Liao Xiaozhong

Prediction of Wind Power Generation based on Chaotic Phase Space Reconstruction Models

The development of wind generation has rapidly progressed over the last decade, but it must be integrated into power grids and electric utility systems. However, it cannot be dispatched like conventional generators because the power generated by the wind changes rapidly because of the continuous fluctuation of wind speed and direction. So it is very important to predict the wind power generation. This paper discusses why the wind power generation can be predicted in short-term, and how to setup the construction of an ANN (artificial neural network) prediction model of wind power based on chaotic time series. The analysis of modeling with low dimensions nonlinear dynamics indicates that time series of wind power generation have chaotic characteristics, and wind power can be predicted in short-term. Phase space reconstruction method can be used for ANN model design. The data from the wind farm located in the Saihanba China are used for this study.

Skid Steering in 4-Wheel-Drive Electric Vehicle

This paper discusses skid steering applied to four wheel drive electric vehicles. In such vehicles, steering is achieved by differentially varying the speeds of the lines of wheels on different sides of the vehicle in order to induce yaw. Skid steer wheeled vehicles require elaborate tire model, so author choose the unite semi-empirical tire model. From this model, longitudinal and lateral tire force can be calculated by slip ratio directly. The vehicle model has 3-DOF, longitudinal, lateral and yaw direction, irrespective of suspension. Induction motor is chosen as the driven motor, and the control method is rotor flux field oriented vector control. To satisfy the requirement of the turn radius, the longitudinal slip must be controlled, so a method of slip limitation feedback is used in the simulation. When the vehicle is turning on a slippery surface, because of the drop at the coefficient of road adhesion, the drive wheels may slip. The traction control system reduces the engine torque and brings the slipping wheels into the desirable skid range. Some simulation results about the steering accuracy and maneuverability are given in the paper.

Neural networks Preisach model and inverse compensation for hysteresis of piezoceramic actuator

The hysteresis nonlinear characteristic of the nanometer positioning system based on piezoceramic actuator decreases the accuracy of the nanometer positioning stage seriously. To compensate the hysteresis nonlinearity and improve the precision of system with hysteresis, this paper studies the modeling of hysteresis and the corresponding inverse compensation. First, a new sorting & taxis model of hysteresis is realized using neural network to describe the hysteresis of the piezoceramic actuator. A BP neural network is introduced to solve the Function F. With this method the error resulting from interpolation is avoided. Secondly, another neural network is promoted do describe the inverse model of hysteresis. The neural network is used in inverse-modeling to replace the reverse checking and interpolation in traditional method, and the hysteresis modeling error is reduced. At last, the inverse Preisach model based on neural networks in used to compensate the hysteresis nonlinearity. Through the experimental results, the effectiveness of the neural networks hysteresis model and inverse model for the piezoceramic actuator is demonstrated. Also the nonlinear characteristic is reduced effectively by the inverse compensation with neural networks.

Short-term wind power prediction with signal decomposition

Wind power is widely used to replace conventional power plant and reduce carbon emission. However, the variability and intermittency of wind makes the wind power output uncertain, which will bring great challenges to the electricity dispatch and the system reliability. So it is very important to predict the wind power generation. Two different signal decomposition methods are introduced into the prediction of wind power generation in this paper. One is wavelet transform (WT), and another is empirical mode decomposition (EMD). Both of them are good at decreasing the non-stationary behavior of the signal. ANN with the capacity of nonlinear mapping is used to model the decomposed time series. The prediction models WT-ANN and EMD-ANN are compared each other and a combined model based on them is tested. The wind power data from the Saihanba wind farm of China is used for this study.

Wind power prediction using time-series analysis base on rough sets

In long-term prediction, dealing with the relevant factors correctly is the key point to improve the wind power prediction accuracy. The key factors that affect the wind power prediction are identified by rough set theory and then the additional inputs of the prediction model are determined. To test the approach, the weather data from Beijing area are used for this study. The prediction results are presented and compared to the chaos neural network model and persistence model. The results show that rough set method will be a useful tool in long-term prediction of wind power.

A rapid design method for high speed aeronautic switched reluctance generator

The aeronautic generator requires many important characteristics, such as reliability, high power density, high capability, low weight, environment adaptability and fault tolerance. Switched reluctance motor (SRM) is simplicity, low cost, high fault tolerance, high power density and high speed capability, whats more, it can realize dual function of both starter and generator easily, which could fully satisfy the request of aeronautic generator. This paper is devoted to illustrate a rapid way to design a low volume and high power density switched reluctance generator. With the assistance of the Ansoft software, a method based on inductance curve is provided to select the parameters of the switched reluctance generator preliminarily, which could help to reduce time of optimization design greatly. Then the ANSYS software and MATLAB software is used to calculate the accurate performance of the generator, the simulation results show that the presented design method is efficient and effective.

Single phase grid-connected PV inverter applying a boost coupled inductor

In this paper, a single-phase grid-connected inverter applying a boost coupled inductor is proposed for photovoltaic (PV) generation system and PV grid connected systems to enhance integration of a Single phase inverter with Photovoltaic panel to form independent embedded photovoltaic modules. DC 20V to 40V is boosted up to DC 400V, successfully inverted to AC 220V. This will efficiently convert solar energy into electrical energy to comply with the requirements of the local load, or connect the converted electrical energy into the grid. Simulations were conducted using PSIM software to testify the effectiveness of the proposed scheme. Also the experiment on boost circuit was conducted to testify the boosting ability of the coupled inductor.

Design and finite element analysis of outer-rotor-type switched reluctance generator in wind power generation

An outer-rotor-type switched reluctance generator (SRG) applied in small wind power generation system is designed in this paper. The formulas using in the design process are given. The finite element method (FEM) is used to optimize the structure, choose the laminated material and analyze the magnetic performance of the SRG. A comparison simulation between the outer-rotor-type SRG and inner-rotor-type SRG is executed and the performances of the two types SRG are demonstrated. In the comparison, the SRGs have the same outer diameter and poles size. The operating performance of the outer-rotor-type SRG in wind power generation is also illustrated at the end of this paper.

Wind power prediction based on multipositon NWP with rough set theory

Wind power prediction is critical to power balance and economic operation of power system when connected to the grid. In order to improve prediction accuracy, NWP information of different positions and height are taken into consideration to predict wind power in wind farms. In this paper, similar day as the prediction day was searched as training sample at first. The key factors of multiposition NWP that affect the wind power prediction are identified by rough set theory. Then the rough set neural network prediction model is built by treating the key factors as the inputs to the model. To test the approach, the NWP data and actual wind power data from a wind farm are used for this study. The prediction results are presented and compared to the single position wind power calculation model, the single position NWP neural network model and persistence model. The results show that rough set method is a useful tool in short term multistep wind power prediction.

A study of grey theory used in prediction of annual wind power generation

With the coming mature of the wind energy technology, wind energy has become one of the most promising renewable energy. In order to conduct post appraisals and operation management to a large wind farm, accurate prediction of the annual wind power generation is necessary. In this paper, grey model GM (1,1) for predicting annual wind power generation is set up. Moreover, in order to improve the prediction accuracy, a effective method of processing the original wind power data series is proposed. The prediction result with the original data series processed is compared to the unprocessed one. We obtain that the normalized average absolute error of the prediction result with the original data series processed is 7.0315%, improved 0.7679% relative to that original data series unprocessed.