In order to solve the problems of uneven fermentation of koji bticks, multiple temperature measurement points, complicated temperature structure and unsuitable fine-tuning and control of the temperature system in the traditional koji bank, a dual neural network control algorithm and internal circulation pipeline structure were proposed to achieve real-time temperature measurement and control of the koji bank. The internal air circulation in the room is also introduced. After experimental testing, the system works stably, and can accurately measure and control the ambient temperature of the koji bank during the fermentation process of the koji bricks, so that the temperature in each area of the koji bank can be kept uniform.