动态属性权重智能前馈神经网络电力控制方法

发布时间：2018-12-24 08:14

【摘要】：在工业园区光伏并网控制中,电力控制驱动过程中呈现阻尼振荡,影响系统响应速度和稳定性。提出一种改进的基于动态属性权重人工智能前馈神经网络的电力控制方法,提高工业园区光伏并网电力控制性能。控制系统设计中,采用两个PI控制器,并结合使用磁链控制器和转速控制器,在转子磁场坐标系下,利用仿射PARK变换,使得工业光伏并网控制电力系统所有运算在转子磁场坐标系下实现。构建人工智能前馈神经网络,通过功率前馈控制加快并网系统的响应速度,采用动态属性权重预测电流控制技术补偿延时,实现无差拍控制和正弦脉宽调制。系统测试表明,采用该控制方法进行工业园区光伏并网电力控制,能有效抑制控制延时和电感量偏差对并网电流造成的畸变,控制系统鲁棒性和稳定性较高。
[Abstract]:In the PV grid-connected control of the industrial park, the damping oscillation occurs in the driving process of the power control, which affects the response speed and stability of the system. An improved power control method based on artificial intelligence feedforward neural network with dynamic attribute weight is proposed to improve the performance of photovoltaic grid-connected power control in industrial park. In the design of the control system, two PI controllers are used, and the flux chain controller and the speed controller are combined. In the rotor magnetic field coordinate system, the affine PARK transformation is used. All the calculation of industrial PV grid-connected control power system is realized in the rotor magnetic field coordinate system. Artificial intelligence feedforward neural network is constructed to speed up the response speed of grid-connected system through power feedforward control. Dynamic attribute weight predictive current control technique is used to compensate delay and realize non-beat control and sinusoidal pulse width modulation. The system test shows that the proposed control method can effectively suppress the distortion caused by the control delay and the deviation of inductance to the grid-connected current, and the control system is robust and stable.
【作者单位】：龙岩学院物理与机电工程学院;
【分类号】：TM615;TM921.5

【参考文献】