Abstract:Objective To solve the problem that the fault diagnosis of switching power supply in medical equipment is difficult due
to lack of drawings, a fault diagnosis method of switching power supply based on feature fusion was proposed. Methods Combined
with the basic working principle of switching power supply, the common faults of switching power supply of medical equipment
were simulated, and the fault signals of key components were collected. Different dimensional features were extracted by using long
short-term memory network and one-dimensional convolution neural network in deep learning, and were adaptively fused into multidimensional
feature vectors. Then, a fault intelligent diagnosis model based on multi-dimensional feature fusion was established.
At the same time, Adam algorithm was used to adaptively optimize the model to realize the automatic identification of the fault
state of switching power supply. Results The switching power supply of Mindray T8 monitor was taken as an example to verify
the proposed method. The results showed that the proposed method had a good diagnosis effect on four kinds of common faults of
switching power supply, and the model accuracy were improved by about 5% compared with the fault diagnosis method based on
single feature. Conclusion The fault diagnosis method of switching power supply based on feature fusion can effectively identify the
common faults of switching power supply, which has certain innovation.
张诗慧,郎朗,种银保. 基于特征融合的开关电源故障诊断方法研究[J]. 中国医疗设备, 2022, 37(9): 27-32.
ZHANG Shihui, LANG Lang, CHONG Yinbao. Research on Fault Diagnosis Method of Switching Power Supply Based on Feature Fusion. China Medical Devices, 2022, 37(9): 27-32.