Search

Fault diagnosis is essential for the safe operation of a permanent magnet synchronous machine (PMSM) system. At present, the fault diagnosis method based on deep learning has been gradually studied due to its ability to automatically extract fault features and an end-to-end diagnostic model. However, this method often uses a single-source signal, resulting in poor fault diagnosis performance. On the other hand, deep learning model typically requires large amounts of storage and computational resources, especially when processing multisource signals. Hence, this article proposes an improved fault diagnosis method for PMSM systems based on lightweight multisource information data layer fusion. In this method, the original multisource information is fused at the data layer. Then the fused information is input into the constructed lightweight deep learning model to implement fault diagnosis of PMSM system, where a one-dimensional convolutional neural network with the combination of depth-separable convolution and global average pooling is first presented to reduce the computational time and complexity of the fault diagnosis method. The simulation and experimental results show that the proposed fault diagnosis method can achieve accurate fault identification of the PMSM system and can reduce the requirement for storage and computational resources, indicating the effectiveness of the proposed fault diagnosis method.

Partial demagnetization fault (PDF) is a common problem for permanent magnet synchronous motor (PMSM). The PMSM usually operates under complex working conditions (dynamic speed and various load), leading to the difficulty in robust PDF diagnosis. Hence, how to reliably and accurately diagnose PDF under complex working conditions has become a key issue in ensuring its safe operation. To address this issue, a robust PDF diagnosis method for PMSM is proposed based on radial air-gap flux density in this article. First, the d-axis magnetic network model of PMSM is established to extract the fault feature from the radial air-gap flux density. Then, by subtracting the offline-calculated radial air-gap flux density of stator current excitation from the online measured value, the open-circuit radial air-gap flux density can be calculated. Next, the equiangular interval resampling method is used to obtain the open-circuit radial air-gap flux density with respect to rotor position angle. Finally, through analysis of open-circuit radial air-gap flux density in angular coordinates, not only PDF can be detected and the fault degree can be estimated, but also the demagnetized PM can be located under complex working conditions. Both the simulation and experimental results validate the effectiveness of the proposed PDF diagnosis method.

The leaf-style slotless resolver has been proposed in the form of the stator permanent magnet (PM) variable reluctance resolver. The prototype is specially designed only for high-speed scenarios, as the PM cannot generate a significant motion electro-motive force (EMF) at zero- and very low speed. This article proposes a more practical prototype and its dedicated system, which has two valuable features. First, the resolver size and cost can further reduce to an ignorable level, including only two tiny pieces of iron and a small printed circuit board. The electrical high-frequency excitation is adopted instead of the magnets; then the decoding system can cover the speed range from zero to several kilo r/min. However, the limited space and slotless structure weaken the excitation and EMF signals. The square waveform injection is thus introduced to solve this issue. The hardware circuit and software computation requirements are friendly to be implanted into a cost-sensitive motor driver. The proposed resolver system has been integrated into a servo motor drive system and tested in experiments.

The function analysis method based on the optimal turn-on angle is one of the most popular efficiency optimization control strategies for switched reluctance motor (SRM) system, but it still has certain shortcomings. Under certain conditions, the optimal turn-on angle proposed by existing research is available, but its analytical formula has either low accuracy or complex solution. In addition, when the operating condition is changed, resulting in the rising rate of the phase flux-linkage at full-voltage is less than the slope of the reference flux-linkage at θm, the optimal turn-on angle proposed by the existing research will be unavailable. Hence, to solve these issues, this article presents an improved efficiency optimization control method of SRM based on a redefined optimal turn-on angle and its corresponding analytical formula. Based on the flux-linkage analysis, the SRM system is divided into two modes according to different operating conditions, with each mode corresponding to different optimal turn-on angles. In mode Ⅰ, the optimal turn-on angle proposed by the existing research is available and still in used. And this article presents a new analytical formula for this optimal turn-on angle, which has high accuracy, convenient solution, and simple structural form compared to the existing analytical formulas. In mode Ⅱ, the existing optimal turn-on angle is unavailable. Thus, this article redefines a new optimal turn-on angle in mode II and also presents its corresponding analytical formula. The analytical formulas for the two modes are similar in form and both are deduced based on the flux-linkage calculation. The simulations and experiments are conducted to verify the effectiveness of the proposed efficiency optimization control strategy.

This article proposes a simplified loss minimization control (LMC) method for interior permanent magnet synchronous machine (IPMSM) by using a virtual current signal. In the proposed method, the LMC problem caused by complex calculation of higher order nonlinear equations due to the unique structure of IPMSM is solved, and the LMC current reference can be directly obtained, making the LMC method can be easily implemented. First, according to the mathematical model of the IPMSM considering iron loss, a virtual current signal iV, which is linear with d- and q-axes active current, is constructed. Next, the expressions of d- and q-axes current references with respect to iV are derived to minimize the total electrical loss. In addition, to further show the performance of the proposed LMC method, the traditional efficiency optimization methods are studied for comparative analysis. Finally, the studies are tested under various operating conditions on a 2.2-kW prototype, and the results show that the proposed LMC method has good efficiency optimization performance and good dynamic performance.

This article proposes an open-circuit fault diagnosis method for parallel inverter switches based on circulating current. The mixed logic dynamic (MLD) model of the parallel inverter system is established. Based on this model, a Luenberger observer is applied to estimate the circulating current of the parallel inverter system. The observed value of the circulating current is compared with the actual value of the circulating current to obtain the residual signal. By using the residual signal, the faulty switches are detected and located. In addition, the reasonable thresholds of the residual signal are set to eliminate the interference of other factors on the fault diagnosis results, which enhances the robustness of the proposed fault diagnosis method. The experimental results verify the effectiveness of the proposed fault diagnosis method. Additionally, the proposed fault diagnosis method does not require additional devices or equipment, and has low cost and good practicability.

This article investigates the influences of the rotor design on air-gap field modulation (AGFM) theory in spoke-type permanent magnet (STPM) machines for traction applications. Due to flux-concentrated effect, STPM machines exhibit abundant air-gap field density harmonics and relatively high torque density. The Maxwell stress tensor (MST) method is adopted to identify and quantify the torque contributions of the various air-gap field harmonics in STPM machines. It reveals a fact that, except the main-order working harmonic is responsible for average torque generation, abundant modulated air-gap field harmonics contribute a non-negligible torque component. Besides, the influences of rotor open slot and PM arrangements on AGFM effects in STPM machines are evaluated and compared by finite element (FE) analysis, which identifies that the rotor slot and PM arrangement designs show a more conspicuous effects on their torque components and AGFM effect. Finally, the experimental results on a STPM prototype with rotor closed slot design are carried out to validate the theoretical and FE analyses.

This article presents a novel flux concentrated stator consequent-pole permanent magnet (FCSCPM) machine with nonuniform stator modulation teeth (NSMT) design, which provides a relative lower magnetic reluctance circuit for low-order working harmonics and hence resulting a higher torque capability. To confirm the merits of the proposed design, the conventional and symmetrical stator modulation teeth (SSMT) FCSCPM machines are selected and compared by finite element (FE) analysis. The simulated results show that the proposed design exhibits a higher torque capability, better field-weakening ability, and a relatively larger high-efficiency (≥90%) region than conventional and SSMT cases. It validates the feasibility of the proposed design with performance improvement.

Open-switch fault is one of the most common faults of dc–dc converter. If the fault cannot be diagnosed in time, it will have a serious impact on the normal operation of the system. To solve this problem, this article proposes an open-switch fault diagnosis method for bidirectional dc–dc converter based on global state observer. In this method, the global mathematical model of two bidirectional dc–dc converters in hybrid energy sources system is first derived, and then a global state observer is constructed based on Luenberger observer. The open-switch fault detection and location are achieved through the residuals of battery current and supercapacitor current. Both the simulation and experimental results verify the effectiveness of the proposed open-switch fault diagnosis method.

Open-circuit fault is one of the most common faults in permanent-magnet synchronous machine (PMSM) drives. The open-circuit fault can cause the obvious change of stator currents of the PMSM. Hence, the previous artificial-intelligence-based-fault diagnosis method mainly relies on the samples extracted from stator currents. However, the large sets of the samples are required due to the variation of the PMSM operating point, increasing the complexity of fault diagnosis. What is more, stator currents are easily affected by the noise, decreasing the accuracy of fault diagnosis. To solve the issues, this article proposes a robust open-circuit fault diagnosis method using the wavelet convolutional neural network with small samples of the normalized current vector trajectory graph. The proposed method uses current normalization to establish small sample sets and combines the convolutional neural network with discrete wavelet transform to enhance the robustness to noise. The proposed fault diagnosis method is validated by simulation and experiment. Both the results show that the proposed method can effectively diagnose 22 kinds of open-circuit fault types (including healthy mode), being with great antinoise ability and robustness to different working conditions.

This article proposes a simplified model predictive voltage control (SMPVC) method without weighting factors for three-phase four-switch inverter-fed permanent magnet synchronous motor drives to suppress the capacitor voltage offset and reduce the current ripple. In this article, the control of electromagnetic torque and magnetic flux in the model predictive torque control is transformed into the control of the voltage vector, and the capacitor voltage offset is suppressed by injecting dc component obtained by an adaptive notch filter into the phase current. On the basis, the voltage vector is taken as the only control term in cost function without the weighting factors to suppress capacitor voltage offset, thus simplifying the calculation amount in cost function. Then the simplified voltage vector synthesis method is presented to synthesize the effective voltage vector to reduce the current ripple. Using the simple sector division, the switching sequence of the voltage vector can be determined quickly. The effectiveness of the proposed SMPVC method is validated by the simulation and experiment.

With the widespread application of the modular multilevel converter (MMC) in the fields of electronic and electrical engineering, the open-circuit failure problem of the power devices in submodules (SMs) has become the main factor restricting its stable operation. Therefore, power device open-circuit fault diagnosis is widely studied as the main means to ensure the ongoing and stable operation of MMCs. The existing fault diagnosis approaches on the basis of voltage of capacitor change are only applicable to fault diagnosis under active power conditions. However, in reactive power scenarios, MMC systems have very important applications, such as power grid steady-state reactive power regulation and transient reactive power support. Therefore, an effective fault diagnosis method based on the capacitance voltage change rate is put forward for the diagnosis of faults under reactive conditions. This method differentiates the capacitor voltage and compares it with the threshold to diagnose faults under reactive power operation. The proposed method is simulated using PSCAD simulation software. Simulation results verify the practicability of the presented approach.

This article investigates the feasibility of interior consequent-pole permanent magnet (ICPM) machines with reduced rare-earth magnet consumption for automotive applications. A new ICPM machine with a double-layer V-shaped permanent magnet (PM) arrangement is proposed and compared with a commercial Nissan Leaf interior PM (IPM) machine. Their basic electromagnetic performances and rotor mechanical strength are evaluated by finite-element (FE) simulation. It shows that the proposed ICPM machine exhibits comparable torque capability, similar high-efficiency region, and significantly improved PM utilization ratio compared with traditional Nissan Leaf IPM case. Besides, to verify the universality of ICPM designs with reduced-rare-earth consumption, two typical ICPM machines with different magnet configurations (single- and double-layer V-shaped) and power levels (80 kW and 2 kW) are, respectively, investigated and compared with their corresponding traditional IPM counterparts. Finally, some tested results on an ICPM machine prototype with V-shaped magnets are carried out to validate the FE analysis.

The energy management of hybrid energy storage system (HESS) and the nonlinear control strategy of the interface circuit are studied in this paper. In order to realize the reasonable distribution of load power between the battery and the supercapacitor, the energy management strategies (EMS) are used to decompose the load power required for electric vehicles under different driving cycle conditions, aiming at choosing an optimal power allocation strategy. Based on the local modeling, the global mathematical model of two bidirectional DC/DC converters under the HESS with active control topology is established, and the nonlinear control strategies are studied for the interface circuit and then the target controller is designed to realize the EMS. The current inner loop controller based on synergetic control theory is to track the battery current and supercapacitor current. The voltage outer loop controller takes power conservation as the criterion and adds the DC bus voltage PI controller to construct the Lyapunov function, where the connection between the DC bus voltage stability and the supercapacitor reference current is established. The supercapacitor replaces the battery to overcome the fluctuation of the DC bus voltage caused by the sudden change of the load power, maintaining the stability of the entire control system, and at the same time the case that the power allocated by the supercapacitor is consistent with the EMS is ensured. Two typical vehicle driving cycle conditions, IM240 and ECE-15, are selected to test the effectiveness of the current controller and the voltage controller. The experimental platform of a small-capacity HESS is built to conduct experimental verification. The simulation and experimental results show that the nonlinear controllers are feasible and have strong robustness and good adaptability under different driving conditions.