This paper presents a novel method for real-time identification of four parameters of the permanent magnet synchronous machines (PMSM) namely stator resistance, d-axis inductance, q-axis inductance and the rotor flux linkage. The proposed method is based on the utilization of the deep neural network to solve the problems of the existing model-based parameter estimation methods, which are caused by the non-linearity of the inverter and the inaccuracy of the measured rotor position. Extensive numerical simulations and experimental studies have been conducted to evaluate the robustness and the accuracy of the proposed online parameters identification solution, compared with the conventional methods such as recursive least square, extended Kalman filter and Adaline neural network.

This paper describes a technique for on-line inductance estimation for an interior permanent magnet synchronous machine (IPMSM) based on a modified square wave voltage injection in the d and q axes. The inductances of an IPMSM can be estimated within each pulse width modulation (PWM) cycle without the knowledge of other parameters, such as the flux linkage, the stator resistance or even the rotor speed. What is more, it is also faster and need fewer computing resources compare with other recursive techniques. Simulation results under different scenarios show that the proposed method can accurately track the inductances not only during steady states, but also during transients.

This paper presents a sensorless control method and a very fast on-line estimation method of inductances of an IPMSM. The current derivatives at one active-voltage vector and one zero-voltage vector are measured every PWM cycle to estimate the rotor speed and position with the aim of increasing the estimation accuracy and reducing total harmonic distortion. In addition, machine inductances can be estimated on-line based on these current derivatives and the DC bus voltage of the inverter. The proposed on-line parameter identification method can overcome the drawback of the existing off-line methods, namely the requirement of a robust mechanical clamping system and test signal generators. The proposed on-line method also addresses the limit of the existing on-line methods, such as slow update and the possibility of incorrect convergence of the estimated parameters. Extensive experimental studies were conducted to verify the effectiveness and robustness of the proposed sensorless control and parameter identification of the IPMSM.

This paper proposes a two-stage optimization of voltage vector selection method integrated with reference stator flux vector calculation (RSFVC) for predictive torque control of a three-level neutral point clamped inverter (3L NPC VSI) fed induction motor. The RSFVC technique simplifies the cost function by avoiding the weighting factor tuning between stator flux and electromagnetic torque and the proposed two-stage method reduces the number of voltage vectors in the prediction stage. Hence, with the proposed control algorithm, two major contributions are achieved. Firstly, the computational burden is reduced due to less number of voltage vector candidates in the prediction stage. Secondly, the design for the cost function becomes simplified as the weighting factor of stator flux error, which is essential in the cost function, is eliminated. Moreover, experimental results confirms that, selection of the two remaining weighting factors required for neutral point voltage (λ cv ) and number of switching transitions (λ n ) becomes less complex. The dynamic and steady-state performances of the proposed control method is investigated in terms of electromagnetic torque and stator flux, total harmonic distortion of stator currents, neutral point voltage variations and robustness of the drive.

This paper investigates the cogging torque and torque ripple in high pole number interior permanent magnet generators, designed for direct-drive applications. Two interior permanent magnet rotor topologies — flat-shaped and V-shaped were considered with distributed wound and fractional slot concentrated wound stators. A comparison of torque performances was made between distributed wound and fractional-slot concentrated wound generators. Cogging torque was minimized by finding an optimum magnet pole arc length and torque ripples were minimized by finding optimum slot-opening and flux barrier shape. Design analysis was carried out in finite element models. It was found that flat-shaped rotor topology in the fractional slot concentrated wound stator can provide the best torque performance regarding low cogging torque and torque ripple. This finding was verified in constructed prototype machine.

In this paper study on solid-state medium-frequency transformer isolated substation is modeled in detail. The substation is formed by AC/DC rectifier, six-pulse DC/AC inverter, high-frequency transformer, DC/AC six-diode rectifier and DC/AC inverter. Both AC/DC rectifier at high AC grid-1 side and DC/AC inverter at lower grid-2 side are controlled by proportional resonant controller and are interfaced with the grids through LCL filters. LCL filters interfaced with both grids are optimized and the controller parameters are also optimized using differential evolution algorithm. It is found by adding passive filter between six-pulse inverter and medium-frequency transformer, it is possible to remove most harmonic voltage components and ensure that the voltage applied across medium- frequency transformer can be quite close to sinusoidal waveform. By doing so, the core losses could be effectively contained, thereby leading to higher frequency and efficiency operation.

A novel combined sliding mode adaptive observer for flux and speed estimation of direct thrust controlled surface mounted Linear Permanent Magnet Synchronous Motor (LPMSM) without using position sensors is proposed. The observer comprises a linear state observer combined with a novel nonlinear sliding mode component. The sliding mode component is improved by using two boundary layers which reduce the chattering without compromising the robustness. The novel observer is experimentally validated. The flux, speed and position estimation errors are small resulting in reliable observer performance.

This chapter brings out the characteristics required for electric machines for traction applications. The focus is on machines and their control issues for hybrid electric vehicles (HEVs) and EVs. It is stressed that high-performance dynamic control requirements of vehicles are met by the decoupled control techniques, such as the rotor flux oriented control (RFOC) and direct torque control (DTC), which have evolved in recent years. Apart from the control of torque, stator and rotor flux linkages in the stator or rotor reference frames, the tractive forces on tyres can be controlled more effectively with electric traction, leading to enhanced energy efficiency and stability. Some of power conversion and control issues related to regenerative braking are also described.

This paper proposes a method of over-voltage protection using power Zener diode for ac-ac matrix converter, ac-dc matrix converter, and matrix-Z-source converter. The analysis and design consideration have been explained, and the discussion on both advantage and drawback of this method also has been given. The effectiveness of this method has been experimentally verified.