In this work, an impact time proportional navigation guidance (ITPNG) law for hypersonic flight vehicles (HFV) is proposed. First, a preset-parameters PNG law is developed by the mapping between initial state, the impact time, and PNG navigation parameters. Subsequently, an accurate approximation of the mapping is achieved through training a deep feedforward neural network (DFNN). Compared to comparable findings, the proposed ITPNG legislation offers enhanced time accuracy and practicality in engineering without requiring time-to-go estimation or relying on simplified assumptions. Ultimately, a series of numerical simulations are conducted to validate the efficacy.