【作者】 王辉；

【导师】 王耀南；

【作者基本信息】 湖南大学 ， 控制理论与控制工程， 2005， 博士

【摘要】 统一潮流控制器(UPFC)是灵活交流输电系统(FACTS)中最具代表性和最多样化的装置。它能实现对输电系统的电压、阻抗、功角、有功功率和无功功率参数的快速动态调节,扩大输电系统的输送能力,提高电力系统的稳定性,优化电力系统运行,是当前FACTS的研究热点之一。论文运用综合智能控制理论,研究了UPFC的控制方法。论文首先研究了UPFC的控制方案。UPFC由并联变换器部分和串联变换器部分组成,可实现对电力系统的并联补偿控制和串联补偿控制。论文以并联补偿和串联补偿的等效电路为基础,研究了串/并联补偿控制的基本关系和控制方案,提出了UPFC的分层递阶控制方案。论文以电流控制电压型PWM(CC-PWM)变换器为基础,研究了UPFC功率变换器的智能电流控制方法。针对常规滞环电流控制器和正弦环宽滞环电流控制器的不足,提出了一种基于模糊逻辑的变环宽的滞环电流控制方法。该方法采用模糊推理方法,根据电流控制误差的大小和变化的方向,自动调整滞环环宽,提高了电流的控制精度和动态性能。PI调节控制是功率变换器线性电流调节器的常用控制方法,但自适应能力较差。因此,论文提出了基于模糊自适应PI的电流控制器和基于模糊神经网络自适应PI的电流控制器两种方法,根据变换器的运行状况自动调整PI调节器的参数,提高了电流控制器的适应能力。UPFC直流电压控制是实现UPFC串联补偿控制功能的重要条件和前提,论文研究了UPFC的直流电压的智能控制方法。论文分析研究了UPFC的电压、电流双闭环控制结构,利用状态反馈和输入前馈方法对电流内环实现了解耦控制,在此基础上,建立了UPFC双闭环控制系统的工程设计方法。将模糊控制和神经网络控制与PID控制相结合,提出了模糊自适应PI控制和单神经元自适应PID控制的两种UPFC直流电压控制方法,仿真研究验证了方法的有效性。潮流控制是UPFC的串联补偿控制的基本功能之一,论文深入研究了UPFC的智能潮流控制方法。论文以串联补偿的等效电路为基础,建立了UPFC的潮流控制的闭环动态结构,利用多变量解耦理论实现了有功功率和无功功率的解耦控制。电力传输线的感抗特性决定了常规PI控制难以获得好的动态特性,本文提出了一种模糊自适应PI控制的UPFC潮流控制方法来解决这一问题。该方法以潮流控制的过渡过程时间、误差和前后两次误差之和作为输入,构成一个三输入两输出的模糊控制器以自动调整PI调节器的参数。仿真结果表明,在同样不产生超调的情况下,模糊自适应PI控制具有更快的响应速度。论文研究了神经网络在UPFC更多还原

【Abstract】 As the most representative and multiplex device of Flexible A.C. Transmission System (FACTS), the Unified Power Flow Controller (UPFC) is one of the research focuses in FACTS recently. UPFC has the characteristic functions as follows: fast and dynamical adjusting the parameters of electricity transmission system, such as voltage, impedance, phase angle, real power and reactive power; expanding the capacity of electricity transmission; improving the stability of power system and optimizing the operation of power system. Based on the comprehensive intelligent control theories, UPFC intelligent control methods were presented in this thesis. Firstly, control scheme of UPFC were described. UPFC consisting of shunt converter and series converter provides shunt compensation and serial compensation for power system. Based on the equivalent circuits of the shunt compensation and the series compensation, the paper investigated the basic relationship between shunt compensation control and serial compensation control, and corresponding control schemes. Consequently, Hierarchic Control Architecture was proposed for UPFC. Based on the CC-PWM converter, intelligent current control method for UPFC power converter was described in the thesis. To meet up the deficiency of the general hysteresis current controller and sinusoid band hysteresis current controller, a band-variable fuzzy hysteresis current control method was presented which regulates the band of hysteresis automatically using fuzzy reasoning on the basis of the value and the mutative direction of control current error. As a result, control precision and dynamic performance of the current are improved. PI regulation law is one of the common methods applied in linear current controller of the power converter while its adaptive performance is poor. To improve adaptive capacity of the current controller, fuzzy adaptive PI current controller and fuzzy neural networks adaptive PI current controller were proposed to adjust the parameters of PI controller automatically according to the running status of the converter. UPFC DC voltage control is the precondition to realize serial compensation control. Intelligent method for DC voltage control was investigated deeply in the paper. Firstly, the double closed-loop control structure of UPFC current and voltage was analyzed comprehensively, and then decoupled-control for inner current loop was achieved by state-feedback and input forward-feed. Consequently, practical design method for UPFC double closed-loop control was derived. In the thesis, combining the fuzzy control and neural control with PID control, adaptive fuzzy PI control and adaptive neural cell PID control were proposed for UPFC DC voltage control. Simulation results have demonstrated the significant performance of proposed method. Power flow control is one of the basis functions of UPFC series compensation control, intelligent power flow control method was studied in the paper. Based on the equivalent circuit of series compensation, this thesis established a closed-loop dynamic structure of UPFC power flow control, which realized the decoupled-control of the active power and reactive power using multi-variable decoupling theory. As the dynamic performance of normal PI control is poor because of the character of transmission lines, this thesis presented an adaptive fuzzy PI control method for UPFC power flow control which takes the transition time, the error and the sum of the last two errors as inputs. Consequently a three-input and two-output fuzzy controller was build to adjust the PI parameters automatically. Simulation results showed that the fuzzy adaptive PI controller has better response performance on the condition of no overshoot. In addition, application of neural network in the UPFC power flow control was also investigated in this paper. Integrating the fuzzy control with neural network control, fuzzy neural networks control and recurrent fuzzy neural networks control method for UPFC power flow control was presented. Simulation results showed that the proposed neural network control method can overcome the disturbance of induction feature of transmission lines and improve dynamic response speed of UPFC power flow control.更多还原