基于DSP自适应模糊控制的直流电源研究

Design of DC power based on DSP self-adaptive fuzzy control

针对传统采用模拟控制或与数字控制相结合的直流电源所存在的"不仅所用元器件比较多,而且硬件电路一旦设计完成,控制策略就随之不能改变,因而控制策略不够灵活"的问题,提出了一种基于自适应神经网络模糊控制理论的PID控制移相方法,并将该方法应用到移相式PWM直流电源中,直流电源主要采用DC/DC变换器来实现;建立了基于该控制策略的直流电源实验仿真模型。验证了该控制方法的合理性。研究结果表明:该方法能够弥补传统传动控制方法的不足,且具有超调量小、稳态精度高的优点,能够有效地解决复杂控制问题,可以动态地适应外界环境的变化,并且能够实现开关管的软开关环境,从而为直流电源提供稳定、可靠的工作环境,且适应于直流电源进一步高频化、智能化、模块化和小型化的发展趋势。

The conventional DC power controlled by analog and combination digital needs more hardware, and when the hardware design is once completed, the tactic would not he changed, so the ploy is not flexible enough. In order to solve the problem, the phased-shifted PID control method was put forward based on adaptive neuro-fuzzy theory, and the strategy was applied to the phased-shifted PWM DC power. The power was mostly composed of uncontrollable rectifier converter, booster and DC/DC converter. At last, the simulation models were constructed to verify the plan. The results indicate that it can availably compensate for the lack of conventional control with little overshoot and very accurate steady-state, and effectively settle complicated control. It has excellent dynamic adaptability in various environments and has made the soft-switching for the switch. So it can provide stability and security for the power, which meets the actual needs of progress on the characteristic of high frequency,intelligence, modularity, and downsizing.