基于BP神经网络的SiC外延炉加热电源动态匹配智能算法研究

Intelligent Algorithm for Dynamic Matching of Heating Power Supply of SiC Epitaxial Furnace Based on BP Neural Network

智能控制算法已普遍应用于各个领域,但是在SiC外延炉中却极少用到。将目前使用最为广泛的BP神经网络模型引入到SiC外延炉加热系统中,对SiC外延炉加热电源匹配电感量进行预测,结合基于变步长搜索最大电流值的方法对频率进行自动跟踪,实现了SiC外延炉加热电源的动态匹配功能。对预测过程和电源系统进行了基于MATLAB的仿真研究,结果验证了匹配电感量预测的精确性和动态匹配算法的有效性,符合设计要求。

Intelligent control algorithm has been widely used in various fields, but it is rarely used in SiC epitaxial furnace. The currently used the BP neural network is widely introduced into the SiC epitaxial furnace heating system, the inductance oft he SiC epitaxial furnace heating power, to predict the maximum current value method with a variable step search for frequency automatic tracking based on realizing dynamic SiC epitaxial furnace heating power matching function. The prediction process and the power system are simulated based on MATLAB. The results verified the accuracy of the prediction and the effectiveness of the dynamic matching algorithm, and meet the design requirements.