开关电源的建模与优化设计研究

Modeling and Optimization Design for Switching Power Supply

针对目前无法建立开关电源精确数学模型的情况下,研究一种有效的开关电源工程化设计方法。给出了一种开关电源拓扑结构,详细介绍了采用实验和仿真结果建立基于Matlab环境下的开关电源神经网络模型的方法,以及利用该非线性神经网络模型结合遗传算法理论对开关电源的电路参数进行优化设计的过程。在此基础上研制了开关电源样机,并对开关电源优化模型和样机进行了仿真与试验研究分析。结果表明,采用这种优化设计方法制造的开关电源的动态性能、静态性能均达到了预期指标。

Aimming at the incapacity of establishing an accurate mathematic model for the switch mode power supply, an efficient method of engineering-oriented design is studied. A topology of switch mode power supply has been described. A method is presented, which is based on the experiment and simulation result to establish a neural network model for the switch mode power supply in Matlab. The process of circuit parameters optimizing is shown, which has combined the nonlinear neural network model with the genetic algorithm. A prototype which is based on the theory mentioned above has been made. Simulation and experiment have been taken to analyze the optimized model and the prototype. The result shows that the dynamic and the static performance of the prototype, which is designed by the new method, achieved the expected success.