BP Neural Network for Temperature Prediction of Alpha Magnetic Spectrometer on Orbit

A BP neural network combined with the Adam optimization algorithm and the Mini-batches Learning algorithm was established for predicting the temperature of BOX-C on Alpha Magnetic Spectrometer (AMS) in this paper. After training, the Mean Squared Error (MSE) of the prediction results under the normal operating condition is 0.14 and this shows that the model can be used to predict the temperature of BOX-C with a satisfying accuracy. The model paves the ground for AMS thermal control on orbit.