微电网并网控制技术及优化研究

发布时间：2018-03-18 01:07

本文选题：微电网　切入点：LCL滤波器　出处：《燕山大学》2014年硕士论文　论文类型：学位论文

【摘要】：针对近年来发生的公用电网大规模停电事故、环境的严重污染以及能源的日益枯竭等现实状况，以太阳能发电、风力发电等可再生清洁微电源和储能装置等构成的微电网系统，不仅能够与公用电网并网运行，还可以与公用电网分离进入孤网运行模式，基于其稳定、灵活及易于管理的特点，越来越受到重视和广泛应用。并网逆变器在整个微电网系统中是核心组成部分，其性能对系统高效、稳定运行起着不可替代的作用。在微电网系统并网运行的过程中，为了能够合理分配各分布式电源模块之间的电应力和热应力，应保持各分布式电源之间的电压、频率和相角同步。同时并网逆变器器件本身呈非线性，会引起系统的电压和电流波动，出现波形畸变，从而造成整个电网系统的谐波污染。基于微电网并网系统的上述状况，本文对微电网并网控制技术及其优化进行了研究。针对微电网中单相LCL型并网逆变器系统，设计了一种鲁棒迭代学习控制器。基于逆变器输出的功率下垂特性，实现各逆变器之间的均流运行；利用迭代学习策略适宜于重复跟踪控制和具有周期性扰动抑制的特点，并结合鲁棒的性能指标设计迭代学习控制器，使其具有克服系统参数不确定性的能力。与重复控制相比，鲁棒迭代学习控制能得到更高精度的输出波形，提高了波形跟踪能力，有效地解决了输出信号的失真问题。针对微电网中多并网逆变器参数未知或慢时变控制问题，首先，采用下垂特性实现各逆变器之间的功率分配，同时引入虚拟电抗，解决线路呈阻性时，有功功率和无功功率的强耦合问题，，提高功率下垂控制的性能。将基于功率下垂控制得到的电压基准变换为电流基准，考虑系统参数不精确及变化情况，采用基于自适应内模的可调PID控制器，实现对电流的精确控制。仿真表明控制方案具有良好的稳态和动态性能。针对以风力发电提供电能的微电网系统，由于风力发电的间断性及可能造成的严重频率波动，需对执行用户端进行频率控制。将微电网系统表示成不确定系统，提出了基于H2/H混合目标函数的优化控制方案，采用单纯形法对各PID控制器参数进行寻优。结果表明基于H2/H混合目标函数的单纯形优化方法，不但解决了PID控制器参数难以整定的问题，同时提高了系统的鲁棒性并能实现实时优化控制，保证了微电网系统的稳定运行和功率的充分利用。
[Abstract]:In view of the recent large-scale power outages, serious environmental pollution and the increasing depletion of energy, the micro-grid system is composed of renewable clean micro-power sources and energy storage devices, such as solar power generation, wind power generation and so on. It can not only be connected with the public power grid, but also can be separated from the public network into the isolated network operation mode, based on its stable, flexible and easy to manage characteristics, Grid-connected inverter is a core part of the whole micro-grid system, and its performance plays an irreplaceable role in the system's high efficiency and stable operation. In the process of grid-connected operation of micro-grid system, grid-connected inverter plays an irreplaceable role. In order to distribute the electric stress and thermal stress between the distributed power modules reasonably, the voltage, frequency and phase angle between the distributed power sources should be kept in sync. At the same time, the grid-connected inverter device itself is nonlinear. It will cause voltage and current fluctuation of the system and cause waveform distortion, resulting in harmonic pollution of the whole power system. Based on the above situation of the microgrid grid-connected system, the microgrid grid-connected control technology and its optimization are studied in this paper. A robust iterative learning controller is designed for single-phase LCL grid-connected inverter system in microgrid. The iterative learning strategy is suitable for repetitive tracking control and periodic disturbance suppression, and a robust iterative learning controller is designed to overcome the uncertainty of system parameters. Robust iterative learning control can obtain higher precision output waveform, improve waveform tracking ability, and effectively solve the distortion problem of output signal. Aiming at the problem that the parameters of multi-grid inverter are unknown or slow time-varying in microgrid, firstly, the droop characteristic is used to realize the power distribution among the inverters, and at the same time, virtual reactance is introduced to solve the problem of line resistance. The problem of strong coupling of active power and reactive power improves the performance of power droop control. The voltage reference based on power droop control is transformed into current reference, and the system parameters are considered to be inexact and variable. The adjustable PID controller based on adaptive internal model is used to realize the accurate control of the current. The simulation results show that the control scheme has good steady-state and dynamic performance. Because of the intermittent wind power generation and the severe frequency fluctuation caused by wind power generation, it is necessary to control the frequency of the executive user side in order to express the microgrid system as an uncertain system. An optimal control scheme based on H _ 2 / H mixed objective function is proposed. The simplex method is used to optimize the parameters of each PID controller. The results show that the simplex optimization method based on H _ 2 / H mixed objective function is used. It not only solves the problem that the parameters of PID controller are difficult to set, but also improves the robustness of the system and realizes the real-time optimal control, which ensures the stable operation of the microgrid system and the full utilization of the power.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM732

【参考文献】