A general model for flow boiling heat transfer in microfin tubes based on a new neural network architecture

•An interpretable neural network architecture was designed, namely DimNet.•A general flow boiling model for microfin tubes was developed based on DimNet.•The model predicts the data with MAE of 13.8% and captures parametric behaviors.•DimNet-based models can be converted to explicit algebraic functions.•DimNet can be used for dimensionless correlation in various complex problems.