Delivering lithium ion batteries capable of fast charging without suffering from accelerated degradation is an important milestone for transport electrification. Recently, there has been growing interest in applying data-driven methods for optimising fast charging protocols to avoid accelerated battery degradation. However, such data-driven approaches suffer from a lack of robustness, explainability, and generalisability, which has hindered their wide-spread use in practice. To address this issue, this paper proposes a method to interpret the fast charging protocols of data-driven algorithms as the solutions of a model-based optimal control problem. This hybrid approach combines the power of data-driven methods for predicting battery degradation with the flexibility and optimality guarantees of the model-based approach. The results highlight the potential of the proposed hybrid approach for generating fast charging protocols. In particular, for fast charging to 80% state-of-charge in 10 minutes, the proposed approach was predicted to increase the cycle life from 912 to 1078 cycles when compared against a purely data-driven approach.