Liu Guohai

Dynamic decoupling control of bearingless switched reluctance motors based on neural network inverse system

Bearingless switched reluctance motors should remain stable levitation force and rotation force in different positions. Not only are the two forces nonlinear functions of positions, but also the levitation forces in two degrees of freedom have strongly coupled nonlinear relationship. Furthermore, the nonlinear coupled relationship exits between the levitation force and rotation force too. In order to realize the stable levitation and controlled rotation of bearingless switched reluctance motors, the first step is to dynamically decouple the levitation forces in different positions and to search for control laws in different positions. Based on basic electromagnetism theory, a radial force and position model of a bearingless switched reluctance motor is presented. Aimed at the nonlinear and strongly coupled characteristics, the model is analyzed with reversibility and proved to be reversible. The nonlinear and strongly coupled multi-variables system can be decoupled and transformed into two linear subsystems without position coupling to each other by connecting a neural network inverse system before a bearingless switched reluctance motor. This neural network inverse system consists of a static neural network (MLN or RBF network) and two integrators, where the static neural network represents the nonlinear mapping relation of the inverse system and the integrators represent the dynamic characteristics of the inverse system. Consequently, the high performance control of the original nonlinear and coupled system can be realized under the help of linear closed-loop controllers for each decoupled subsystem. The results of simulation show that this system can realize the stable levitation of bearingless switched reluctance motors.

Application of an adaptive controller with a single neuron in control of multi-motor synchronous system

The paper takes the multi-variable of synchronous system of the AC induction motors as study object, focuses on the system of induction motors powered by current-tract SPWM transducers, establishes the unified model of the system of two motor depends on the alpha, beta static reference frame. It adopts the adaptive PSD controller with multi-variable single neural unit to perform the decoupling control of speed and tension of the system. The results of simulation and experiment prove that the controller has higher speediness, fine control precision and good robust.

A novel navy ship power system restoration using multi-agent approach

A reconfiguration system is needed to solve power balance problems and try to satisfy vital loads as much as possible. And it is important to restore the power system as soon as possible to a target network configuration after the fault, especially after combat damage occurs. Current reconfiguration techniques are centralized methods and cannot meet the Navys needs for fight-through survivability and high reliability [1]. This paper presents a novel multi-agent based reconfiguration approach to perform system restoration for navy ship. Each zone agent in this system communicates with other agents. With the same global information and the same target, all zone agents use the same algorithm and collaborate with each other to achieve the reconfiguration objective as soon as possible.

Lifting wavelet transform based adaptive filter for active power filters

This paper presents a new harmonic measurement scheme on lifting wavelet transform based adaptive filter (LWTAF) for active power filters (APFs). As one of methods to build second-generation wavelets, lifting wavelet transform (LWT) can be easy to achieve in time domain for its original bit location calculation. LWT could lessen calculation complexity and depress course due to its particular structure, compared with first generation wavelet algorithm (WT). Then the performance of the LWTAF is studied and the harmonics detection working schematic of LWTAF is presented. LWTAF can abstract harmonic components effective and follow up the transient component more quickly than adaptive filter (AF). The simulation results validate the conclusion above.

Estimation of induction motor speed based on artificial neural networks inversion system method

Rotation speed is necessary for high-performance induction motor control, how to estimate the speed quickly and accurately is concerned by most scholars. On the analysis of theoretic invertibility of the induction motorpsilas mathematic model, a speed estimation based on artificial neural networks inversion is proposed. The structure of multi-layer feed-forward neural network (MFNN) is trained by advanced Back Propagation arithmetic. Also the the achievement mehod and experiment results were given. The results show that the responses based on ANN inversion method can track the rotation speed quickly and accurately. The method proposed is attractive in application.

Multi-motor synchronous system based on neural network decoupling control

Taking the multi-variable of synchronization system of the AC induction motors as study object, focusing on the system of induction motors powered by current-tract SPWM transducers, the mathematical model of the system of two motors is established. Combining decoupling technology of adaptive neuron decoupling compensator, RBF neural network adaptive PID controller is adopted to design the neural network controller of two-motor synchronous system. The experiment results show that the two-motor synchronous system is decoupled based on the controller designed with better dynamic and static characteristics and following performance.

Static Decoupling Control Based on Hammerstein Model and Neural Network for Induction Motors

A novel static decoupling control strategy based on Hammerstein model and neural network for induction motors was proposed in this paper. Hammerstein model, consisting of a static nonlinear module and a dynamic linear module, can be used to model many nonlinear systems. In the proposed method, firstly, neural network and auto-regressive moving-average (ARMA) model were employed to construct the static nonlinear module and the dynamic linear module respectively. Further, neural network inverse model of the static nonlinear module can be trained on the static dataset collected in the framework of the Hammerstein model. Finally, the inverse model was utilized to offset the nonlinear characteristic of an induction motor, decoupled into a rotor speed subsystem and a rotor flux subsystem. Simulations show that the proposed static decoupling control strategy has satisfactory decoupling performances and robustness to load disturbance in close loop control.

Hybrid Power System Composed of Photovoltaic and Commercial Power

As an important application of solar energy, photovoltaic generation has attracted more and more attention. But there are some disadvantages in the perspective of solar energy, such as dispersion, intermittent, and randomness, which cannot provide stable and consistent power and should be equipped with additional energy. This chapter chooses commercial power as the back-up energy and proposes a hybrid power system with both photovoltaic and commercial power. The system is composed of a solar cell, the commercial power, a DC–DC(direct current) converter, a power factor correction (PFC) converter and DC load. In order to utilize the solar energy as much as possible, a power management is necessary for the hybrid power system. The core of the energy management strategy is to keep the system work under suitable mode to control the energy flow of the system according to the work status of the solar cells and the load. The experimental results verified the effectiveness of the energy management strategy.

Data reconciliation by two-step risk analysis of modeling

A new method for data reconciliation by risk analysis of modeling is presented in this paper. Yamarura designed an integer programming model for gross error detection and data reconciliation based on Akaike information criterion. But much computational cost is needed for its combinational nature. To reduce computation burden, a new method by two-step risk analysis of modeling is proposed. Measurement modeling risk is analyzed in the first step. Then gross error modeling analyzed based on the minimum measurement modeling risk is considered. The proposed method could effectively reduce the scale of the integer programming problem. Simulation shows the efficiency of the proposed method.