Sun Zhenxing

Bio: Sun Zhenxing is an academic researcher from Nanjing Tech University. The author has contributed to research in topics: Torque sensor & Direct torque control. The author has an hindex of 1, co-authored 1 publications receiving 12 citations.

Efficiency optimization of direct torque controlled induction motor drives for electric vehicles

Abstract: Aim at the demands for high efficiency and fast dynamic response of electric vehicle drives, an efficiency optimization control strategy of the direct torque controlled induction motors is proposed. Firstly, the mathematical model of induction motors is established in stator-field-oriented reference frame considering core losses. Secondly, the function relationship between the power losses and the developed torque, rotor speed and stator flux is analyzed, and the computational formula of the stator flux amplitude which makes the power losses minimization is derived at different operating conditions. Consequently, the efficiency optimization of the driving system is achieved. The experimental results show that the proposed control strategy not only preserves fast torque dynamic response, but also improves operating efficiency of the driving system and lengthens the travel distance of electric vehicles.

The effect of flux optimization on energy efficiency of induction motors in fan and pump applications

Abstract: In this paper we discuss the effect of flux optimization on energy efficiency of AC drives. The focus of the study is flux optimization as a part of VSD drive control of induction motors, specifically, for fan and pump applications. Using a very detailed motor model, we present flux optimization results for various speed and load scenarios, and translate these results into practical power savings. We study parameter sensitivity issues of flux optimization and make recommendations for its online implementation. In the final version of the paper we support our simulation study by experimental results. This study is expected to result in a useful electronic tool for industry.

Efficiency Analysis of Induction Motor Control Strategies Using a System-Level EV Model

Abstract: This paper presents a comparative study on the effect of two commonly used induction motor (IM) control strategies on motor and inverter efficiency of battery electric vehicles over standard drive cycles. An electric vehicle (EV) model is created for the 2015 Chevrolet Spark EV and verified using experimental data. After model verification, the Spark permanent magnet synchronous motor is replaced with a detailed IM and controller model. Simulation results for both Field Oriented Control (FOC) with constant rated flux and Maximum Torque Per Ampere (MTPA) control over a test drive cycle are given to validate the good tracking capability of IM current and speed controllers. The effect of control method and drive cycle on motor and inverter efficiency is illustrated by comparing efficiency calculation results for three standard drive cycles (UDDS, HWFET and US06).

Developing Induction Motor State Observers with Increased Robustness

Abstract: This paper presents the results of recently conducted research on Luenberger observers with non-proportional feedbacks. The observers are applied for the reconstruction of magnetic fluxes of an induction motor. Structures of the observers known from the control theory are presented. These are a proportional observer, a proportional-integral observer, a modified integral observer, and an observer with additional integrators. The practical application of some of these observers requires modifications to their structures. In the paper, the simulation results for all mentioned types of observers are presented. The simulations are performed with a Scilab-Xcos model which is attached to this paper. The problem of gains selection of the observers is discussed. Gains are selected with the described optimization method based on a genetic algorithm. A Scilab file launching the genetic algorithm also is attached to this paper.

Neural network SVPWM-DTC of induction motor for EV load model

Abstract: A three phases induction motor is used as an electric vehicle propulsion system in this paper. The proposed neural network speed controller is design based on the space vector modulation technique on direct torque control. Since the electric drive performance significantly lean against on the design of speed controller, thus the improvement on the speed controller become the core of this research and so it is enhanced by replace the generic speed controller by the adaptive neural network controller. The performance of the control system in addition with the neural network learning scheme are depict in this paper. The complete schemes of both conventional and proposed scheme are simulated using Mathlab. The comparison system performance between the conventional direct torque control and the proposed intelligent neural network speed control have been investigated and show a satisfy result in both steady state and transient response. The simulation results verify that the proposed direct torque controller with the adaptive speed controller for induction motor satisfy and effectiveness enough as the candidate for electric vehicle propulsion.

Induction Motor Drive Based Neural Network Direct Torque Control

Abstract: A neural network based direct torque control of an induction motor was presented in this paper. The paper trained a neural network for speed controller of the machine to use in the feed-back loop of the control system. The description of the control system, training procedure of the neural network is given in this paper. The complete neural network based direct torque control scheme of induction motor drive is simulated using MATLAB. The acquired results compared with the conventional direct torque control reveal the effectiveness of the neural network based direct torque control schemes of induction motor drives. The proposed scheme improved the performance of transient response by reduces the overshoot. The validity of the proposed method is verified by the simulation results.