Search

Copper-loss-minimizing torque control (CLMTC) method has been widely studied to reduce electromagnetic torque ripple and minimize copper loss of a brushless dc motor (BLDCM), but still suffers challenges to improve its performance. This paper presents a quantitative analysis of stator iron loss and current-loop delay effects on CLMTC. In the analysis, the iron-loss-included torque equation that only suits for sinusoidal back electromotive force (EMF) is modified to cover the case of nonsinusoidal back EMF. It is concluded that the iron loss effect brings about significant torque generation error and the delay effect leads to torque ripple that increases linearly with the delay angle. Two improvements are proposed, one of which redefines the condition of an optimal motor current vector to compensate for the iron loss effect. The other one aims to suppress the delay effect by preshaping current reference based on delay characteristic of the loop. The improvements can be performed effectively on a microprocessor to realize a better CLMTC method with reduced torque generation error and torque ripple. Both the analysis and improvements are experimentally validated on a 0.95 kW BLDCM setup. [ABSTRACT FROM AUTHOR]

This paper extends the application of a frame-angle-based (FAB) direct torque controller (DTC) for permanent magnet synchronous motor (PMSM) drives to low and very low speeds. The FAB-DTC is intended for PMSM drives that are fed by $3\phi$ six-pulse wavelet-modulated dc–ac converters. The foundations of the FAB-DTC are based on regulating the $d\text{--}q$ -axis stator voltages, in response to variations in the load torque and/or drive speed. The adjustments in $d\text{--}q$ -axis stator voltages are created by changing the angle $\vartheta$ of the frame spanned by the $d\text{--}q$ -axis stator voltages. In order to extend the FAB-DTC to low and very low speeds, the changes in $\vartheta$ are accompanied by changes in the maximum scale. The values of maximum scale and $d\text{--}q$ -axis stator voltages are used to update or change the reference signals employed by the wavelet modulation to generate switching pulses to operate a $3\phi$ dc–ac converter. The complete PMSM drive system incorporating the modified FAB-DTC is implemented for a 10-hp PMSM drive system. The performance of the FAB-DTC is investigated for different changes in the command torque for various low and very low speeds. Performance results demonstrate stable, fast, dynamic, and accurate responses, which have minor sensitivity to variations in load torque and/or drive speed. [ABSTRACT FROM AUTHOR]

Conventional finite control set-model predictive control (FCS-MPC) applies single voltage vector within each control period. This leads to relatively high steady-state ripples and requires fast sampling rate. Additionally, enumeration-based optimal vector selection is computationally intensive. Recently, double-vector-based schemes have been proposed to improve the steady-state performance of FCS-MPC. However, they are usually complicated in vector selection and duty ratio calculation. In this paper, a universal multiple-vector-based MPC (UMV-MPC) is proposed, which achieves the same performance as the state-of-the-art double-vector-based MPC, but executes in a much more efficient and universal way. Unlike conventional FCS-MPC, enumerating process and state predictions for candidate voltage vectors are not required in the proposed UMV-MPC to select the best voltage vectors. In UMV-MPC, the optimal vectors and duty ratios are directly constructed from deadbeat control (DBC) based on space vector modulation (SVM), which is easy to follow and quick to use. The proposed UMV-MPC is not only more efficient than prior methods, but also reveals the inherent relationship between double-vector-based MPC and DBC with SVM. A comparative study of UMV-MPC and prior double-vector-based MPC is carried out in this paper. The theoretical analysis as well as simulation and experimental tests on a 2.2-kW induction motor drive are demonstrated to validate the effectiveness of the proposed UMV-MPC. [ABSTRACT FROM AUTHOR]

Flux-switching permanent magnet machine typically suffers from relatively high cogging torque due to its special doubly salient structure and high air-gap flux density, causing undesired torque ripples, as well as acoustic noise and vibration, especially at low speeds. In this paper, an analytical expression of cogging torque is derived based on a magnetomotive force-permeance model, through which the optimal design parameters and dimensions, such as combinations of stator slots and rotor poles, skewing angle, stator tooth width, rotor tooth width, and magnet thickness, can be determined. Based on the derived analytical model, two novel approaches for cogging torque reduction are proposed, i.e., asymmetric magnetomotive force and asymmetric permeance, respectively, and consequently, the optimal designs are also acquired analytically. However, it is also found that the optimal choice of the proposed techniques depends on the combinations of stator slots and rotor poles of the machines. Finally, the effectiveness of the proposed approaches is verified by both finite element analytical predictions and experimental results. [ABSTRACT FROM PUBLISHER]

Back-electromotive force (back-EMF) self-sensing is commonly used in induction machine (IM) drive systems for its maintenance cost and safety considerations. In recent years, compared with standard indirect field oriented control, deadbeat-direct torque and flux control (DB-DTFC) has been shown to be a highly effective method for IM control with a fixed switching frequency. It has advantages such as fast torque dynamics with one step dynamic response for torque and flux with no current regulator. This paper evaluates the synergy between back-EMF self-sensing technology and DB-DTFC for IM drive systems. In this paper, an observer-based closed-loop back-EMF tracking self-sensing control in an IM DB-DTFC drive system is presented. It includes a back-EMF state filter, back-EMF tracking observer, and cascaded motion observer. Back-EMF harmonic decoupling is explored to improve the low-speed performance. The band-pass filter method (BPF) for back-EMF self-sensing is also presented. Finally, the closed loop system dynamic stiffness at very low speeds, with and without the BPF method, is experimentally evaluated. [ABSTRACT FROM AUTHOR]

Permanent magnet synchronous machines (PMSMs) are popular due to their high efficiency and favorable dynamic properties. In practice, these machines often have a back-electromotive force (EMF) waveform, which deviates from a theoretically ideal sinusoid or trapezoid. In the past, current waveform shaping techniques were developed for these machines to calculate nonconventional current waveforms to reduce the torque ripple. However, in the light of global warming, energy efficiency has become more important than torque ripple in many applications. Therefore, current waveform shaping techniques can be used to focus on maximum efficiency. In this paper, a technique is presented that maximizes the efficiency of both the PMSM and the drive simultaneously. A time domain approach is used instead of the classical frequency domain approach, which allows us to include many machine aspects that had to be neglected previously, i.e., armature reaction, reluctance, zero-sequence components, and iron losses. [ABSTRACT FROM AUTHOR]

This paper presents and experimentally validates a new control scheme for electrical drive systems, named cascaded predictive speed and current control. This new strategy uses the model predictive control (MPC) concept. It has a cascaded structure like that found in field-oriented control or direct torque control. Therefore the control strategy has two loops, external and internal, both implemented with model predictive control. The external loop controls the speed, while the inner loop controls the stator currents. The inner control loop is based on Finite Control Set Model Predictive Control, and the external loop uses MPC deadbeat, making full use of the inner loop's highly dynamic response. Experimental results show that the proposed strategy has a performance that is comparable to the classical control strategies but that it is overshoot-free and provides a better time response. [ABSTRACT FROM PUBLISHER]

This paper presents a load torque estimation method for self-sensing brushless dc drives. Torque ripples in brushless dc machines can be reduced using load torque information. This method uses the terminal voltage, the virtual neutral point voltage, and the dc bus current of the machine. The algorithm uses the variation of successive back electromotive force (back EMF) samples to estimate the rotor speed. The rotor position is estimated by defining an intermediate function of estimated speed and back EMF samples. An estimate of acceleration is used to estimate load torque. The load torque information is used for increasing the dynamic stiffness of the drive. The mathematical background is given and discussed, and the simulations and the experimental results prove the performance of the proposed method. [ABSTRACT FROM PUBLISHER]

A new method of hierarchically partitioning the numerical calculations associated with the finite-element analysis of rotating electrical machinery is set forth. In this method, the device is partitioned geometrically and hierarchically into its constituent subsystems. Each Newton–Raphson iteration in a magnetically nonlinear device involves the hierarchical solution of relatively compact linear algebraic equations. This is contrasted with conventional approaches involving the formation or updating and solution of a single but large-dimensional sparse equation. In addition to an improvement in the computational efficiency, this approach facilitates efficient management of rotation. The proposed approach is demonstrated to require less than one-fourth the computation time, while producing identical results to those of a conventional formulation when solving for cogging torque and back electromotive force (back-EMF) of a permanent-magnet synchronous machine. The numerical results are also shown to be in agreement with the measured data. [ABSTRACT FROM AUTHOR]