This paper proposes a modulation strategy for a 9-level Active Neutral Point Clamped and H-bridge (9L-ANPC-H) converter that features harmonic optimization and low total harmonic distortion (THD), while also incorporating a phase capacitor voltage balancing control. The strategy decomposes the original modulation scheme using a variable threshold calculation method, which equivalently adjusts the voltage contribution of different power stages by varying the threshold values. This enables effective voltage balancing of the three-phase floating capacitors. The algorithm introduces the concept of switching sequence planning and integrates modulation characteristics based on phase arrangement. A hybrid modulation process based on independent modulation, carrier phase shifting and sub modulated signal normalization processing is designed to achieve SPWM-based switching state planning, the output current has lower harmonic components and the overall THD is better than that of the conventional modulation, and the feasibility and validity of the proposed modulation strategy are verified by simulation.

Conventional motor drive systems have numbers of disturbances and uncertainties in all frequency band, which seriously affect current control performance. To solve this problem, this paper proposes a composite controller based on the combination of modified equivalent-input-disturbance (MEID) and fractional-order phase-adaptive feedforward repetitive control (FPFRC) to suppress various types of disturbances and harmonics. Firstly, a unit proportional internal mode feedforward term is introduced in conventional repetitive controller to essentially improve resonance gain and convergence speed. Secondly, the resonant frequencies are modified in real time according to motor speed by using a phase-adaptive mechanism based on the fractional-order IIR filter. Thirdly, the state estimation accuracy and noise immunity of conventional EID strategy are improved by designing a proportional-integral augmentation observer based on the integral term of measured current as an augmentation variable. Finally, a modified filter with phase lead compensation is proposed to realize phase delay-free estimation of low-frequency disturbances in MEID. Experiments results verify the effectiveness of the proposed method, with suppressed current harmonics and improved output torque performance upon system disturbance.

Due to the skin effect, parameters of motor windings change over frequency. In order to solve the problem in traditional Kalman-filter based deadzone compensator that the performance is sensitive to parameter mismatch, a novel deadzone compensation scheme with high robustness is proposed. Firstly, a deadzone compensation scheme based on Kalman filter technology is presented, and motor parameter variation over frequency changes is studied. Secondly, the impact of motor parameter mismatch is analyzed, and the verification of the performance deterioration is done via computer simulation. Thirdly, an Adaline neural-network(NN) based robustness enhancement algorithm is proposed to analyze the current error components. The performance deterioration is compensated by using the analyzed results of the robustness enhancement algorithm. Finally, the robustness against parameter sensitivity of the proposed method under various parameter mismatch conditions is fully studied and verified.

Iterative feedforward tuning (IFFT) enables high performance for motion systems that perform varying tasks without the need for system models. In this paper, IFFT is employed for a wafer stage to achieve good trajectory tracking performance and excellent disturbance compensation ability. Recently, the instrumental variable (IV) approach has been introduced into IFFT algorithms (IV-IFFT), enabling unbiased estimates for the parameters of a feedforward controller in the presence of stochastic noise. However, the estimation variances achievable with IV-IFFT are larger than zero. The aim of this paper is to develop an IFFT algorithm that enables unbiased estimates with zero asymptotic variances, which can be achieved by the simultaneous use of the Kalman filtering (KF) approach and the IV approach in IFFT, yielding the KF-IV-IFFT algorithm. The different roles of KF and IV approaches to improve the noise-tolerant capability of IFFT are also revealed. Experimental results obtained on a wafer stage confirm the practical relevance of the proposed KF-IV-IFFT algorithm.

This paper presents a novel thrust ripple estimation method for PMLSM based on second-order sliding-mode disturbance observer. The super-twisting algorithm (STA) with no sliding-mode surface derivative term is superior in reducing chattering while both retaining strong robustness of conventional sliding-mode, thus it is utilized to design the observer. Then, the equivalent q-axes current according to observed thrust ripple are configured as feedforward compensation term to suppress the force ripple and obtain smooth velocity response within the velocity and current double-closed-loop control system. Coincident observation results with the measured one and obvious velocity ripple reduction verifies the validity and effectiveness of the proposed method on an air-bearing PMLSM test stage.

This paper presents a prototype of magnetostrictive material-based vibration energy harvester, which is used to generate electricity during human walking with high power density. According to Villari effect (inverse-magnetostrictive effect), the harvester consists of a coil-wound magnetostrictive rod. In order to improve the harvesting effect, a magnetostrictive/electromagnetic hybrid configuration with an adjustable air gap is designed. It is followed with detailed analysis of mode effects on ΔB induced by single Δσ and ΔH for optimized design. Then, the magnetic field and leakage of the harvester are analyzed, mechanical parameters of the harvester are optimally determined. Lastly, a prototype of harvester is fabricated and tested. The proposed structure has larger piezomagnetic effect than traditional harvesters, especially under the 5-10MPa compress stress (impact of human walking is set about 5-12.5MPa). Also, it has good harvesting effects in 0.01-0.1s impact vibration time, and can be safely used in up to 20MPa transient impact, which is suitable for future researches of power generating floor systems.