Supersonic combustion chamber flow field reconstruction method capable of explaining deep learning

The invention discloses a supersonic combustion chamber flow field reconstruction method capable of explaining deep learning, and belongs to the technical field of supersonic combustion chamber flow field reconstruction, in order to improve data-driven reconstruction precision, the method comprises the following steps: S1, constructing a flow field data set of a combustion chamber configuration under different incoming flow conditions by using a ground wind tunnel test and CFD numerical simulation software; s2, performing data preprocessing operation on the flow field data set to obtain a preprocessed flow field data set; s3, disorganizing the preprocessed flow field data set and segmenting the preprocessed flow field data set into a training set and a test set according to a proportion; s4, training a supersonic combustion chamber flow field reconstruction model by using the training set to obtain a trained model and a training weight; and S5, according to the test set, embedding the training weight and the