模块化多电平整流器优化控制策略研究
发布时间：2018-09-08 10:04
【摘要】：随着现代文明对电能需求的增加,电能变换装置的研究变得越来越重要。电压源型变换器经历了两电平至多电平的发展阶段。相对于两电平变换器,多电平变换器具备更好的基波特性、较低的开关频率与器件耐压要求。模块化多电平变换器在多电平变换器家族中具有举足轻重的地位,其模块化的设计方法、易于功率和电压等级扩展等优点吸引了国内外学者的广泛关注。本文以模块化多电平整流器(Modular Multilevel Rectifier,MMR)为研究对象,分别研究了其在电网平衡与不平衡情况下的工作原理,并提出了改进系统控制策略的方法。系统启动时,如果子模块电容电压值低于安全工作时电压值,将会引起较大的冲击电流,造成功率开关器件与电容器的损坏。本文推导了不控充电方式下子模块电容电压的表达式,指出了不控充电方式存在子模块电容电压稳定值受网侧线电压峰值与桥臂子模块个数限制的缺点,并提出了一种选择充电方法。为避免传统的模型预测控制(Model Predictive Control,MPC)策略权重因子的参数调节困难与遍历寻优复杂等问题,本文提出了一种优化模型预测控制策略(Optimized Model Predictive Control,OMPC)。该控制策略在每个计算周期,无需借助价值函数即可得到最优的子模块导通个数。因此,很大程度上降低了控制器的设计难度与运算量。借助MMR数学模型,推导了桥臂电压与相电流、环流实际值及其参考值的关系。借助最近电平逼近或模块统一脉宽调制策略确定子模块导通个数与占空比,通过排序均压策略维持子模块电容电压稳定。针对理想电网情况,建立了MMR的数学模型,并基于此研究了传统双闭环控制策略控制器参数的设计方法。用优化模型预测控制策略取代传统的电流内环前馈解耦控制器,降低了电流内环控制器参数调节的复杂性,提高了电流内环的动态性能。针对非理想电网情况,建立了MMR的数学模型,介绍了基于信号延时法的正负序分离方法。对抑制直流侧有功功率波动与抑制交流电流不平衡两个控制目标分别提出了基于优化模型预测控制的电网不平衡控制方法。方法的有效性通过软件仿真与物理实验得到了验证。
[Abstract]:With the increasing demand of modern civilization, the research of power conversion devices becomes more and more important. The voltage source converter has experienced the development of two-level maximum level converter. Compared with the two-level converter, the multilevel converter has better fundamental characteristics, lower switching frequency and device voltage requirements. Modularized multilevel converters play an important role in the multilevel converter family. The advantages of modularized design method, easy to extend power and voltage levels have attracted wide attention of scholars at home and abroad. In this paper, the principle of modularized multilevel rectifier (Modular Multilevel Rectifier,MMR) under the condition of power network balance and unbalance is studied, and the method of improving system control strategy is put forward. When the system starts up, if the capacitance voltage of the sub-module is lower than the voltage value of the safe operation, it will cause a large impulse current, resulting in the damage of power switch devices and capacitors. In this paper, the expression of capacitor voltage of submodule in uncontrolled charging mode is derived. It is pointed out that the voltage stability of submodule capacitance is limited by the peak voltage of line voltage on the grid side and the number of submodules on the bridge arm. A selective charging method is proposed. In order to avoid the difficulty of parameter adjustment and ergodic optimization in the traditional model predictive control (Model Predictive Control,MPC) strategy, an optimized model predictive control strategy (Optimized Model Predictive Control,OMPC) is proposed in this paper. The optimal number of submodules can be obtained without the help of the value function in each calculation period of the control strategy. Therefore, to a large extent, the controller design difficulty and operation. Based on the MMR mathematical model, the relationship between the voltage of the bridge arm and the phase current, the actual value of the circulation and its reference value is derived. The number and duty cycle of the submodule are determined by using the nearest level approximation or the uniform pulse width modulation strategy, and the capacitor voltage stability of the sub-module is maintained by the sort voltage equalization strategy. In this paper, the mathematical model of MMR is established for the ideal power network. Based on this, the design method of the parameters of the traditional double-closed-loop control strategy controller is studied. The optimization model predictive control strategy is used to replace the traditional current inner loop feedforward decoupling controller, which reduces the complexity of adjusting the parameters of the current inner loop controller and improves the dynamic performance of the current inner loop. In this paper, the mathematical model of MMR is established for the non-ideal power network, and the method of positive and negative sequence separation based on signal delay method is introduced. To suppress the active power fluctuation of DC side and to suppress the imbalance of AC current, the optimal model predictive control method is proposed for the unbalanced control of power grid. The effectiveness of the method is verified by software simulation and physical experiments.
【学位授予单位】：中国矿业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM461
[Abstract]:With the increasing demand of modern civilization, the research of power conversion devices becomes more and more important. The voltage source converter has experienced the development of two-level maximum level converter. Compared with the two-level converter, the multilevel converter has better fundamental characteristics, lower switching frequency and device voltage requirements. Modularized multilevel converters play an important role in the multilevel converter family. The advantages of modularized design method, easy to extend power and voltage levels have attracted wide attention of scholars at home and abroad. In this paper, the principle of modularized multilevel rectifier (Modular Multilevel Rectifier,MMR) under the condition of power network balance and unbalance is studied, and the method of improving system control strategy is put forward. When the system starts up, if the capacitance voltage of the sub-module is lower than the voltage value of the safe operation, it will cause a large impulse current, resulting in the damage of power switch devices and capacitors. In this paper, the expression of capacitor voltage of submodule in uncontrolled charging mode is derived. It is pointed out that the voltage stability of submodule capacitance is limited by the peak voltage of line voltage on the grid side and the number of submodules on the bridge arm. A selective charging method is proposed. In order to avoid the difficulty of parameter adjustment and ergodic optimization in the traditional model predictive control (Model Predictive Control,MPC) strategy, an optimized model predictive control strategy (Optimized Model Predictive Control,OMPC) is proposed in this paper. The optimal number of submodules can be obtained without the help of the value function in each calculation period of the control strategy. Therefore, to a large extent, the controller design difficulty and operation. Based on the MMR mathematical model, the relationship between the voltage of the bridge arm and the phase current, the actual value of the circulation and its reference value is derived. The number and duty cycle of the submodule are determined by using the nearest level approximation or the uniform pulse width modulation strategy, and the capacitor voltage stability of the sub-module is maintained by the sort voltage equalization strategy. In this paper, the mathematical model of MMR is established for the ideal power network. Based on this, the design method of the parameters of the traditional double-closed-loop control strategy controller is studied. The optimization model predictive control strategy is used to replace the traditional current inner loop feedforward decoupling controller, which reduces the complexity of adjusting the parameters of the current inner loop controller and improves the dynamic performance of the current inner loop. In this paper, the mathematical model of MMR is established for the non-ideal power network, and the method of positive and negative sequence separation based on signal delay method is introduced. To suppress the active power fluctuation of DC side and to suppress the imbalance of AC current, the optimal model predictive control method is proposed for the unbalanced control of power grid. The effectiveness of the method is verified by software simulation and physical experiments.
【学位授予单位】：中国矿业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM461
【参考文献】
相关期刊论文 前10条
1 伍小杰;杨超;公铮;戴鹏;;基于多谐振控制器的MMC简化环流抑制策略[J];电工技术学报;2016年13期
2 杨晓峰;郑琼林;薛尧;林智钦;陈博伟;;模块化多电平换流器的拓扑和工业应用综述[J];电网技术;2016年01期
3 刘普;王跃;丛武龙;雷万钧;;模块化多电平换流器优化模型预测控制策略研究[J];中国电机工程学报;2014年36期
4 马为民;吴方R，
本文编号：2230201
本文链接：https://www.wllwen.com/kejilunwen/dianlidianqilunwen/2230201.html
