Data-informed deep optimization
Fig 5
The distribution of the simulated actual flow value obtained by DiDo.
The distribution of the simulated actual flow value on sampled data used for training DNN surrogate and the candidates of optimal parameters obtained by DiDo finally. The light green bars correspond to the training samples and the dark green bars correspond to the candidates of optimal parameters. The simulated actual flow values of all candidates obtained by DiDo are greater than the largest simulated actual flow value in the training data, demonstrating the effectiveness of our DiDo approach.