Hybrid distribution transformer (HDT) is a promising device for constructing the future active distribution network. To overcome the limitations of the existing HDT control schemes, a compound control scheme is proposed for a HDT with improved circuit design. In this design, the voltage source converter (VSC) for low-voltage (LV) side voltage regulation is moved to the medium-voltage (MV) side. The compound control scheme is developed based on the dynamic model of this new HDT, which explicitly considers the external disturbances. In this control scheme, two compound controllers are established to regulate the grid current and load voltage, respectively. For each compound controller, feed-forward loops are integrated with the PI regulator, thus suppressing the disturbances. The key feature of this control scheme is that, by using the feed-forward loops, the effect of external step disturbance can be equivalently weaken to an impulse disturbance and, hence, the system robustness can be improved. Simulation and experimental results validate the effectiveness of the proposed control scheme.

DC-link voltage control is a key issue for the reliable operation of hybrid distribution transformers (HDTs). This article uncovers the basic power flow principle of HDT based on the established dynamic power balance equations and the proposed power flow diagrams. Based on the derived differential equations and the transfer function models, the dc-link voltage control system with PI controller is developed. With the pure PI controller, the grid currents distortion and asymmetry cannot be well eliminated. To address this, an improved controller that includes a low-pass filter and three notch filters is designed. A disturbance feedforward loop is integrated into the improved controller, thus formulating a novel controller. This novel controller not only controls the dc-link voltage and grid currents with better dynamic performance but also eliminates distortion and asymmetry. Simulation and experiment results verify the correctness of the analyses and proposed methods.

Hybrid distribution transformer (HDT) is realized by augmenting the traditional low-frequency transformer with a fractionally rated converter. This article improves the HDT topology by assembling the LCL -type converters and presents the grid currents and load voltages quasi-proportional-resonant (PR) control strategies for this improved HDT. The PR control schemes are developed based on the established stationary frame dynamic models of HDT, and the effects of the controller parameters and LCL -filter parameters variation on the HDT's control performance and harmonic suppression ability are analyzed by the Bode diagrams and step responses. The key feature of this article is that, by using the resonant controller and LCL -type filters, the complex coordinate transformation can be avoided and the high-frequency harmonics can be well suppressed in the HDT application. The simulation and experiment results verify the effectiveness of the improved HDT and the proposed control strategies.

This paper presents a novel algorithm for industrial robot trajectory planning based on the NURBS(Non-Uniform Rational B-Spline) curve and Slerp interpolation aiming at the problems that the trajectory of a six-axis industrial robot is not smooth enough in the operation process, the posture planning process is non-uniform, and the six-axis industrial robot starts and stops frequently. Firstly, aiming at the first problem, the trajectory planning algorithm based on the NURBS curve is presented to improve the smoothness of the trajectory curve. Combined with Slerp posture planning based on quaternion description, which realizes the uniform change of posture on the robot’s end-effector. Secondly, aiming at the second problem, the S-velocity planning algorithm is presented in the interpolation interval of the robot, which realizes the operation process of complex curves continuously, and improves the operation quality. Finally, this paper uses Bernoulli’s lemniscate as the incentive trajectory, and the contrast experiment of trajectory planning between two incentive profiles is designed, which are the NURBS curve and the five-order polynomial curve. The result of the experiment indicates that the planning algorithm proposed in this paper could effectively improve the smoothness of trajectory in a Cartesian workspace, decrease the impact and tremulous in a Cartesian workspace, and effectively improve the performance of the robot working process. The results drawn from this paper lay a certain foundation for the future high-precision control of industrial robots.