Search

Key objects’ semantic segmentation and power line extraction from the electric power corridor point cloud are critical steps in power line inspection. However, the massive amount of point cloud data and missing power line points pose a challenge to the object extraction. To complete the extraction of power lines and other essential objects, a method called SS-IPLE is proposed, which is based on a pointwise multilayer-perceptron semantic segmentation network. The method consists of two main parts: electric power corridor semantic segmentation and individual power line extraction. In the segmentation step, SCF-Net is employed as our primary segmentation network, and the network can process large-scale point clouds. To further improve the segmentation ability of SCF-Net in the corridor, a local coding module is designed to construct the SCFL-Net. In the individual power line extraction step, individual power lines are extracted by flexible grid filtering, effectively overcoming the point-missing problem. The corridor point cloud semantic segmentation and individual power line extraction experiments are conducted in different corridors collected from suburban areas. Promising results are obtained for both semantic segmentation and individual power line extraction, with an mIou of point cloud semantic segmentation and a mean extraction rate of 96.70% and 96.56%, respectively.

To improve the robustness of UAVs in rotating and motion scenes, we propose a stereo simultaneous localization and mapping (SLAM) for UAVs with a new keyframe strategy that differs from current SLAM systems. Moreover, it has two strategies. 1) In the dominant strategy based on image quality, we filter and retain the strong feature points in each image frame by our own defined rules; this leads to longer survival time and attribute invariance in image tracking, and we save the image frames containing more strong feature points as keyframes. 2) In the secondary strategy based on motion state, we quantify the motion state during the UAV’s motion (called the compound rotation amount), and characterize the intensity of the motion. Since we want the UAV to have better robustness in the rotating scene, this strategy generates keyframes when the compound rotation amount meets the threshold. These two strategies are used to cope with gentle motion scenes and rotational motion scenes, respectively. Thus, the insertion of our keyframes is determined by the motion state; our SLAM system is proposed based on the motion state (MS-SLAM). In the back-end part of the system we construct a new weighted cost function to optimize the pose. Finally, through comparison experiments on the public dataset EuRoc, we demonstrate that our algorithm is more advantageous than some current mainstream algorithms. In difficult sequences, our algorithm compares the absolute trajectory error with ORB-SLAM2, SVO+gtsam, and VINS-Mono, the absolute trajectory error of our algorithm can be reduced by 87%.

Voltage sag is one of the most common power quality disturbance in industry, which causes huge inrush current and inrush torque of induction motors, and has serious impacts on the safe and stable operation of induction motor. To further study the detailed transient characteristics of induction motors during sag events, an improved 2-D multi-slice field-circuit-motion coupling time-stepping finite element method (FEM) is proposed. Then, taking a 5.5 kW, a 55 kW and a 135 kW induction motors as examples, the influence of phase-angle jumps and initial point-on-wave on the stator current peak, torque peak, speed loss and critical clearing time of the induction motor are discussed by using the proposed model. The simulation results show that stator current peak and torque peak increase obviously, while speed and critical clearing time decrease apparently, when considering the phase angle jump. And different initial points have little influence on these operation indicators. Finally, the proposed method is verified by comparing the measured and calculation data of the 5.5 kW three-phase squirrel-cage motor, respectively. The results show that the proposed method can reach higher precision with considering nonlinear characteristics.

This paper introduces the structural characteristics of a 6-D force sensor based on an E-type membrane, analyzes the calibration results of each bridge, and determines the coupling relationship between bridges of a sensor. In the case that a sensor has no fault in its bridges, a decoupling matrix is calculated by identifying a linear decoupling model. In the case that the fault happens, a linear neural network method is used to discard the faulty bridge to calculate the reduced-dimensional decoupling matrix, and the BP neural network nonlinear method is used to compensate the faulty bridge signal for fault-tolerant decoupling. Simulation results indicate that the reliability of the sensor under the fault condition has been significantly improved.

Interaction between wind farms and AC systems can cause oscillatory instabilities across a wide frequency band. Owing to the high-order nature of integrated wind power system, it is almost impossible to determine the system stability directly with the criterion based on the zeroes of the determinant of the impedance matrix. To address this issue, a stability analysis method of zeros identification based on frequency response characteristics of the determinant of the impedance matrix is proposed in this paper. It divides the concerned frequency band into several sub-bands. In each sub-band, a unified reduced-order frequency response model is established, which retains the dominant oscillation mode and a simplified equivalent term of other modes. Therefore, the unknown parameters are significantly reduced, and then the remaining parameters can be easily identified by the fitting algorithm. Finally, zeros of the reduced-order frequency response model in each sub-band can be solved to analyze the system stability. Practical procedure of the proposed method for application in the high order system is proposed. Both the eigenvalue analysis and time-domain simulations performed in three representative systems have validated effectiveness and accuracy of the proposed procedure. [ABSTRACT FROM AUTHOR]

As the wind-integrated power system is a strongly nonlinear system with uncertain operating conditions that vary with time, it is difficult to guarantee the effectiveness of the conventional model-based sub-synchronous damping control (SSDC). To solve this problem, a model-free adaptive control (MFAC) method for static synchronous compensator (STATCOM) is proposed to realize sub-synchronous oscillation (SSO) mitigation. The method is driven by a limited number of online input/output (I/O) data rather than full technical and operational details of the system. First, the structure of the proposed MFAC is described in detail with control algorithms of all modules. Particularly, a new form of one-step-ahead weighted predictive control algorithm of MFAC for SSO mitigation is proposed. Then, the convergence of tracking error along with the stability of the closed-loop system with the improved MFAC are analyzed. Finally, the performance of the proposed MFAC for STATCOM is demonstrated and compared with that of the conventional SSDC under various operating conditions. Time domain simulation results validate the effectiveness of the proposed MFAC and demonstrate its better adaptability than the conventional SSDC. [ABSTRACT FROM AUTHOR]

The fuzzy broad learning system (FBLS) is a recently proposed neuro-fuzzy model that shares the similar structure of a broad learning system (BLS). It shows high accuracy in both classification and regression tasks and inherits the fast computational nature of a BLS. However, the ensemble of several fuzzy subsystems in an FBLS decreases the possibility of understanding the fuzzy model since the fuzzy rules from different fuzzy systems are difficult to combine together while keeping the consistence. To balance the model accuracy and complexity, a synthetically simplified FBLS with better interpretability, named compact FBLS (CFBLS), is developed in this article, which can generate much fewer and more explainable fuzzy rules for understanding. In such a way, only one traditional Takagi–Sugeno–Kang fuzzy system is employed in the feature layer of a CFBLS, and the input universe of discourse is equally partitioned to obtain the fuzzy sets with proper linguistic labels accordingly. The random feature selection matrix and rule combination matrix are employed to reduce the total number of fuzzy rules and to avoid the “curse of dimensionality.” The enhancement layer is kept in the CFBLS that helps to add a nonlinear transformation of input features to the traditional first-order polynomial used in the consequent part of a fuzzy rule. The pseudoinverse is also used to determine the parameters of CFBLS guaranteeing its fast computational nature. The experiments on the popular UCI and KEEL datasets indicate that the CFBLS can generate a smaller set of comprehensible fuzzy rules and achieve much higher accuracy than some state-of-the-art neuro-fuzzy models. Moreover, the advantage of CFBLS is also verified in a real-world application. [ABSTRACT FROM AUTHOR]

The broad learning system (BLS) has been proved to be effective and efficient lately. In this article, several deep variants of BLS are reviewed, and a new adaptive incremental structure, Stacked BLS, is proposed. The proposed model is a novel incremental stacking of BLS. This invariant inherits the efficiency and effectiveness of BLS that the structure and weights of lower layers of BLS are fixed when the new blocks are added. The incremental stacking algorithm computes not only the connection weights between the newly stacking blocks but also the connection weights of the enhancement nodes within the BLS block. The Stacked BLS is considered as the increment of “layers” and “neurons” dynamically during the training for multilayer neural networks. The proposed architecture along with the training algorithms that utilizes the residual characteristic is very versatile in comparison with traditional fixed architecture. Finally, experimental results on UCI datasets, MNIST dataset, NORB dataset, CIFAR-10 dataset, SVHN dataset, and CIFAR-100 dataset indicate that the proposed method outperforms the selected state-of-the-art methods on both accuracy and training speed, such as deep residual networks. The results also imply that the proposed structure could highly reduce the number of nodes and the training time of the original BLS in the classification task of some datasets. [ABSTRACT FROM AUTHOR]

In this article, the problem of the stochastically finite time stabilization for an uncertain single-input and single-output stochastic system in presence of input quantization is studied. The broad learning system (BLS) is first applied to identify the uncertain system with unknown dynamics. The problem of unmeasured states can be solved by establishing a novel BLS-based state observer. Combining the stochastically finite time theorem with Itô formula, a new finite time design method is proposed, which can reduce the difficulty in designing controllers by traditional methods. A stochastically finite time quantized control method is presented by utilizing a new finite time design Lemma and quantized input decomposition technique. The developed control approach can guarantee that the closed-loop system is semi-global finite-time stable in probability, and the convergence performances are well in presence of actuator quantization. The simulation on a chemical reactor is utilized to verify the proposed scheme, which demonstrates the advantage of BLS, as well as the validity of our control method. [ABSTRACT FROM AUTHOR]

We establish a fuzzy deep model called the fuzzy deep belief net (FDBN) based on fuzzy restricted Boltzmann machines (FRBMs) due to their excellent generative and discriminative properties. The learning procedure of an FDBN is divided into a pretraining phase and a subsequent fine-tuning phase. In the pretraining phase, a group of FRBMs is trained in a greedy layerwise way: the first FRBM is trained by original samples, and the average values of the left and right probabilities produced by its hidden units are treated as the training data for subsequent FRBMs. The resulting FDBN is either a generative or a discriminative model depending on the choice of training a generative or a discriminative type of FRBM on top. Then, a hybrid learning approach is proposed to fine-tune this novel fuzzy deep model: the well pretrained fuzzy parameters are first defuzzified, and the FDBN with defuzzified parameters is fine-tuned by the wake–sleep or stochastic gradient descent algorithm. This hybrid strategy not only avoids learning an intractable fuzzy neural network, but also greatly improves the classification capability of the FDBN. The experimental results on MNIST, NORB, and 15 Scene databases indicate that the FDBN with the hybrid learning approach can handle high-dimensional raw images directly. It inherits the fine nature of the FRBM and outperforms some state-of-the-art discriminative models in classification accuracy. Moreover, it shows better capability of robustness than a deep belief net when encountering noisy data. [ABSTRACT FROM AUTHOR]

The restricted Boltzmann machine (RBM) is an excellent generative learning model for feature extraction. By extending its parameters from real numbers to fuzzy ones, we have developed the fuzzy RBM (FRBM) which is demonstrated to possess better generative capability than RBM. In this paper, we first propose a generative model named Gaussian FRBM (GFRBM) to deal with real-valued inputs. Then, motivated by the fact that the discriminative variant of RBM can provide a self-contained framework for classification with competitive performance compared with some traditional classifiers, we establish the discriminative FRBM (DFRBM) and discriminative GFRBM (DGFRBM) that combine both the generative and discriminative facility by adding extra neurons next to the input units. Specifically, they can be trained into excellent stand-alone classifiers and retain outstanding generative capability simultaneously. The experimental results including text and image (both clean and noisy) classification indicate that DFRBM and DGFRBM outperform discriminative RBM models in terms of reconstruction and classification accuracy, and they behave more stable when encountering noisy data. Moreover, the proposed learning models show some promising advantages over other standard classifiers. [ABSTRACT FROM AUTHOR]

The three-phase buck rectifier (3ph-BR) is suitable for applications where a voltage step-down function is required. In this paper, an improved 3ph-BR topology is proposed to reduce the voltage stress on the transistors. The freewheeling diode in the conventional topology is split into two diodes in series and the input neutral point is connected to the common point of the two diodes. With the proposed topology and the corresponding modified modulation scheme, the transistors only need to withstand the input phase voltage instead of the line-to-line voltage, bringing about the significant reduction of voltage stress. The proposed topology enables a more cost-efficient and flexible selection of the transistors. Experimental results have verified the validity of the modified topology and associated modulation scheme. [ABSTRACT FROM AUTHOR]

A matrix converter (MC) with model predictive control based on the source reactive power control usually fails to show sinusoidal source currents. The analysis presented in this paper shows that this common combination of a converter and control has the inherent inability to suppress some harmonics in the source currents, even with additional passive or active damping control. Direct source current control can be implemented to give sinusoidal source currents and intrinsic active damping. However, the issue of steady-state error in output currents then arises, as the MC topology does not allow of the independent control of source and output currents. Therefore, feedback control of load active power is proposed to address this issue without degrading the fast dynamic performance. Benefiting from the direct source current control, a simplified implementation is also proposed to decrease the number of candidate switching states from 27 to 5, which significantly reduces the computational burden. Experimental results have verified the theoretical analysis and the effectiveness of the proposed control scheme. [ABSTRACT FROM AUTHOR]