直驱式永磁风电机组智能控制研究

发布时间：2018-03-21 18:59

本文选题：风力发电　切入点：变桨距控制　出处：《北京交通大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着能源的日益消耗和环境污染的加重,清洁型的新能源在当今社会正稳步的取得愈来愈重要的地位。作为一种可持续发展的新能源,风力发电由于风能的环保无污染,以及取之不尽用之不竭的特性得到了广泛关注。另外,风力发电具有较好的社会效益和经济效益,使其得到快速发展。直驱式永磁风电机组由于其具有无齿轮箱和可靠性高等优点,在风力发电市场的主流机型中占据着重要地位。然而,风速的随机性、风电机组的非线性和复杂性,会对风电机组的输出稳定性和快速性产生影响。因此,先进的控制系统所蕴含的技术在风电机组中的应用就显得尤为关键。本文在分析直驱式永磁风电机组控制策略研究现状的基础上,受中央高校基本科研业务费专项资金“变桨距风电机组智能控制器研究(2013JBM078)”资助,主要做了以下几项工作： 1、分析了直驱式永磁风电机组的系统结构,深入研究了机组的建模过程,建立了包括风速、风力机、永磁同步电机、全功率变流器等模型,将各部分的数学模型有机地连接起来,得到了直驱式永磁风电机组的整体非线性数学模型。另外,研究了风电机组的运行特性,为后续章节的以恒功率控制为目标进行桨距角控制和以最大风能捕获为目标进行电机转矩控制的优化控制方法研究提供了基础。 2、通过分析风力机的运行特性以及变桨距控制要求,采用模糊前馈与模糊自适应PID相结合的复合变桨距控制策略对额定风速之上的风电机组恒功率阶段进行变桨距控制。模糊自适应PID控制器能够依据控制规则改善PID参数整定困难的问题。模糊前馈控制器可根据扰动的大小和方向,补偿扰动对被控量的影响,使被控量因扰动而产生的偏差降低至最低。在模糊自适应PID控制器基础上加入前馈控制,可以大大消除随机性外扰对系统的影响,进一步提高控制品质。 3、对最大风能跟踪阶段的转矩控制进行研究,将支持向量机、粒子群优化算法、模型预测控制相结合,提出了基于粒子群优化算法的支持向量机模型预测控制策略。机组在给定的参考转速与参考功率轨迹下,根据当前时刻测得的是实际输出功率和电机转速以及最优控制电压信号来预测机组的功率与转速预测输出值。仿真结果表明,在滚动优化得到的最优电压控制信号作用下,机组的输出功率与转速能够较好地跟踪功率与转速给定参考值,实现风电机组快速、平稳的捕获最大风能。 4、研究了风力机特性模拟的技术原理,设计并搭建了基于无刷直流电机的风力机模拟试验平台,并基于LabVIEW软件编写上位机界面,实现了风力发电控制系统的数据采集、风机运行控制等功能。在试验平台上对小功率风力机进行了样机试验,验证了实验平台的可行性。
[Abstract]:With the increasing consumption of energy and the aggravation of environmental pollution, clean new energy is steadily gaining more and more important position in today's society. As a kind of new energy of sustainable development, wind power generation has no pollution due to the environmental protection of wind energy. In addition, wind power generation has good social and economic benefits, which makes it develop rapidly. The direct-drive permanent magnet wind turbine has the advantages of no gearbox and high reliability, which plays an important role in the mainstream models of wind power market. However, the randomness of wind speed, the nonlinearity and complexity of wind turbine, and so on. Therefore, the application of advanced control system in wind turbine is very important. On the basis of analyzing the present situation of the control strategy of direct-drive permanent magnet wind turbine, this paper is funded by the special fund of basic scientific research business expense of central university, "Research on Intelligent Controller of variable Propeller Wind Turbine" (2013JBM078). The main works are as follows:. 1. The system structure of direct-drive permanent magnet wind turbine is analyzed, the modeling process of the unit is deeply studied, and the models including wind speed, wind turbine, permanent magnet synchronous motor, full power converter and so on are established. The integral nonlinear mathematical model of direct-drive permanent magnet wind turbine is obtained by organically connecting the mathematical models of each part. In addition, the operation characteristics of wind turbine are studied. It provides the foundation for the following chapters to study the optimal control methods of pitch angle control with constant power control as the target and motor torque control with maximum wind energy capture as the target. 2. By analyzing the operating characteristics of the wind turbine and the control requirements of the pitch of the variable propeller, The variable pitch control strategy based on the combination of fuzzy feedforward and fuzzy adaptive PID is used to control the pitch of wind turbine at constant power stage above the rated wind speed. The fuzzy adaptive PID controller can be improved according to the control rules. The problem of PID parameter tuning is difficult. The fuzzy feedforward controller can be based on the magnitude and direction of the disturbance. By compensating the effect of disturbance on the controlled quantity, the deviation caused by disturbance is minimized. The influence of random disturbance on the system can be greatly eliminated by adding feedforward control on the basis of fuzzy adaptive PID controller. Further improve the quality of control. 3. The torque control of the maximum wind energy tracking stage is studied. Support vector machine, particle swarm optimization algorithm and model predictive control are combined. A predictive control strategy for support vector machine (SVM) model based on particle swarm optimization (PSO) algorithm is proposed. According to the actual output power, motor speed and optimal control voltage signal measured at the present time, the predicted output value of the unit power and speed is predicted. The simulation results show that under the action of the optimal voltage control signal obtained by rolling optimization, The output power and rotational speed of the unit can track the given reference value of the power and speed well, so that the wind turbine can capture the maximum wind energy quickly and smoothly. 4. The technical principle of wind turbine characteristic simulation is studied, the wind turbine simulation test platform based on brushless DC motor is designed and built, and the upper computer interface is written based on LabVIEW software, which realizes the data acquisition of wind power generation control system. The small power wind turbine is tested on the test platform, and the feasibility of the experimental platform is verified.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM614

【参考文献】