Humidity Predictive Control of Humidifier Based on Multi-model BP Neural Network Algorithm
The humidity control effect of high-flow respiratory humidifier is an important standard to measure the quality of hu-midifier.Since humidity sensor is not used to collect data in real time at the air outlet of humidifier,the humidity of air-oxygen mix-ture at the air outlet of humidifier cannot be monitored in real time,so a complete closed-loop negative feedback system cannot be built.Only an open-loop system can be constructed with control based on estimated data.For humidity predictive control of humidifi-er,a BP neural network algorithm based on multi-model switching is proposed,which is linearized at the target temperature equilib-rium point,and the model is selected according to the change of flow rate.The experimental results show that the humidity predic-tive control system of humidifier has higher control quality and is suitable for different temperature and flow rate.
国家科技期刊平台
NSTL
万方数据
