Convolutional neural networks (CNNs) have demonstrated a great potential in fault detection for a wide type of multilevel inverters. Despite the remarkable performance of CNNs, their interpretability remains a challenge. This is due to networks, which have complicated black boxes behaviors. Consequently, they present a substantial challenge for widespread adoption of different models in practical applications. Moreover, relying solely on accuracy is insufficient, especially in critical applications where maintaining trust and robustness is vital for protecting a system against potential damage. Therefore, this study implements a visual explanation method called gradient weighted class activation map (Grad-CAM) for fault detection of multilevel inverter. The Grad-CAM method can identify the model’s important features and interpret the detection of fault types. The proposed method was validated by both simulation and experimental results for three-level neutral-point clamped inverters, demonstrating that a reliable CNN achieved high classification accuracy and effectively identified fault types.

This article presents an advanced speed filtering technique for servo drives independent from the system model information with the reduction of the performance tuning difficulty. The contributions are two-fold: (a) the main filtering system outputs the filtered rotor position from the encoder measurement to recover the rotational speed of the rotor through the order reduction by the pole-zero cancellation property, resulting in the filtering error diagonalization, and (b) the subsystem, which forms the second-order disturbance observer, attenuates the disturbance to improve the acceleration filtering accuracy. The prototype 500-W servo drive validates the improvement in filtering performance under several load conditions.

This research work designs an improved iterative learning control (ILC) with dc-link balancing capability for three-level neutral-point-clamped inverter-fed permanent magnet synchronous motors (PMSMs). Unlike the classical ILC, which depends only on information captured from the last iteration, the proposed ILC approach is capable of storing the data of the state errors from a number of preceding iterations. Henceforth, the designed control method incorporates the simple dc-link balancing algorithm. It encompasses the integration of essential dynamics, with a dual focus on achieving not only rapid dynamic response but also maintaining a satisfactory steady-state performance. The efficacy of the proposed control approach is validated by experimental findings implemented with a PSIM software package and an experimental PMSM test rig with a TMS320F28335-DSP, respectively. Therefore, the improved transient response and steady-state performance indicate that the suggested control approach outperforms the standard approach.

This article proposes a generalized hybrid method (GHM) for permanent magnet synchronous motors that leverages the advantages of field-oriented control (FOC) in steady-state operation and model predictive control (MPC) during transient-state operation. The proposed GHM aims to achieve fast dynamic response, zero average steady-state error, low overshoot and computation time, and reduced voltage vectors (VVs) without the need for weighting factors tuning for multilevel inverters. To achieve these objectives, the GHM creates a single VV using FOC and merges it in the MPC. Furthermore, the fuzzy logic chooses the optimal VV and inserts its input–output to train the artificial neural network to directly select the optimal value and achieve fast selection behavior. The GHM was experimentally evaluated against conventional MPC, FOC, and recent conventional hybrid methods (CHMs) in different operating conditions. The results showed that the proposed GHM demonstrated a low computation time, fast dynamic response, and good steady-state performance. Additionally, the recent CHMs were found to suffer from large overshoot, whereas the proposed method exhibited stable performance and low overshoot.

Convolutional neural networks (CNNs) have demonstrated a great potential in fault detection for a wide type of multilevel inverters. Despite the remarkable performance of CNNs, their interpretability remains a challenge. This is due to networks, which have complicated black boxes behaviors. Consequently, they present a substantial challenge for widespread adoption of different models in practical applications. Moreover, relying solely on accuracy is insufficient, especially in critical applications where maintaining trust and robustness is vital for protecting a system against potential damage. Therefore, this study implements a visual explanation method called gradient weighted class activation map (Grad-CAM) for fault detection of multilevel inverter. The Grad-CAM method can identify the model’s important features and interpret the detection of fault types. The proposed method was validated by both simulation and experimental results for three-level neutral-point clamped inverters, demonstrating that a reliable CNN achieved high classification accuracy and effectively identified fault types.

This article presents an advanced speed filtering technique for servo drives independent from the system model information with the reduction of the performance tuning difficulty. The contributions are two-fold: (a) the main filtering system outputs the filtered rotor position from the encoder measurement to recover the rotational speed of the rotor through the order reduction by the pole-zero cancellation property, resulting in the filtering error diagonalization, and (b) the subsystem, which forms the second-order disturbance observer, attenuates the disturbance to improve the acceleration filtering accuracy. The prototype 500-W servo drive validates the improvement in filtering performance under several load conditions.

This research work designs an improved iterative learning control (ILC) with dc-link balancing capability for three-level neutral-point-clamped inverter-fed permanent magnet synchronous motors (PMSMs). Unlike the classical ILC, which depends only on information captured from the last iteration, the proposed ILC approach is capable of storing the data of the state errors from a number of preceding iterations. Henceforth, the designed control method incorporates the simple dc-link balancing algorithm. It encompasses the integration of essential dynamics, with a dual focus on achieving not only rapid dynamic response but also maintaining a satisfactory steady-state performance. The efficacy of the proposed control approach is validated by experimental findings implemented with a PSIM software package and an experimental PMSM test rig with a TMS320F28335-DSP, respectively. Therefore, the improved transient response and steady-state performance indicate that the suggested control approach outperforms the standard approach.

This article proposes a generalized hybrid method (GHM) for permanent magnet synchronous motors that leverages the advantages of field-oriented control (FOC) in steady-state operation and model predictive control (MPC) during transient-state operation. The proposed GHM aims to achieve fast dynamic response, zero average steady-state error, low overshoot and computation time, and reduced voltage vectors (VVs) without the need for weighting factors tuning for multilevel inverters. To achieve these objectives, the GHM creates a single VV using FOC and merges it in the MPC. Furthermore, the fuzzy logic chooses the optimal VV and inserts its input–output to train the artificial neural network to directly select the optimal value and achieve fast selection behavior. The GHM was experimentally evaluated against conventional MPC, FOC, and recent conventional hybrid methods (CHMs) in different operating conditions. The results showed that the proposed GHM demonstrated a low computation time, fast dynamic response, and good steady-state performance. Additionally, the recent CHMs were found to suffer from large overshoot, whereas the proposed method exhibited stable performance and low overshoot.

This paper proposes stable and efficient I/F control for dual parallel surface-mounted permanent magnet synchronous motor (SPMSM) drives fed by a single inverter. The frequency modulation method using power perturbation of active power is used to stabilize the speed oscillation of a motor. The power perturbation is obtained from the active power of the two motors through a high-pass filter. Using a full-order flux observer, the motor position is estimated, and the torque angle is controlled using estimated position, to minimize the current. Simulations are conducted to verify the presented method.