无源网络接入MMC-HVDC的模糊自适应PI控制策略研究

发布时间：2018-01-09 01:27

本文关键词：无源网络接入MMC-HVDC的模糊自适应PI控制策略研究　出处：《武汉大学》2017年硕士论文　论文类型：学位论文

【摘要】：在实际生活中,我们通常使用该地的柴油发电机或燃气轮机来为远离陆地的孤岛或者钻井平台等无源网络提供电能。但是此类发电机的输电效率通常不高,导致其构成供电系统的安全无法达到期望的效果。因此对于此类无源网络,通过大电网对无源网络供电会是更安全可靠和经济性高的方式。基于模块化多电平的高压直流输电系统(MMC-HVDC)是一种新型的柔性直流输电方式,电压源换流器采用可关断器件可以实现自换相,这种结构的换流器兼具能够灵活操作和在高密度情况下输电等优点,可以很好的解决以上问题,所以采用电压源换流器十分适合向无源网络供电。本文首先阐述了课题背景、研究意义和国内外研究现状,与传统的两电平换流器和箝位型多电平换流器相比较,模块化多电平换流器(MMC)的技术性能更为理想,以及其功能特性上所有的诸多优势,使得MMC更加适合我国实际工程中的高压柔性直流输电系统。接着论文研究了 MMC-HVDC系统拓扑结构、运行机理、工作特性和控制方式,介绍了适用于MMC的调制方式,重点介绍了 NLM(最近电平逼近),并将NLM选为本文所适用的调制策略。其次建立了在保持交流电压稳定时的MMC-HVDC系统的数学模型,在此模型上研究了 MMC-HVDC系统的多种控制策略并且设计了相应的控制器,从换流站级控制策略和换流阀级控制策略两个方面进行研究,设计了接入无源网络的稳态运行时的控制策略,即接入无源网络侧MMC采用定交流电压控制和定频率控制,系统侧MMC采用定直流电压控制及定无功功率控制。然后详细研究了模糊控制的原理、结构和不同类型,并且对比了一般模糊控制和模糊自适应控制的区别,模糊自适应控制能够弥补一般模糊控制规则粗糙或者不够完善的缺陷。因此论文将模糊控制和PI控制两者结合设计了模糊自适应PI控制器,并将模糊自适应PI控制器应用于接入无源网络的MMC-HVDC逆变侧的外环控制器上。在PSCAD上搭建了基于模糊自适应PI控制的无源网络接入MMC-HVDC的系统模型,对稳态情况和故障情况下分别进行了仿真验证,最后的结果证明了本文提出的控制策略的有效性,使得无源网络接入MMC-HVDC后能够有良好的动态响应和抗干扰性,验证了对于系统整体性能的改进,模糊自适应PI控制器有很实际的可行性。
[Abstract]:In real life, we usually use diesel generators or gas turbines to power passive networks far from land, such as isolated islands or drilling platforms, but the transmission efficiency of such generators is usually low. As a result, the safety of the power supply system can not achieve the desired results. Therefore, for such passive networks. It is more safe, reliable and economical to supply power to passive network through large power grid. MMC-HVDC based on modularized multilevel HVDC is a new flexible DC transmission mode. The voltage source converter can use turn-off device to realize self-commutation. This kind of converter has the advantages of flexible operation and transmission under high density, so it can solve the above problems well. Therefore, voltage source converter is very suitable for power supply to passive network. Firstly, this paper describes the subject background, research significance and domestic and foreign research status. Compared with traditional two-level converters and clamped multilevel converters, the modular multilevel converters (MMCs) have better technical performance and many advantages in their functional characteristics. It makes MMC more suitable for HVDC system in our country. Then, this paper studies the topology, operation mechanism, working characteristics and control mode of MMC-HVDC system. The modulation mode suitable for MMC is introduced, with emphasis on NLM (nearest level approximation). NLM is chosen as the suitable modulation strategy in this paper. Secondly, the mathematical model of MMC-HVDC system is established when the AC voltage is stable. In this model, the various control strategies of MMC-HVDC system are studied and the corresponding controllers are designed. The control strategy of converter station level and the control strategy of converter valve level are studied from two aspects: the control strategy of converter station and the control strategy of converter valve level. A control strategy is designed to access the passive network when it is running stably, that is, the constant AC voltage control and the constant frequency control are used to access the passive network side MMC. The system side MMC adopts constant DC voltage control and constant reactive power control. Then, the principle, structure and different types of fuzzy control are studied in detail, and the differences between general fuzzy control and fuzzy adaptive control are compared. Fuzzy adaptive control can make up for the defects of rough or imperfect fuzzy control rules. Therefore, fuzzy adaptive Pi controller is designed by combining fuzzy control and Pi control in this paper. The fuzzy adaptive Pi controller is applied to the outer loop controller of MMC-HVDC inverter side to access the passive network. The passive network access M based on fuzzy adaptive Pi control is built on PSCAD. The system model of MC-HVDC. Finally, the simulation results show the effectiveness of the proposed control strategy. The passive network can have good dynamic response and anti-jamming after accessing to MMC-HVDC. It is proved that the fuzzy adaptive Pi controller is feasible to improve the overall performance of the system.
【学位授予单位】：武汉大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM721.1;TM46

【参考文献】