Hybrid rotor position observer for wide speed-range sensorless PM motor drives including zero speed

This paper addresses the problem of wide speed-range sensorless control of a surface-mount permanent-magnet (SMPM) machine including zero-speed operation. A hybrid structure integrating a flux observer and signal-injection techniques is proposed, which results in a rotor position signal independent of motor parameters at low and zero speed. Although the SMPM machine typically has a very low geometric saliency, the injection technique is effective in tracking the saturation-induced saliency produced by the stator flux. Experimental results are presented showing an excellent performance for both the sensorless speed and position control using an off-the-shelf SMPM machine.     

Anisotropy Comparison of Reluctance and PM Synchronous Machines for Position Sensorless Control Using HF Carrier Injection

Position sensorless control of reluctance and permanent magnet synchronous machines at zero and low speed is possible using HF voltage injection and proper demodulation. The so-called saliency position, which is tracked by the HF sensorless scheme, is different from the actual rotor position: the difference contains both offset and rotor-position-varying components, which may be explained by carefully considering the HF behavior of the machine and the effect that fundamental excitation and rotor position have upon it. This paper gives insight into the HF behavior of synchronous machines and serves as a practical guide for implementation of stable and robust position estimation at zero and low speed.      

Suppression of saturation saliency effects for the sensorlessposition control of induction motor drives under loaded conditions

This paper presents an automated commissioning procedure used for the elimination of the saturation saliency effects in the sensorless position control of field-orientated cage induction motor drives. The position control itself is based on extracting a rotor position estimate from a high-frequency signal injection interacting with natural or engineered rotor position saliencies within the machine. The paper shows that this estimate cannot be robustly or accurately obtained if saturation saliencies are present. The paper introduces a method for suppressing the effects of the saturation saliency through information gained in a prior commissioning procedure. The effectiveness of the procedure is demonstrated through experimental results showing both good suppression of the saturation harmonics and true sensorless position control under high load torques     

无速度传感器控制的外转子永磁同步电动机的优化设计

机车牵引、启动发电机和直接驱动等诸多应用场合都需要比较大的转矩。本文介绍了外转子永磁同步电动机和无速度传感器控制算法的优点，特别是使用INFORM方法时低速和零速时的性能。因此必须考虑饱和效应，并用于增强INFORM方法的性能。     

Comparative analysis of experimental performance and stability of sensorless induction motor drives

This paper compares the experimental performance of three flux and speed observers for speed-sensorless induction motor drives and discusses the cause of their differences. The small signal analysis using the linearized model is carried out to analyze stability. Three methods are generally accepted to be representative candidates for high sensorless performance, namely: 1) rotor-flux model reference adaptive system (MRAS); 2) torque-current MRAS; and 3) adaptive nonlinear flux observer. Many other sensorless methods improved these methods. The paper discusses baseline conditions for the experiments and the stability analysis, which include matched load inertia, specified speed estimator dynamics, and sensorless operation within a speed control loop. For the comparison tests in the paper, the speed estimation dynamics of the methods are the same; this is important for parameter sensitivity. The paper concentrates on the low-speed performance, and all results shown are under sensorless speed control.     

Sensorless torque control of salient-pole synchronous motor atzero-speed operation

A position and speed sensorless control using the counter electromotive force of a permanent-magnet motor (PM) debases the control performance at a low speed. We propose a controllable system at full speed, including a zero speed using saliency. At low speed, the sensorless control is made by observing a current ripple at a time when alternating voltage has been applied to a salient-pole motor. Also, for discriminating the S and N poles of the magnet, magnetic saturation is used. A device has been applied to the motor rotor to allow the magnetic saturation to come about easily. Furthermore, at a time of high speed, drive at a full-speed range has been accomplished by switching smoothly over to a sensorless driving system making use of counter electromotive force. All algorithms are implemented by software, and this system can operate successively from starting to high-speed operation. The paper discusses the operational principles at a low speed, analysis and experimental results, the control scheme, how to changeover the control mode at high speed, and the experimental results     

Sensorless position control of induction motors-an emergingtechnology

Concepts for the sensorless position control of induction motor drives rely on anisotropic properties of the machine rotor. Such anisotropies can be incorporated as periodic variations of magnetic saliencies in various ways. The built-in spatial anisotropy is detected by injecting a high-frequency flux wave into the stator. The resulting stator current harmonics contain frequency components that depend on the rotor position. Models of the rotor saliency serve to extract the rotor position signal using phase-locked loop techniques. A different approach makes use of the parasitic effects that originate from the discrete winding structure of a cage rotor. It has the merit of providing high spatial resolution for incremental positioning without sensor. The practical implementation of sensorless position identification and of a high-accuracy position control system are reported     

A general algorithm for speed and position estimation of AC motors

A computationally efficient speed and position estimation algorithm, generally applicable to AC motor drives, is designed and analyzed. Applications include: (a) sensorless permanent-magnet and reluctance synchronous motor drives using the fundamental excitation as information source; (b) sensorless drives using saliency and signal injection; and (c) sensored drives using resolvers. Particular attention is given for case (a). Low parameter sensitivity in the entire speed range (except at low speeds for the reluctance motor)-implying a small position estimation error-and good dynamic properties at nominal speeds are verified     

MRAS Observers for Sensorless Control of Doubly-Fed Induction Generators

This paper addresses the analysis and performance of several model reference adaptive system (MRAS) observers for sensorless vector control of doubly-fed induction machines. Small signal models allow the formal analysis of the observers for a given dynamic. The performance of each MRAS observer is analyzed, considering grid-connected and stand-alone operation. The MRAS observers are implemented in a 3.5 kW experimental prototype composed of a doubly-fed induction generator and a wind turbine emulator. Experimental results validate the predictions of the small signal models and demonstrate the performance of the sensorless methods during both steady state and variable speed wind energy generation.     

面向显著性目标检测的SSD改进模型

传统显著性目标检测方法常假设只有单个显著性目标,其效果依赖显著性阈值的选取,并不符合实际应用需求。近来利用目标检测方法得到显著性目标检测框成为一种新的解决思路。SSD模型可同时精确检测多个不同尺度的目标对象,但小尺寸目标检测精度不佳。为此,该文引入去卷积模块与注意力残差模块,构建了面向多显著性目标检测的DAR-SSD模型。实验结果表明,DAR-SSD检测精度显著高于SOD模型;相比原始SSD模型,在小尺度和多显著性目标情形下性能提升明显;相比MDF和DCL等深度学习框架下的方法,也体现了复杂背景情形下的良好检测性能。     

A self-tuning closed-loop flux observer for sensorless torquecontrol of standard induction machines

This paper proposes a self-tuning closed-loop flux observer, which provides field-oriented torque control for induction machines without a tachometer. The proposed algorithm combines the best features of harmonic detection and stator voltage integration through the use of a new tuning scheme. The observer accuracy and robustness is augmented by a parameter-independent accurate-speed detector, which analyzes magnetic saliency harmonics in the stator current. The harmonic-detection scheme provides accurate rotor-speed updates during steady-state operation down to 1-Hz source frequency. This additional speed information is used to tune the rotor-resistance parameter of the observer. The tuned observer exhibits improved dynamic performance, accurate steady-state speed control and an extended range of control near zero speed. The algorithm requires no special machine modifications and can be implemented on most existing low- and medium-performance drives. The closed-loop nature of the flux observer, combined with the harmonic-detection scheme, provides flux and speed error feedback, which significantly increases the robustness of sensorless control across the entire speed range     

Saliency detection for panoramic landscape images of outdoor scenes

Saliency detection has been researched for conventional images with standard aspect ratios, however, it is a challenging problem for panoramic images with wide fields of view. In this paper, we propose a saliency detection algorithm for panoramic landscape images of outdoor scenes. We observe that a typical panoramic image includes several homogeneous background regions yielding horizontally elongated distributions, as well as multiple foreground objects with arbitrary locations. We first estimate the background of panoramic images by selecting homogeneous superpixels using geodesic similarity and analyzing their spatial distributions. Then we iteratively refine an initial saliency map derived from background estimation by computing the feature contrast only within local surrounding area whose range and shape are changed adaptively. Experimental results demonstrate that the proposed algorithm detects multiple salient objects faithfully while suppressing the background successfully, and it yields a significantly better performance of panorama saliency detection compared with the recent state-of-the-art techniques.     

飞轮电机速度的无传感器控制方法

简述了飞轮电机的无传感器控制的基本原理,分类介绍了假定旋转坐标法、模型参考自适应系统、观测器、扩展的卡尔曼滤波、滑模观测器、高频信号注入法等各种用于飞轮电机速度无传感器控制的方法、效果及研究应用状况.由控制效果可见,用无传感器控制可以使飞轮电机的整体性能得到显著改善,是高性能飞轮电机调速系统开发的一个重要方向.     

Use of a matrix converter to enhance the sensorless control of a surface-mount permanent-magnet AC motor at zero and low frequency

This paper addresses the use of a matrix converter (MC) with high-frequency voltage injection for zero- and low-speed sensorless control of surface mount permanent magnet alternating current motors. This paper also proposes a dual compensation technique that can practically eliminate zero-current-crossing distortions in MCs, discusses the effect of such distortions on sensorless control using injection techniques, and presents experimental results that illustrate the enhanced low-speed sensorless performance that can be obtained from exploiting the effective linearity of the MC operation.     

Two-layer average-to-peak ratio based saliency detection

Visual saliency plays an important role in pattern recognition tasks such as rapidly seeking prominent regions in a complex scene to discover the meaningful objects. In this paper, we present a new method to detect visual saliency from an image. This saliency is modeled as two parts, i.e., average-to-peak ratio (APR) saliency and chrominance-aware (CA) saliency. The first term is designed to describe the global contrast, which is computed based on pixel-level saliency maps. To compute the CA saliency, the luminance component is first removed by subtracting it from each color channel. Then the difference follows the APR saliency computation. Finally, a two-layer saliency model is built by combining the two saliency maps. To evaluate our proposed method, we do extensive experiments on three well-known image data sets including MSRA image set, PASCAL VOC image set, and human fixation dataset. Experimental results show that our method outperforms the state-of-the-art methods and achieves the good performance on the visual saliency detection task.     

基于加权的K近邻线性混合显著性目标检测

显著性目标检测旨在于一个场景中自动检测能够引起人类注意的目标或区域,在自底向上的方法中,基于多核支持向量机(SVM)的集成学习取得了卓越的效果。然而,针对每一张要处理的图像,该方法都要重新训练,每一次训练都非常耗时。因此,该文提出一个基于加权的K近邻线性混合(WKNNLB)显著性目标检测方法:利用现有的方法来产生初始的弱显著图并获得训练样本,引入加权的K近邻(WKNN)模型来预测样本的显著性值,该模型不需要任何训练过程,仅需选择一个最优的K值和计算与测试样本最近的K个训练样本的欧式距离。为了减少选择K值带来的影响,多个加权的K近邻模型通过线性混合的方式融合来产生强的显著图。最后,将多尺度的弱显著图和强显著图融合来进一步提高检测效果。在常用的ASD和复杂的DUT-OMRON数据集上的实验结果表明了该算法在运行时间和性能上的有效性和优越性。当采用较好的弱显著图时,该算法能够取得更好的效果。     

图像显著性检测方法解析

图像显著性检测是一种通过对图像颜色、强度、方向等特征进行分析生成图像显著性图的技术。其生成的显著性图可以用于图像分割、图像压缩以及图像识别等图像处理领域,从而改善图像处理的性能。为了对图像显著性检测技术及其发展有一个全面深入的了解,使用文献研究法和比较研究法对其概念及方法进行了探究。针对几种具有代表性的图像显著性检测算法进行了简要的概述和分析,用流程图简明扼要地表示显著性检测算法的基本框架。研究结果显示,图像显著性检测技术的效率在不断提升,算法越来越多样化,在图像处理领域的应用越来越广泛,这些对于图像处理自动化具有重要意义。     

无速度传感器感应电机控制系统的现状与展望

新技术的发展给电机控制行业带来新的发展机遇。从电机控制器、无速度传感器技术方面,介绍了国内外无速度传感器感应电机控制系统中这两个领域内的研究现状及发展趋势。回顾了DSP处理器的发展历程,介绍了DSP的发展现状,从设计和应用角度分析了DSP的特点,介绍了DSP的发展前景;分析了无速度传感器技术的研究现状,指出电机低速下速度估计性能差及低频范围稳定运行问题是目前无速度传感器技术的研究热点,并提出了相应的解决方法。     

基于区域协方差的视频显著度局部空时优化模型

显著度检测在计算机视觉中应用非常广泛,图像级的显著度检测研究已较为成熟,但视频显著度因其高度挑战性研究相对较少。该文借鉴图像级显著度算法的思想,提出一种通用的空时特征提取与优化模型来检测视频显著度。首先利用区域协方差矩阵构造视频的空时特征描述子,然后计算对比度得出初始显著图,最后通过联合前后帧的局部空时优化模型得到最终的显著图。在2个公开视频显著性数据集上的实验结果表明,所提算法性能优于目前的主流算法,同时具有良好的扩展性。