In commercial electrical equipment, interior permanent magnet (IPM) synchronous motors undergo variations in temperature and parameters under different operating conditions. Conventional maximum torque per ampere (MTPA) control using fixed parameters lacks accurate motor parameter information, which can negatively impact the operating efficiency of electrical equipment. To address the above problem, this paper proposes a simple online MTPA control method without motor parameters. To approximate the interface structure of the conventional MTPA controller, a back propagation neural network (BP-NN) adjuster with a space vector decomposer is designed as the MTPA controller. Moreover, by leveraging the similarity between the loss function of the BP-NN and the stator current representing the efficiency of the IPM motor, the teacher signal is cleverly chosen to update the parameters of the BP-NN online. In the discrete domain, the motion mechanism of the proposed method is analyzed and characterized by the perturb & observe technique. Compared to the conventional MTPA control using fixed parameters, the feasibility and high efficiency of the proposed method for copper-loss-minimization control were validated in the simulation. The feasibility of the proposed method was further validated in simulated scenarios involving sudden changes in torque and speed. Furthermore, the transient characteristics of the output stator current were analyzed. Overall, the proposed method can achieve true online MTPA control without motor parameters.

Although significant efforts have been made to enhance industrial air conditioning systems, there are still efficiency and transient response issues in vehicle air conditioning systems using IPMSM compressors. This paper focuses on the neural network compensator-based control in an interior permanent-magnet synchronous motor (IPMSM) to address the occurrence of reduced power copper loss efficiency and degraded velocity response in the motor system when confronted with periodic dynamic disturbances of step signals. This paper encompasses two main objectives: the first objective is to introduce a neural network (NN) compensator to improve the power copper loss efficiency. The NN compensator is developed using the velocity loop and current loop control model equation of an IPMSM, and trained to implement optimal compensation control based on the back propagation algorithm. The second objective is to optimize the dynamic performance of velocity response compared to the traditional maximum torque per ampere (MTPA) current control method under step disturbance and dynamic control conditions by building an experimental system for validation, incorporating both hardware and simulation. Another significant advantage is the low computational load introduced by the neural network compensator, rendering it well-suited for implementation within low-order DSP systems. The results indicate that the neural network compensator surpasses conventional MTPA control method in both simulation and hardware-based implementations concerning power copper loss and velocity response in an IPMSM control system.

In addition to the basic voltage vector modulation technique, virtual vectors can be generated through the discrete space vector modulation (DSVM) technique. Consequently, DSVM-based model predictive control (MPC) can achieve the reduction in current harmonics and torque ripples in permanent magnet synchronous machine (PMSM) drives. However, as the number of virtual candidate voltage vectors becomes excessively large, the computational burden increases significantly. This paper proposes an artificial neural network (ANN) control algorithm, in which massive input and output datasets generated by an existing DSVM-MPC algorithm are utilized for ANN offline training. In this way, the ANN can efficiently select the optimal voltage vector without enumerating all candidate voltage vectors, thereby reducing the heavy online computation of the DSVM-MPC controller and significantly reducing the computational burden. Finally, the effectiveness of the proposed ANN controller is validated.

Heaters are critical components in various heating control systems, and their faults are often a primary cause of system failure, drawing significant attention from engineers and researchers. Early and accurate fault diagnosis is crucial to prevent cascading failures. Many diagnostic methods target faults under generally stable and simple operating conditions, such as constant load or steady-state temperature. However, real-world scenarios are often complex and variable, involving dynamic loads, nonlinear temperature rises, and other challenges, which limit diagnostic accuracy. To address this issue, this paper proposes an intelligent fault diagnosis model based on a one-dimensional convolutional neural network (CNN), using the heater’s current and voltage as the input to the neural network. The effectiveness and accuracy of the proposed model were validated through experimental data under two different conditions, achieving an average accuracy rate of 98%. The disconnection faults were generated during actual operation and occurred in the early stages, differing significantly from artificially simulated faults, thereby increasing the difficulty of accurate diagnosis. Analysis and comparison of the experimental results demonstrate the feasibility of the intelligent diagnostic model and its high diagnostic accuracy.

In commercial electrical equipment, the popular sensorless drive scheme for the interior permanent magnet synchronous motor, based on the quasi-sliding mode observer (QSMO) and phase-locked loop (PLL), still faces challenges such as position errors and limited applicability across a wide speed range. To address these problems, this paper analyzes the frequency domain model of the QSMO. A QSMO-based parameter adaptation method is proposed to adjust the boundary layer and widen the speed operating range, considering the QSMO bandwidth. A QSMO-based phase lag compensation method is proposed to mitigate steady-state position errors, considering the QSMO phase lag. Then, the PLL model is analyzed to select the estimated speed difference for transient position error compensation. Specifically, a transient position error compensator based on a feedback time delay neural network (FB-TDNN) is proposed. Based on the back propagation learning algorithm, the specific structure and optimal parameters of the FB-TDNN are determined during the offline training process. The proposed parameter adaptation method and two position error compensation methods were validated through simulations in simulated wide-speed operation scenarios, including sudden speed changes. Overall, the proposed scheme fully mitigates steady-state position errors, substantially mitigates transient position errors, and exhibits good stability across a wide speed range.