The ramp-up is a critical phase in the operations of a Tokamak, during which engineering and physics aspects must be taken into account to ensure stability, minimize flux consumption and avoid disruptions. Predicting ramp-up phases faces challenges such as nonlinearity, uncertainty on boundary and initial conditions and changes in the magnetic equilibrium. Our work uses the High-Fidelity Pulse Simulator (HFPS), a Python workflow based on JINTRAC. The input and output are in machine and code generic IMAS data format. The HFPS predicts the evolution of the current, temperature, main ion density and impurity density up to the separatrix. The self-consistent prediction of the density during the ramp-up represents the main element of novelty in this work. To this end, a closed feedback loop is set to match experimental line averaged density. QuaLiKiz (Citrin et al 2017 Plasma Phys. Control. Fusion 59 124005), TGLF (Staebler et al 2016 Phys. Plasmas 23 062518) and FRANTIC (Tamor 1981 J. Comput. Phys. 40 104119) are used to calculate turbulent fluxes and the source of neutrals respectively. QuaLiKiz and TGLF predict a transition from Trapped Electron Mode early in the discharge to Ion Temperature Gradient dominated turbulence. The results are compared to higher fidelity simulations with GKW, which show qualitative agreement. Good general agreement is reached between integrated modelling and experimental data, quantified by proposed measures of agreement. A large set of sensitivities to modelling choices and initial and boundary conditions is performed on four different discharges, to assess the robustness of the approach.