Iet Electric Power Applications最新文献

A Taguchi-based robust design strategy is proposed to minimise the torque ripple of a 6-slot/2-pole modular high-speed permanent magnet motor in mass production, accounting for manufacturing tolerances of split gap (Δg), misalignment (Δm), and offset angle (Δα). Firstly, the effects and interactions of manufacturing tolerances are calculated, indicating that Δg has the highest effect followed by Δm, positive Δg and negative Δm have a strengthening effect, and Δα has no effect, and subsequently, the worst-case scenario of manufacturing tolerances with the highest torque ripple is obtained. Afterwards, tooth circumferential positions are optimised for minimising torque ripple without jeopardising average torque, considering the tradeoff between the cases without manufacturing tolerance and with the worst-case scenario of manufacturing tolerances. As will be demonstrated, torque ripples are reduced significantly, that is, they are particularly reduced by 40% in the worst-case scenario. Under hypothetical 100 sets manufacturing tolerances as Gauss distributions, the optimised machines have significantly reduced torque ripples (maximum and average reductions are 33% and 16%, respectively) with more concentrated distribution. The correctness of the methods is verified by experimental validation.

Sensorless induction motor control has been widely applied to the rail transit field. However, achieving a safe stop of a train using electric braking without applying air braking has been an urgent problem to be solved. The current research only considers the stability of the speed identification in the low-speed region and does not consider the impact of inaccurate parameters on the stability, which cannot ensure the stable braking and parking of the train under all working conditions. To address this problem, the coupling relationship between the motor speed and the stator resistance is used and an adaptive rate of them is designed based on the Lyapunov stability design law. In addition, aiming to reduce the torque ripple, a torque ripple elimination link is designed to cancel the torque ripple caused by the small-signal injection. Experiments show that the proposed parallel identification strategy of speed, stator resistance, and rotor resistance can ensure the system operation stability in the low- and zero-speed regions without increasing the torque ripple.

In the in-service induction motors (IMs), friction and windage losses (FWL) value should be determined non-intrusively. Hence, in the in-service IMs, instead of measuring FWL, empirical equations are used to estimate FWL value. A novel technique is proposed for estimating the FWL value in low-voltage three-phase IMs based on obtained data from applying the no-load test on 425 simulated IMs in the MATLAB software. The simulated IMs are 380 V, 50 Hz, with different numbers of poles in the power range of 0.37–400 kW. The FWL value for simulated IMs is calculated based on the IEEE 112 standard and by applying the no-load test. Then, based on the dispersion of the obtained data from the no-load test and using non-linear regression, according to the number of IM poles for each number of poles, a third-degree equation for FWL estimation is fitted to the test data. The proposed method to estimate the FWL value only needs the nominal output power listed on the IM nameplate. Also, unlike existing empirical relationships, the proposed approach estimates the FWL value for IMs with high accuracy and non-intrusively. The effectiveness of the suggested technique is confirmed by simulation and practical results.

The authors present a model for diagnosing motor faults based on machine learning, demonstrating advantages over other algorithms in terms of both improved fitness values and reduced running time. The structure of the model involves three primary phases: feature extraction, feature selection and classification. During the feature extraction phase, crucial features are identified using empirical mode decomposition, fast Fourier transform and multiresolution analysis, resulting in a total of 144 features. The feature selection stage employs a new strategy that combines symmetrical uncertainty in the filter approach with the binary grey wolf optimiser and emperor penguin optimiser in the wrapper approach. Finally, a support vector machine is used for classification to generate fitness values. To validate the model's effectiveness and accuracy, motor fault current signal datasets, case Western Reserve University (CWRU) benchmark datasets and mechanical failure prevention technology benchmark datasets are utilised. In the motor fault current signal dataset, the highest average accuracy achieved is 99.95%, with a minimum average running time of 88.02 s obtained under ∞dB conditions. Regarding benchmark datasets and mechanical failures at CWRU, using the prevention technology benchmark dataset resulted in classification accuracies of 99.54% and 99.52%, respectively. Comparative analysis with traditional algorithms reveals that symmetric uncertainty and emperor penguin–grey wolf optimisation model outperforms traditional models in terms of performance.

A novel method is presented for determining the derating factors of a three-phase induction motor under the condition of the unbalanced supply voltages. In this method, a mechanical system is used which consist of the a centrifugal pump, two valves, a DC motor, which are connected to the shaft of the three-phase induction motor. A sliding mode control system is used for position control of the DC motor for adjusting the valve angle for derating the induction motor. The authors present the results of an experiment in which a three-phase induction motor was subjected to various unbalanced voltage conditions. The results of simulations were used to look into what happened when there were different levels of imbalanced voltage. This was done to determine how these situations changed an induction motor's speed, torque, and efficiency. For this system, the stator current would be greater than the rated current if there was an imbalance in the supply voltage. Therefore, to reduce the amount of power that the three-phase induction motor can produce, the control system uses a DC motor to reduce the angle of one of the two valves. This decreasing angle continues until the root mean square value of the stator current returns to the rated current. At this point, the derating factor may be calculated by dividing the output power of the three-phase induction motor in the unbalanced condition by the output power when there are ideal sinusoidal. The MATLAB SIMULINK environment is utilised to perform simulations of the proposed system.

2-D air-gap magnetic field distribution is the essential prerequisite of electrical machine analysis. Because of the natural periodicity of rotary machines, Fourier analysis is a suitable choice for air-gap field prediction. However, due to the end-effect, periodicity is not present in linear machines and the Fourier series is not appropriate. In engineering mathematics, a rigorous method of solving partial differential equations is the separation of the variables method (SVM) which is based on the hypothesis that the field solution is in the form of the product of two functions of orthogonal directions; for example, in an axisymmetric structure, a longitudinal harmonic function (LHF) and a radial harmonic function (RHF). A particular case of these functions is trigonometric functions, which result in the Fourier series. SVM is imposed in a different manner, where RHF is approximated by the Bessel–Fourier series and consequently LHF has a (piecewise) exponential behaviour. This choice of basis functions not only serves the purpose of modelling the end-effect but also removes the Gibbs phenomenon, that is, it precisely models discontinuities of flux density at the surface of PMs. A numerical case-study of a slotless double-sided linear tubular surface-PM machine shows that the piecewise exponential approximation can attain 1% error while utilising only three harmonics, whereas the trigonometric approximation, due to the Gibbs phenomenon, slowly arrives at 20% error with 100 harmonics for approximating flux density. Results of both methods were validated using the finite element method (FEM). Additionally, it was discovered that the computational complexity of the model is independent of varying position; therefore, the analytical method could yield the back electromotive force and electromagnetic thrust force for a number of 320 positions at merely 0.2 s while the FEM software required 5 min.

In view of the cost and reliability issues caused by mechanical position sensors, position sensorless control (PSC) of permanent magnet synchronous motors (PMSMs) has received widespread attention. A series of rotor position estimation (RPE) schemes have been developed, in which the phase-locked loop (PLL) plays an important role. However, the conventional phase-locked loops face two thorny issues, one is the degradation of RPE performance under motor acceleration and deceleration conditions, and the other is high noise sensitivity. To this end, an adaptive gain PLL (AG-PLL) is proposed for PSC of PMSM drives. The gains of the loop filter in the proposed AG-PLL are adaptively adjusted with the estimation error in order to find a compromise between the estimation performance under motor acceleration and deceleration conditions and the noise sensitivity. Moreover, the proposed scheme avoids increasing the order of the PLL, so its dynamic performance is also guaranteed. The proposed scheme is tested based on the hardware-in-the-loop platform.

An analytical and optimisation method is presented to improve the air-gap magnetic field of the surface-mounted permanent magnet machine with multi-level array magnets, which can reduce air-gap magnetic field harmonics and torque ripple. Different from the conventional analytical method (AM) around the entire single-magnet pole, the modelling method of multi-level equal-thickness ferrite magnets with different remanences for improving the air-gap flux density is investigated. Considering the influence of air-gap relative permeability by stator slots and multi-level magnet remanences on the air-gap flux density, the proposed AM for improving air-gap flux density is derived by analysing the slotless air-gap flux density and slotted air-gap relative permeability. The cogging torque is predicted with energy method by analysing the tangential magnetic field distribution at stator slots. Then, the back electromotive force is predicted by analysing the winding distribution and radial air-gap flux density. Based on the proposed AM, the simulated annealing particle swarm optimisation algorithm is used by optimising the widths and thicknesses of the multi-level array magnets to further reduce torque ripple and increase output torque. Finally, the accuracy of the proposed AM is verified by comparing with the finite element method.

Insulation deterioration, which is mainly caused by partial discharge (PD) occurring inside power transformers, is one of the prime reasons to cause transformer faults. Therefore, an effective diagnosis of PD is crucial to ensure the safe and stable operation of transformers. To extract more effective features that characterise transformers PD signals and enhance the recognition accuracy, a novel transformer PD fault diagnosis model based on improved adaptive local iterative filtering (ALIF) and bidirectional long short-term memory (BILSTM) neural network is proposed. Addressing the issue of predetermined decomposition levels and accuracy in ALIF decomposition, the golden jackal optimisation (GJO) algorithm is introduced to optimise the parameters. The proposed fault diagnostic model extracts dominant PD features employing the improved ALIF and Refined Composite Multi-Scale Dispersion Entropy and improves the diagnostic accuracy with the optimised BILSTM by introducing GJO. Experimental data evaluates the performance of support vector machine, long short-term memory and BILSTM. The results verify the effectiveness and superiority of the proposed model.

The authors present a low-loss dual-permanent-magnet-excited vernier (DPMEV) machine with segmented stator design to meet the requirements of low iron loss of electric aircraft based on the field modulation theory. The stator topology features and losses of both the original and segmented DPMEV machines are comparatively investigated. Then, the armature air-gap flux density is deduced by the magnetic motive force-permeance model, and the influence of each harmonic on the losses are analysed. It is found that the harmonic which produces losses in the original machine reduced greatly after introducing the segmented structure. Furthermore, the electromagnetic performances of two DPMEV machines are comparatively analysed by finite element analysis. Finally, two prototypes of the original and segmented DPMEV machines are built and tested to verify the theoretical analysis.