逆变电源的双重学习控制技术研究

Research on the Integrated Control Strategy for Inversion Power Supply

为了提高逆变电源的输出电压波形质量,提出了一种基于迭代学习补偿和神经网络学习自整定PID参数的双重学习控制方案,分别用于改善稳态时输出正弦电压波形的质量和提高系统处于振荡或过冲时的动态性能。采用TMS320LF2407AG型DSP芯片在一台3.5kWPWM逆变电源上验证了该双重学习控制方案。实验结果表明,采用提出的控制方案不仅能在稳态时获得谐波畸变率(THD)低的输出电压波形,而且能在负载突变时获得较好的动态响应。

In order to obtain higher quality output voltage waveform of inverter system, an integrated control strategy based on iterative learning compensator technique and neural network learning adaptive PID controller （NN-PID） is proposed to generate high-quality sinusoidal output voltage in steady state and improve transient performance whenever the system exhibits an oscillatory or overshoot behavior respectively.Experimental results on a 3.5 kW PWM inverter whose control system is based on DSP TMS320F2407A prove that the proposed control scheme can achieve output voltage waveform with low total harmonic distortion （THD） during steady-state operation and fast transient response subject to load step change.