This paper presents a novel current feedforward control strategy for a three-phase pulse-width modulation (PWM) DC voltage-type converter based on phase and amplitude control (PAC). With right-angle triangle relation ...This paper presents a novel current feedforward control strategy for a three-phase pulse-width modulation (PWM) DC voltage-type converter based on phase and amplitude control (PAC). With right-angle triangle relation of phasors and principle of conservation of energy, a phasor adjustment method and the relevant low-frequency mathematical model of the system are analyzed in detail, both in rectification and regeneration modes for the converter, are discussed. For improving the traditional PAC dynamic performance, variable load current is detected indirectly by the change of the DC voltage, which is fed to the control system as an additional control variable to generate modulation index and phase angle. Also, the algorithm is derived and the system principle is introduced. Experimental results from a 3 kw laboratory device are included to demonstrate the effectiveness of the proposed control strategy.展开更多
The Vienna rectifier is a widely adopted solution for high-power rectification due to its efficiency and straightforward design.However,its performance can degrade under unbalanced three phase voltage conditions,leadi...The Vienna rectifier is a widely adopted solution for high-power rectification due to its efficiency and straightforward design.However,its performance can degrade under unbalanced three phase voltage conditions,leading to current zero-crossing distortion and compromised dynamic response.This paper investigates the causes of these distortions,identifying a phase shift between the input current and the grid voltage as a primary factor,and proposes an effective distortion phase identification strategy.Furthermore,the dynamic performance is enhanced through improved current reference calculations and a refined power feedforward strategy.This approach optimizes the system's response to load changes and maintains output voltage stability under unbalanced conditions.Simulation results validate the effectiveness of the proposed methods in reducing current distortion and improving overall performance.展开更多
随着电力系统对灵活性资源需求的增加,核电作为一种稳定、可控的清洁能源未来在电力系统中不可避免要承担调峰的任务。在调度需求下,主蒸汽阀门开度的频繁变化对一回路冷却剂平均温度等参数产生了一定程度的扰动,而反应堆冷却剂平均温...随着电力系统对灵活性资源需求的增加,核电作为一种稳定、可控的清洁能源未来在电力系统中不可避免要承担调峰的任务。在调度需求下,主蒸汽阀门开度的频繁变化对一回路冷却剂平均温度等参数产生了一定程度的扰动,而反应堆冷却剂平均温度控制系统中的功率失配通道难以满足快速控制的要求。为了提高堆机协调中反应堆调节的快速响应能力,降低主蒸汽阀门开度变化对一回路冷却剂平均温度等参数的扰动,依据大型压水堆(Pressurized Water Reactor,PWR)核电机组的模型,在传统的R棒控制系统的基础上引入以汽轮机负荷为输入信号的前馈环节。一方面,采用理论推导方法,在合理简化R棒控制系统的基础上,借助梅逊公式精确计算前馈环节传递函数;另一方面,采用粒子群算法,依托MATLAB与Simulink之间数据交互,直接对前馈环节传递函数中的待定参数进行寻优。最后,在80%工况下引入±5%、±10%和±20%的负荷阶跃,仿真结果表明:6种负荷阶跃工况下,理论推导的前馈环节让一回路冷却剂温度和主蒸汽压力调节时间缩短53.7%⁓89.6%,波动降22.8%⁓100%;粒子群优化算法计算的前馈环节使调节时间优化64.4%⁓95.3%,波动降23.4%⁓100%。由两种方法得出的前馈环节引入到R棒控制系统中均能有效大幅提升一回路平均温度与主蒸汽压力的调节速度,降低其波动性。所提出的理论计算方法与粒子群优化算法的结果相近,且具有计算成本低、操作简便的优势,在工程中更具应用价值。展开更多
随着智能配电网的快速发展,配电自动化终端的数量逐渐增多,终端入网调试的技术难度不断加大。为了提高自动化终端入网调试的效率,研究提出一种基于门循环单元及前馈神经网络的终端入网智能联调模型(Intelligent Joint Debugging Model f...随着智能配电网的快速发展,配电自动化终端的数量逐渐增多,终端入网调试的技术难度不断加大。为了提高自动化终端入网调试的效率,研究提出一种基于门循环单元及前馈神经网络的终端入网智能联调模型(Intelligent Joint Debugging Model for Terminal Network Access Based on Gate Recurrent Unit and Feedforward Neural Network,GRUFNN)。模型利用门循环单元(Gate Recurrent Unit,GRU)对配电网络终端基本数据进行采集和处理,再通过前馈神经网络(Feedforward Neural Network,FNN)结构完成对数据性能的匹配,判断该点位调试是否合格,从而完成入网认证。模拟测试结果显示,研究所提出的模型在数据节点数量达到1000和2500时分别能够保持86.5%和84.2%的运算精度,且随节点数量增长的影响较小。在实际的应用中,模型对15种调试项目的成功率均在0.9以上,其中对于故障告警的成功率达到0.98,反映出模型全面稳定的运行能力。以上结果表明,研究所提出的模型具有优秀的运行性能,能够完成对实际配电数据的精准处理,对配电入网工况检测工作的自动化和智能化改造具有积极意义。展开更多
文摘This paper presents a novel current feedforward control strategy for a three-phase pulse-width modulation (PWM) DC voltage-type converter based on phase and amplitude control (PAC). With right-angle triangle relation of phasors and principle of conservation of energy, a phasor adjustment method and the relevant low-frequency mathematical model of the system are analyzed in detail, both in rectification and regeneration modes for the converter, are discussed. For improving the traditional PAC dynamic performance, variable load current is detected indirectly by the change of the DC voltage, which is fed to the control system as an additional control variable to generate modulation index and phase angle. Also, the algorithm is derived and the system principle is introduced. Experimental results from a 3 kw laboratory device are included to demonstrate the effectiveness of the proposed control strategy.
文摘The Vienna rectifier is a widely adopted solution for high-power rectification due to its efficiency and straightforward design.However,its performance can degrade under unbalanced three phase voltage conditions,leading to current zero-crossing distortion and compromised dynamic response.This paper investigates the causes of these distortions,identifying a phase shift between the input current and the grid voltage as a primary factor,and proposes an effective distortion phase identification strategy.Furthermore,the dynamic performance is enhanced through improved current reference calculations and a refined power feedforward strategy.This approach optimizes the system's response to load changes and maintains output voltage stability under unbalanced conditions.Simulation results validate the effectiveness of the proposed methods in reducing current distortion and improving overall performance.
文摘随着电力系统对灵活性资源需求的增加,核电作为一种稳定、可控的清洁能源未来在电力系统中不可避免要承担调峰的任务。在调度需求下,主蒸汽阀门开度的频繁变化对一回路冷却剂平均温度等参数产生了一定程度的扰动,而反应堆冷却剂平均温度控制系统中的功率失配通道难以满足快速控制的要求。为了提高堆机协调中反应堆调节的快速响应能力,降低主蒸汽阀门开度变化对一回路冷却剂平均温度等参数的扰动,依据大型压水堆(Pressurized Water Reactor,PWR)核电机组的模型,在传统的R棒控制系统的基础上引入以汽轮机负荷为输入信号的前馈环节。一方面,采用理论推导方法,在合理简化R棒控制系统的基础上,借助梅逊公式精确计算前馈环节传递函数;另一方面,采用粒子群算法,依托MATLAB与Simulink之间数据交互,直接对前馈环节传递函数中的待定参数进行寻优。最后,在80%工况下引入±5%、±10%和±20%的负荷阶跃,仿真结果表明:6种负荷阶跃工况下,理论推导的前馈环节让一回路冷却剂温度和主蒸汽压力调节时间缩短53.7%⁓89.6%,波动降22.8%⁓100%;粒子群优化算法计算的前馈环节使调节时间优化64.4%⁓95.3%,波动降23.4%⁓100%。由两种方法得出的前馈环节引入到R棒控制系统中均能有效大幅提升一回路平均温度与主蒸汽压力的调节速度,降低其波动性。所提出的理论计算方法与粒子群优化算法的结果相近,且具有计算成本低、操作简便的优势,在工程中更具应用价值。
文摘随着智能配电网的快速发展,配电自动化终端的数量逐渐增多,终端入网调试的技术难度不断加大。为了提高自动化终端入网调试的效率,研究提出一种基于门循环单元及前馈神经网络的终端入网智能联调模型(Intelligent Joint Debugging Model for Terminal Network Access Based on Gate Recurrent Unit and Feedforward Neural Network,GRUFNN)。模型利用门循环单元(Gate Recurrent Unit,GRU)对配电网络终端基本数据进行采集和处理,再通过前馈神经网络(Feedforward Neural Network,FNN)结构完成对数据性能的匹配,判断该点位调试是否合格,从而完成入网认证。模拟测试结果显示,研究所提出的模型在数据节点数量达到1000和2500时分别能够保持86.5%和84.2%的运算精度,且随节点数量增长的影响较小。在实际的应用中,模型对15种调试项目的成功率均在0.9以上,其中对于故障告警的成功率达到0.98,反映出模型全面稳定的运行能力。以上结果表明,研究所提出的模型具有优秀的运行性能,能够完成对实际配电数据的精准处理,对配电入网工况检测工作的自动化和智能化改造具有积极意义。
