Iet Electric Power Applications最新文献

Heat pipe is a kind of heat transfer element with very high thermal conductivity, which has a good application prospect in high power density driving motors. However, the complicated heat transfer process involves a variety of mechanisms, which undoubtedly increases the difficulty of temperature field modelling. Therefore, the equivalent thermal conductivity model of flat heat pipe (FHP) considering the influence of some factors was established. First, the mathematical model can fully consider the influence of the thickness of shell, the thickness of vapour chamber, the length of endothermic section and exothermic section on the equivalent thermal conductivity of FHP. Second, a test platform is built to test the equivalent thermal conductivity of FHP and the accuracy of established analytical model is verified. Finally, a 30-pole 36-slot high power density outer rotor permanent magnet synchronous motor (PMSM) is designed. Based on the FHP cooling method, FHPs placed in stator yoke are analysed, and the influence of different cooling structures on motor temperature rise is compared. At the same time, the effect of forced air cooling on the temperature rise of the motor is discussed, which provides a valuable reference for the application of FHP in the motor cooling.

Reactive and preventive maintenance strategies have been applied to avert transformer failures and safeguard their operations. However, these approaches have limitations of high operational downtimes, over- and under-maintenance issues, maintenance fatigue and revenue loss. The advancements in machine learning and artificial intelligence have positively altered the machine and equipment maintenance landscape. Thus, predictive maintenance (PdM), in contrast to the above-listed maintenance approaches, has laid the foundation for improving transformer maintenance by identifying incipient failures to solve the existing challenges. Recent developments in predictive maintenance of distribution power transformers have made great strides, but to solve the current challenge of accurate fault identification, this study proposed a new model architecture (DMSA CNN-LSTM) using multimodal data fusion to address anomaly detection. A classification accuracy, F1-score, precision and recall of 0.9917, 0.9714, 0.9781 and 0.9647, respectively, were produced on a fused multimodal dataset at a computational time of 619.898 s. The performance was afterwards evaluated against other state-of-the-art benchmark models. The significance of this study lies in providing a scalable data-driven architecture suitable for real-time deployment in providing predictive solutions for transformers at a higher performance efficiency. This approach leverages deep neural networks that provide a comprehensive diagnostic and prognostic approach to mitigate transformer faults and breakdowns.

The multi-vector-based direct model predictive torque control (DMPTC) inherits the high control precise of vector control and the fast dynamic response of model predictive control. However, the cascade optimisation mode is difficult to guarantee the global optimal performance of the voltage vector (VV) combination, and results in invalid VV time duration in some cases as well as high calculation complexity. In this paper, for the predictive control of permanent magnet synchronous motors widely used in electric propulsion aircraft, a parallel optimisation DMPTC method is proposed to ensure that the multi-vectors applied are all optimal in each control cycle. The optimal double vectors are directly chosen locating the sub-region of the sector of reference VV. With only the absolute value of stator VV error term, the time duration is always in the valid range and robust to motor parameters. In addition, a cost function with a single control objective is designed to control the torque and flux cooperatively, eliminating the stator flux weighting factor and reducing the computation burden. The proposed method is compared with the recent predictive torque control strategies. The experimental results confirm the effectiveness of the proposed method in improving the dynamic response and steady-state performance while reducing computation burden and complexity.

Servo controllers are essential components in robotics, manufacturing, and various industrial applications. However, achieving fast and accurate reference tracking in servo systems remains challenging due to modelling uncertainties and external disturbances. In this paper, a hybrid control strategy is proposed that combines a Linear Quadratic Regulator (LQR) state-feedback controller with deep learning to address these challenges. The LQR controller utilises system state measurements to optimise the control input, while the integration of a deep neural network enhances accuracy and dynamic response by adapting to changing system conditions. This approach provides robust control performance, effectively mitigating the impact of uncertainties and disturbances on servo system behaviour. The proposed method was validated using AC servo motors, among the most common servo systems, though the approach is adaptable to other servo-like systems. Comparative evaluations are conducted against existing methods, including SIMC-SMC, 2DOF-IMC-SMC, 2DOF-IMC-PID, and SIMC-PD controllers, focusing on the angular position control of a servo motor. Simulation results demonstrate that the proposed controller outperforms these methods in terms of robustness, precision, and disturbance rejection. These findings highlight the potential of the proposed LQR-deep learning framework to significantly improve servo system performance across a wide range of applications.

Recently, there has been a growing interest in new permanent magnet (PM) motor topologies. However, these newly developed PM motors are still in their early design stages and face various challenges. In order to overcome these drawbacks, further development of these motor topologies is necessary. One such new motor topology is the ring winding axial flux permanent magnet (RWAFPM) motor. Enhancing the performance of this motor requires optimising its geometry, which can be a time-consuming process when using the three-dimensional (3D) finite element method (FEM). It is essential to propose a two-dimensional model that offers faster processing but lower accuracy to address this issue compared to 3D FEM for this motor. Due to the three-dimensional structure of ring-winding axial flux machine|axial flux motors, creating a two-dimensional (2D) model for this motor presents a significant challenge. In this article, while studying the RWAFPM motor, the simplifications to the motor's geometry have been proposed in order to create a 2D model that provides sufficient accuracy as a substitute for 3D FEM. To validate our findings, the Schwarz–Christoffel conformal mapping technique has been employed to extract the modelling results. Finally, these findings have been compared with the outcomes obtained from the 3D simulation. Validation. Validation of 3D finite element results has been done by experimental results.

This paper proposes a multilayer analytical model for calculating the magnetic field and equivalent circuit parameters of solid rotor induction motors (SRIM), and the B–H curve of rotor material is used to consider the influence of saturation effect of solid rotor. The equivalent circuit of SRIM based on surface impedance theory can fully consider the influence of rotor material. Then, the magnetic vector potential in the rotor is solved in the three-dimensional electromagnetic field, and the rotor is segmented along the axial direction to obtain the end coefficient considering the axial non-linear effect of the rotor. The air gap magnetic field distribution, rotor loss, rotor impedance and torque were analysed by finite element simulation, and the accuracy of the multilayer analytical model and end coefficient was verified. Compared with the finite element method, the analytical method proposed in this paper is time saving, resulting in that it is suitable for the optimised design. Finally, an experimental prototype is developed and tested, confirming the accuracy of both the finite element simulation and the analytical model.

The T-type three-level rectifier has garnered significant attention due to its ability to enhance the voltage waveform quality in power systems and reduce electromagnetic interference with other equipment. To ensure high reliability in high-power wind and photovoltaic power generation systems, conducting fault diagnosis for T-type three-level rectifiers is crucial. This paper first analyzes the input current characteristics of both in-phase and out-of-phase dual transistor open-circuit faults. A current-extended observer is developed using a hybrid logic dynamic model to estimate the current value during normal operation. The residual state equation is derived by comparing this estimated current with the actual fault current. A residual information table for fault currents is created through differential solutions of the residual state equation, enabling fault localisation by comparing the residual information against predefined threshold values. Finally, the feasibility and accuracy of the proposed fault diagnosis method are validated through simulations and experiments.

The acoustic signals of power transformers serve as key indicators for assessing operational statuses and detecting internal mechanical issues. However, low-frequency noise, such as fan noise, often obscures critical features. This study introduces an adaptive low-frequency denoising algorithm based on the filtered-x least mean square (FxLMS) model. By optimising the step size and convergence factor, the algorithm resolves the peak offset issue in traditional LMS methods, enhances denoising performance and computational efficiency, and demonstrates effectiveness in practical scenarios. Using denoised acoustic data, quantitative analysis based on information entropy (frequency complexity analysis (FCA)) evaluates changes in mechanical properties. The analysis indicates that healthy transformers exhibit lower FCA values, whereas aged transformers show values approximately double those of healthy units, reflecting the mechanical changes associated with normal ageing. Further analysis of transformers commissioned in 2004 reveals that FCA values of abnormally aged transformers exceed three times those of typically aged units, indicating severe mechanical degradation. These findings demonstrate that combining the FxLMS algorithm with FCA analysis effectively extracts denoised acoustic features and distinguishes between healthy states, normal ageing, and abnormal ageing in transformers.

In order to reduce the transformer winding hot spot temperature (HST), a dynamic radial basis function surrogate model optimisation strategy is proposed to optimise the structure of the forced oil circulation transformer winding's block washers in this paper. The whale optimisation algorithm (WOA) is introduced to obtain the optimal hyperparameter, and the WOA-DRBF surrogate model optimisation strategy is used to improve the fitting and optimisation ability. After optimisation, the HST corresponding to the size of the block washers decreased by 3.67°C, and the maximal temperature rise decreased by 12.93%, effectively ameliorating the phenomenon of excessive local temperature rise within each partition. Comparing the above optimisation results with those of the WOA-RBF static surrogate model and the genetic algorithm (GA), the proposed method shows superior performance in both optimisation search capability and efficiency. Comparing the optimisation efficiencies of the three methods, the number of model calls for the WOA-DRBF method is 23.89% of the WOA-RBF method and 1.77% of the GA method, respectively. The proposed method significantly enhances optimisation efficiency while improving global optimisation search capability. It offers a feasible solution for efficiently and accurately solving the optimisation problem of the transformer winding's block washer.

Harmonic injection matrix converter (HIMC) achieves direct AC-AC power conversion with maximum voltage transfer and reduced transistor capacity. However, the existing open-loop current control method for HIMC may bring in low-order harmonics if the injected harmonic current deviates from its desired value. To address this issue, this paper proposes a closed-loop input current control strategy based on the waveform similarity index (WSI) for these emerging MC topologies. The WSIs are defined as the ratios of the corresponding input currents against the maximum and minimum input voltages. Multi resonant controllers are adopted to regulate the WSI difference so that its low-order harmonics are eliminated, which generate an additional signal to correct the injected harmonic current. By adopting the proposed current control strategy, the 5th and 7th harmonics are suppressed to lower than 1.0% under the sinusoidal input voltages and nominal load scenario. Even with inaccuracies in the reference harmonic current generation, the proposed strategy enables the input currents to approach sinusoidal waveforms with significantly reduced WSIs difference, thus enhancing the input power factor over the conventional open-loop method. Besides, the proposed method is effective under unbalanced and distorted conditions through the experimental verification.