convolutional neural network (CNN) based human saliency features. Next, temporal features of an activity in the sequence of frames are extracted by utilizing the convolutional layers of a FlowNet2 CNN model. Finally, a multilayer long short-term memory is presented for learning long-term sequences in the temporal optical flow features for activity recognition. Experiments 1 https://github.com/Aminullah6264/Activity%5fRec%5fML-LSTM. are conducted using different benchmark action and activity recognition datasets, and the results reveal the effectiveness of the proposed method for activity recognition in industrial settings compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Interconnected dc grids are studied in this paper, which comprise resistive and constant-power loads (CPLs) fed by photovoltaic (PV) units. All the sources and CPLs are connected to the grid via dc–dc converters. Nonlinear behavior of PV units in addition to the effect of negative-resistance CPLs can destabilize the dc grid. Thus, the decentralized nonlinear model and intelligent control are proposed using adaptive output-feedback controller to stabilize the grid. The use of the output-feedback control makes possible the utilization of other available signals, in case of loss of main signal, at the converter location and creates measurement redundancy that improves reliability of the dc network. The switching between measured signals of different types are performed through using the neural network (NN) controllers without the need to further tuning. The stability of the entire network is assured through Lyapunov stability method while each converter employs only local measurements. The adaptive NNs are utilized to overcome the unknown dynamics of the dc–dc converters at distributed energy resources and CPLs and those of the interconnected network imposed on the converters. Simulation and experimental results are provided on a small-scale dc grid to show the effectiveness of the developed model and the proposed controller. [ABSTRACT FROM AUTHOR]

This paper focuses on developing an advanced nonlinear controller for hydraulic system to achieve asymptotic tracking with various disturbances. To accomplish this study, a multilayer neural-networks (NNs) estimator is first developed to improve the compensation accuracy of model-based feedforward control terms, which can greatly reduce the uncompensated disturbance, then a continuous robust integral of the sign of the error (RISE) control approach is effectively integrated with the multilayer NNs estimator to deal with the residual mismatched disturbance, in which the RISE feedback gain is adapted online to further decrease the high-gain feedback. At last, by considering the inwardness of matched disturbance in hydraulic systems, it is estimated by another multilayer NNs fully with the residual functional reconstruction inaccuracies handled by a novel adaptive term. As a result, theoretical analysis reveals that the proposed controller guarantees a semiglobal asymptotic stability. Extensively comparative experimental results verify the priority of the proposed control strategy, and a 0.2% dynamic tracking accuracy is achieved. [ABSTRACT FROM AUTHOR]

This paper proposes the use of an artificial neural network (ANN) for solving one of the ongoing research challenges in finite set-model predictive control (FS-MPC) of power electronics converters, i.e., the automated selection of weighting factors in cost function. The first step in this approach is to simulate a detailed converter circuit model or run experiments numerous times using different weighting factor combinations. The key performance metrics [e.g., average switching frequency ($f_{{\rm sw}}$) of the converter, total harmonic distortion, etc.] are extracted from each simulation. This data is then used to train the ANN, which serves as a surrogate model of the converter that can provide fast and accurate estimates of the performance metrics for any weighting factor combination. Consequently, any arbitrary user-defined fitness function that combines the output metrics can be defined and the weighting factor combinations that optimize the given function can be explicitly found. The proposed methodology was verified on a practical weighting factor design problem in FS-MPC regulated voltage source converter for uninterruptible power supply system. Designed weighting factors for two exemplary fitness functions turned out to be robust to load variations and to yield close to expected performance when applied both to detailed simulation model (less than 3% error) and to experimental test bed (less than 10% error). [ABSTRACT FROM AUTHOR]

Automated early detection and identification of switch faults are essential in high-voltage applications. Modular multilevel converter (MMC) is a new and promising topology for such applications. MMC is composed of many identical controlled voltage sources called modules or cells. Each cell may have one or more switches and a switch failure may occur in anyone of these cells. The steady-state normal and fault behavior of a cell voltage will also significantly vary according to the changes in the load current and the fault timing. This makes it a challenging problem to detect and identify such faults as soon as they occur. In this paper, we propose a real-time and highly accurate MMC circuit monitoring system for early fault detection and identification using adaptive one-dimensional convolutional neural networks. The proposed approach is directly applicable to the raw voltage and current data and thus eliminates the need for any feature extraction algorithm, resulting in a highly efficient and reliable system. Simulation results obtained using a four-cell, eight-switch MMC topology demonstrate that the proposed system has a high reliability to avoid any false alarm and achieves a detection probability of 0.989, and average identification probability of 0.997 in less than 100 ms. [ABSTRACT FROM AUTHOR]

This paper investigates the optimal consensus control problem for discrete-time multi-agent systems with completely unknown dynamics by utilizing a data-driven reinforcement learning method. It is known that the optimal consensus control for multi-agent systems relies on the solution of the coupled Hamilton–Jacobi–Bellman equation, which is generally impossible to be solved analytically. Even worse, most real-world systems are too complicated to obtain accurate mathematical models. To overcome these deficiencies, a data-based adaptive dynamic programming method is presented using the current and past system data rather than the accurate system models also instead of the traditional identification scheme which would cause the approximation residual errors. First, we establish a discounted performance index and formulate the optimal consensus problem via Bellman optimality principle. Then, we introduce the policy iteration algorithm which motivates this paper. To implement the proposed online action-dependent heuristic dynamic programming method, two neural networks (NNs), 1) critic NN and 2) actor NN, are employed to approximate the iterative performance index functions and control policies, respectively, in real time. Finally, two simulation examples are provided to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM PUBLISHER]

In order to improve the pulp quality and to reduce the energy consumption, the fiber length distribution (FLD) is generally employed as one of the important technological indexes in the refining process. Considering that the traditional mean and variance of fiber length are unable to adequately characterize the non-Gaussian distribution properties, this paper proposes a novel geometric analysis based double closed-loop iterative learning control (ILC) method for probability density function (PDF) shaping of output FLD in the refining process. Primarily, a radial basis function (RBF) neural network with Gaussian-type is utilized to approximate the square root PDF in the inner loop, where the RBF basis function parameters (center and width) are tuned between any two adjacent batches by using an ILC law, and the subspace identification method can be applied to establish the state-space model of weight vector. Then, for the sake of accelerating the convergence rate of the closed-loop system, a geometric analysis based ILC method is adopted in the outer loop. Finally, both simulation and experiments demonstrate the effectiveness and practicability of the proposed approach. [ABSTRACT FROM AUTHOR]

In this paper, an incremental centralized formation system is developed for controlling the multirobot formation with joining robots, and a nonlinear model predictive control (NMPC) method is implemented as the controller. The incremental updating method is used to update the system's state in real time, when there is a new robot joining during the formation process. Then, an NMPC approach is developed to reformulate the formation system into a convex nonlinear minimization problem, which can be further transformed into a quadratic programming (QP) with constraints. Then, a general projection neural network (GPNN) is implemented for solving this QP problem online to get the optimal inputs. In the end, two examples of incremental multirobot formation are demonstrated to verify the effectiveness of this method. [ABSTRACT FROM AUTHOR]

Time series classification is an important research topic in machine learning and data mining communities, since time series data exist in many application domains. Recent studies have shown that machine learning algorithms could benefit from good feature representation, explaining why deep learning has achieved breakthrough performance in many tasks. In deep learning, the convolutional neural network (CNN) is one of the most well-known approaches, since it incorporates feature learning and classification task in a unified network architecture. Although CNN has been successfully applied to image and text domains, it is still a challenge to apply CNN to time series data. This paper proposes a tensor scheme along with a novel deep learning architecture called multivariate convolutional neural network (MVCNN) for multivariate time series classification, in which the proposed architecture considers multivariate and lag-feature characteristics. We evaluate our proposed method with the prognostics and health management (PHM) 2015 challenge data, and compare with several algorithms. The experimental results indicate that the proposed method outperforms the other alternatives using the prediction score, which is the evaluation metric used by the PHM Society 2015 data challenge. Besides performance evaluation, we provide detailed analysis about the proposed method. [ABSTRACT FROM AUTHOR]

This paper studies the two-timescales operational optimal control problem for the flotation industrial process with unknown operational model in the presence of setpoint constraints on the device layer. A novel dual-rate data-driven algorithm based on lifting technology and reinforcement learning (RL) is proposed. First, a dual-rate flotation process model including the device layer and the operational models is formulated. Then, a proportional integral (PI) controller for device layer is designed, by using lifting technology, a unified timescale controlled plant model with partially unknown dynamics is established. Based on such a model, an online learning algorithm using neural network is presented so that the operational indices, namely concentrate and tail grades, can be kept in the target range while maintaining the setpoints of the device layer within the specified bounds. At last, emulation experiments in a hardware-in-the-loop system are used to verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Tool condition monitoring is critical in ultraprecision manufacturing in order to optimize the performance of the overall process, while maintaining the desired part quality. Recently, deep learning has been successfully applied to numerous classification tasks in manufacturing, often to forecast part quality. In this paper, a novel deep learning data-driven modeling framework is presented, which includes a fusion of multiple stacked sparse autoencoders for tool condition monitoring in ultraprecision machining. The proposed computational framework consists of two main structures. First, a training model that is designed with the ability to process multiple parallel feature spaces to learn the lower-level features. Second, a feature fusion structure that is used to learn the higher-level features and associations to tool wear. To achieve this learning structure, a modified loss function is utilized that enhances the feature extraction and classification tasks. A dataset from a real manufacturing process is used to demonstrate the performance of the proposed framework. Experimental results and simulations show that the proposed method successfully classifies the ultraprecision machining case study with over 96% accuracy, while also outperforming comparable methodologies. [ABSTRACT FROM AUTHOR]

Efficient and accurate planetary gearbox fault diagnosis is the key to enhance the reliability and security of wind turbines. Therefore, an intelligent and integrated approach based on deep belief networks (DBNs), improved logistic Sigmoid (Isigmoid) units, and impulsive features is proposed in this paper. The vanishing gradient problem is an inherent drawback of conventional Sigmoid units, and it usually occurs in the backpropagation process of DBNs, resulting in that the training is considerably slowed down and the classification rate is reduced. To solve this problem, Isigmoid units are designed to combine the merits of unsaturation from leaky rectified linear (LReL) units. The results of handwritten digit recognition experiments show the superiority of Isigmoid over Sigmoid on convergence speed and classification accuracy. Since impulses contain much useful fault information, especially for early failures, an integrated approach using the optimized Morlet wavelet transform, kurtosis index, and soft-thresholding is applied to extract impulse components from original signals to improve the diagnosis accuracy. Then, the features extracted from original signals and impulsive signals are employed to train and test the DBNs with Isigmoid, Sigmoid, and LReL units for comparison. Finally, the results of planetary gearbox fault diagnosis show that Isigmoid has higher comprehensive performance than conventional sigmoid and LReL. [ABSTRACT FROM AUTHOR]

For the safe operation of redundant manipulators, physical constraints such as the joint angle, joint velocity, and joint acceleration limits should be taken into account when designing redundancy resolution schemes. Velocity-level redundancy resolution schemes are widely adopted in the kinematic control of redundant manipulators due to the existence of the well-tuned inner loop regarding the joint velocity control. However, it is difficult to deal with joint acceleration limits for velocity-level redundancy resolution methods. In this paper, a recurrent-neural-network-based velocity-level redundancy resolution method is proposed to deal with the problem, and theoretical results are given to guarantee its performance. By the proposed method, the end-effector position error is asymptotically convergent to zero, and all the joint limits are not violated. The effectiveness and superiority of the proposed scheme are validated via simulation results. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of asymptotic trajectory tracking with prescribed transient behavior for a fully actuated marine surface vessel subject to unmodeled dynamics and external disturbances. A prescribed transient performance constraint is introduced to specify the transient behavior of the output tracking errors. Neural network (NN) approximation and adaptive control techniques are employed to compensate for the unmodeled dynamics. In general, the NN approximation errors always exist in the NN-based adaptive control systems, and typically result in uniformly ultimate boundedness, where the tracking errors cannot be stabilized asymptotically. In order to obtain an asymptotic tracking performance, a robust integral of the sign of the error (RISE) feedback term is introduced in feedback control design to compensate for the NN approximation errors and external disturbances. Consequently, RISE-based adaptive NN controller with compensations of unmodeled dynamics and external disturbances is proposed such that an asymptotic tracking performance is achieved with guaranteeing prescribed transient behavior. The performance of the proposed tracking controller is shown by simulation results. [ABSTRACT FROM AUTHOR]

In a wireless power transfer (WPT) system via coupled magnetic resonances, the power transfer efficiency (PTE) drastically decreases with the transfer distance or the load changing. In this paper, the causes of efficiency degradation are analyzed, and an automatic impedance matching method based on the feedforward-backpropagation (BP) neural network is proposed to maintain the PTE at a reasonable level. To validate and test the performance of the proposed method, a WPT automatic impedance matching simulation system is implemented. Moreover, a prototype based on the proposed method is built and dynamic matching experiments were performed. The simulation results show that the algorithm efficiency of the proposed BP method is 108.5% higher than that of the genetic algorithm. The experimental results show that the PTE is improved up to 78.33% and this is closely maintained within a distance of 10–30 cm, which is consistent with the simulation result. [ABSTRACT FROM AUTHOR]

Bearing remaining useful life (RUL) prediction plays a crucial role in guaranteeing safe operation of machinery and reducing maintenance loss. In this paper, we present a new deep feature learning method for RUL estimation approach through time frequency representation (TFR) and multiscale convolutional neural network (MSCNN). TFR can reveal nonstationary property of a bearing degradation signal effectively. After acquiring time-series degradation signals, we get TFRs, which contain plenty of useful information using wavelet transform. Owing to high dimensionality, the size of these TFRs is reduced by bilinear interpolation, which are further regarded as inputs for deep learning models. Here, we introduce an MSCNN model structure, which keeps the global and local information synchronously compared to a traditional convolutional neural network (CNN). The salient features, which contribute for RUL estimation, can be learned automatically by MSCNN. The effectiveness of the presented method is validated by the experiment data. Compared to traditional data-driven and different CNN-based feature extraction methods, the proposed method shows enhanced performance in the prediction accuracy. [ABSTRACT FROM AUTHOR]

This paper is focused on the linearization of the radio frequency power amplifier of a professional digital handheld by means of an artificial neural network. The simplicity of the neural network that is used, together with the fact that a feedback path is unnecessary, makes this solution ideal to reduce both the cost of a handheld and its hardware complexity, while fully maintaining its performance. A compensation system is also needed to keep the linearization characteristics of the neural network stable against frequency, temperature, and voltage variations. The whole solution that comprises both the neural network and the compensation system has been implemented in the digital signal processor of a real handheld and afterward fully tested. It has proved to be satisfactory to meet the telecommunication standard requirements in all frequency, temperature, and voltage ranges under consideration while efficient to lower the computational cost of the handheld and to make its internal hardware simpler in comparison with other traditional linearization techniques. The results obtained demonstrate that a neural network can be used to linearize the power amplifiers that are used in transmitters of telecommunication equipment, leading to a significant reduction of both their hardware cost and complexity. [ABSTRACT FROM AUTHOR]

This paper proposes a novel approach for detection of stator short-circuit faults in three-phase induction motors. The method is based on two stages: feature extraction and classification by intelligent systems. First, mutual information is estimated from delayed stator current signals, which are used as inputs of C4.5 decision trees and multilayer perceptron neural networks in the second step. Several offline and online experimental tests are presented considering voltage unbalance, load torque variations, and 1% to 10% short-circuit levels. The obtained results corroborate the effectiveness of this new diagnostic approach. [ABSTRACT FROM AUTHOR]

In addition to diagnosing a wiring of the vehicle in operation, it is also very important to find wire mismatches during the assembly process. In this paper, we propose a new method combining time–frequency-domain reflectometry and deep learning to verify that the wire is connected to the proper port of the underhood electrical center. Considering the frequency characteristics of each wire (black, blue, red, and yellow), we develop an optimization signal design algorithm. Using the time–frequency cross correlation (TFCC), the reflected signal generated at the impedance discontinuities is acquired and converted into the Wigner–Ville distribution image. Through the proposed algorithm, the existing images are converted into new images, which are easy to distinguish among groups. The new images are used as input of the convolutional neural network and trained to learn the feature of each group. The lengths, compensation filters, and the port information to be connected to each wire are stored in the filter bank. If the distance derived using the TFCC is different from the stored length, the wire is considered defective, and the acquired signal is restored by the compensation filter designed by the overcomplete wavelet transform method. Experimental results demonstrate the effectiveness of the proposed method for detecting the wire mismatch and fault location. [ABSTRACT FROM AUTHOR]

In this paper, a data-driven fault diagnosis model dealing with chemical imbalanced data streams is investigated. Different faults occur with varied frequencies by continuous arrival in chemical plants, while this issue has been hardly addressed in developing a diagnosis model. A novel incremental imbalance modified deep neural network (incremental-IMDNN) is proposed to promote the fault diagnosis to the imbalanced data stream. The first step in designing the incremental-IMDNN is the employment of an imbalance modified method combined with active learning for the extraction and generation of the most valuable information keeping in view the model feedback. DNN is utilized as a basic diagnosis model to excavate potential information. Then for the continuous arrival of new fault modes, DNN is promoted in an incremental hierarchical way. Unlike the traditional model that trained on a static snapshot of data, this model inherits the existing knowledge and hierarchically expands the diagnosis model by the similarity of faults. Similar faults that are judged by fuzzy clustering merge into a superclass, and every submodel shares the same architecture that is prevalent in previous research, which can be trained in parallel. We validate the performance of the proposed method in a Tennessee Eastman (TE) dataset, and the simulation results indicate that the proposed incremental-IM-DNN is better than the existing methods and possesses significant robustness and adaptability in chemical fault diagnosis. [ABSTRACT FROM AUTHOR]

In this paper, a neural-dynamic optimization-based nonlinear model predictive control (NMPC) is developed for controlling leader–follower mobile robots formation. Consider obstacles in the environments, a control strategy is proposed for the formations which includes separation-bearing-orientation scheme (SBOS) for regular leader–follower formation and separation-distance scheme (SDS) for obstacle avoidance. During the formation motion, the leader robot shall track a desired trajectory and the desire leader–follower relationship can be maintained through SBOS method; meanwhile, the followers can avoid the collision by applying the SDS. The formation-error kinematics of both SBOS and SDS are derived and a constrained quadratic programming (QP) can be obtained by transforming the MPC method. Then, over a finite-receding horizon, the QP problem can be solved by utilizing the primal-dual neural network (PDNN) with parallel capability. The computation complexity can be greatly reduced by the implemented neural-dynamic optimization. Compared with other existing formation control approaches, the developed solution in this paper is rooted in NMPC techniques with input constraints and the novel QP problem formulation. Finally, experimental studies of the proposed formation control approach have been performed on several mobile robots to verify the effectiveness. [ABSTRACT FROM PUBLISHER]

In this paper, a discrete-time neural network for solving convex quadratic programming (QP) problems in constrained model predictive control (MPC) technology is investigated and implemented on a digital signal processor (DSP) device. This makes it possible to apply MPC technology to local control for high-dimensional multiple-input–multiple-output systems. The convergence issue of the discrete-time neural network is first studied. By choosing a proper error function, a sufficient condition is obtained under which the neural network converges to the exact optimal solution globally. This is the theoretical basis of this paper. An integrated hardware and software design method to implement the neural network on a DSP chip as a universal QP solver is then developed. With the QP solver handling the computational tasks in MPC problems, a general DSP-based MPC controller is achieved. A prototype system is built on a TMDSEVM6678L DSP development board. It is then applied to an air-separation-unit system and achieves satisfactory control performance. This verifies the effectiveness of the whole design. [ABSTRACT FROM PUBLISHER]

In this paper, a real-time implementation of selective harmonic elimination pulsewidth modulation (SHEPWM) using generalized Hopfield neural network (GHNN) as applied to the single-phase inverter is designed and discussed. Finding the location of the switching instants in the case of SHEPWM involves the solution of a set of nonlinear algebraic transcendental equations. The problem is redrafted as an optimization problem, and it is solved by using GHNN. Though, in principle, the $n$ number of harmonics can be eliminated using the presented methodology, this paper tries for the case of the elimination of the 5th-, 7th-, 11th-, and 13th-order harmonics while retaining the desired fundamental. An energy function is formulated for this problem and the set of ordinary differential equations (ODEs) describing the behavior of GHNN is obtained. A program in KEIL C was developed to solve these systems of ODEs by the Runge–Kutta fourth-order method to give the switching instants for continuously varying modulation indexes $(M)$. A MATLAB simulation was carried out, and an experimental setup was also constructed in order to validate the simulated results. The fast Fourier transform analysis of the simulated output voltage waveform and the experimental output voltage waveform confirms the effectiveness of the proposed method. Hence, the proposed method proves that it is much applicable in the industrial applications by virtue of its suitability in real-time applications. [ABSTRACT FROM PUBLISHER]

Considering the unknown dynamics of the multiple-input-multiple-output strict-feedback nonlinear systems, this paper proposed the neural composite learning control using the online recorded data. The control structure follows the back-stepping scheme, while neural networks (NNs) are employed for uncertainty approximation. Through the error dynamics analysis, the critical prediction error is constructed to indicate the performance of intelligent learning over the interval. Furthermore, the novel composite learning law is proposed for NN weights update. The stability of the closed-loop system is analyzed via the Lyapunov approach and the signals are guaranteed to be bounded. Through simulation test of two-input and two-output systems, the proposed controller can achieve better tracking performance, while with the composite learning algorithm, the NNs can efficiently approximate nonlinear functions. Similar conclusions are obtained on the control of the hypersonic reentry vehicle. [ABSTRACT FROM PUBLISHER]

It is known that adverse environments such as high reverberation and low signal-to-noise ratio (SNR) pose a great challenge to indoor sound source localization (SSL). To address this challenge, in this paper, we propose an SSL algorithm based on a probabilistic neural network, namely a generalized cross-correlation classification algorithm (GCA). Experimental results for adverse environments with high reverberation time T60 up to 600 ms and low SNR such as -10 dB show that the average azimuth angle error and elevation angle error by GCA are only 4.6° and 3.1°, respectively. Compared with three recently published algorithms, GCA has increased the success rate on direction of arrival estimation significantly with good robustness to environmental changes. These results show that the proposed GCA can localize accurately and robustly for diverse indoor applications where the site acoustic features can be studied prior to the localization stage. [ABSTRACT FROM PUBLISHER]

In this paper, we design an automatic pearl classification machine, composed of four parts: feeding mechanism, delivering mechanism, vision-based detection device, and classification mechanism. Pearls can be delivered to the detection device one by one, where multiview images of each pearl can be collected. A novel multistream convolutional neural network (MS-CNN) is developed to cope with these multiview images, with each stream processing an image of particular viewing angle and different streams sharing part of weights to fuse high-order features without losing too much diversity. Using the machine, we collect 52 500 multiview images for 10 500 pearls, i.e., each pearl has five images of top, left, right, main, and rear views. These pearls were labeled by the experienced professionals in advance, and grouped into two classes with rough rules and seven classes with fine rules. Experimental results show that, compared with the support vector machine and backpropagation neural network, our MS-CNN behaves much better in both classification tasks, obtaining 92.14% and 91.24% accuracies. Moreover, the visualization of activations of convolutional kernels suggests that MS-CNN, imitating the manual process, can indeed recognize relatively complex features. These results indicate the potential value of our machine in the pearl industry. [ABSTRACT FROM PUBLISHER]

This paper concentrates on the problem of control of a hybrid energy storage system (HESS) for an improved and optimized operation of load-frequency control applications. The HESS consists of a supercapacitor serving as the main power source and a fuel cell serving as the auxiliary power source. First, a Hammerstein-type neural network is proposed to identify the HESS, which formulates the Hammerstein model with a nonlinear static gain in cascade with a linear dynamic block. It provides the model information for the controller to achieve the adaptive performance. Second, a feedforward neural network based on a back-propagation training algorithm is designed to formulate the proportional-integral-derivative (PID)-type neural network, which is used for the adaptive control of the HESS. Meanwhile, a dynamic antiwindup signal is designed to solve the operational constraint of the HESS. Then, an appropriate power reference signal for the HESS can be generated. Third, the stability and the convergence of the whole system are proved based on the Lyapunov stability theory. Finally, simulation experiments are followed through on a four-area interconnected power system to demonstrate the effectiveness of the proposed control scheme. [ABSTRACT FROM PUBLISHER]

The key challenge of intelligent fault diagnosis is to develop features that can distinguish different categories. Because of the unique properties of mechanical data, predetermined features based on prior knowledge are usually used as inputs for fault classification. However, proper selection of features often requires expertise knowledge and becomes more difficult and time consuming when volume of data increases. In this paper, a novel deep learning network (LiftingNet) is proposed to learn features adaptively from raw mechanical data without prior knowledge. Inspired by convolutional neural network and second generation wavelet transform, the LiftingNet is constructed to classify mechanical data even though inputs contain considerable noise and randomness. The LiftingNet consists of split layer, predict layer, update layer, pooling layer, and full-connection layer. Different kernel sizes are allowed in convolutional layers to improve learning ability. As a multilayer neural network, deep features are learned from shallow ones to represent complex structures in raw data. Feasibility and effectiveness of the LiftingNet is validated by two motor bearing datasets. Results show that the proposed method could achieve layerwise feature learning and successfully classify mechanical data even with different rotating speed and under the influence of random noise. [ABSTRACT FROM PUBLISHER]

This paper describes a method for developing a novel, low-cost, hand-held electronic nose (E-Nose) for black tea flavor estimation using metal–oxide–semiconductor gas sensors. In this research, a first-level experimentation has been conducted with a computer-assisted-technology-based system for sensor selection and validation. In the second level, a novel hand-held E-Nose is designed and developed for generation of tea aroma, capturing the response patterns and signal processing in a microcontroller (PIC 18F45K22) for tea flavor grading. The results of the first-level experimentation provide the sensor selection, where only four out of 12 sensors were selected based on their performance. Moreover, the classification network was predetermined for implementation in the microcontroller. Experiments conducted on the developed hand-held E-Nose show the classification on eight different grades of tea (each with ten samples) varying from 70% to 100%, revealing an average classification efficiency of 90%. The repeatability of the system was also found to be 80%. The developed system also overcomes the problem of heating the tea sample, since the volatile organic compounds of tea are generated due to the emitted heat of the sensors in the proposed system. [ABSTRACT FROM PUBLISHER]

Recently, it has been verified by simulation results that the ability of recognizing and classifying patterns by artificial neural network can be successfully used for anti-islanding protection of distributed synchronous generators. However, since this type of protection differs from the available commercial relays, it is important to test the algorithm performance implemented in hardware. So, this paper presents a performance evaluation of the hardware implementation of an ANN-based algorithm for anti-islanding protection of distributed generators. For the evaluation process, a complete digital relay architecture is developed in LabVIEW, thus allowing its implementation in the commercial real-time platform cRIO-9025/cRIO-9018. The proposed scheme is tested under real conditions by means of the CMC-256-plus test set, as performed for conventional intelligent electronic devices. The proposed algorithm implemented in hardware shows high effectiveness, detecting all islanding conditions within 1 s, sharing the same good performance verified by computational simulations. The algorithm implementation shows robustness, as it remains blocked for other power grid disturbances, such as frequency variation in the grid utility, massive connection or load relief. [ABSTRACT FROM PUBLISHER]