This document constitutes a verification and validation report for the MegaRoller WEC numerical model developed under WP1, describing the approach and the results of the comparison of the numerical model with already validated tools and experimental measurements. The numerical models were developed in a custom version of the numerical simulation package WEC-Sim (Wave Energy Converter SIMulator) and was documented in [D1.1, 2018] and [D1.2, 2019]. In [D1.1, 2018], a preliminary validation of the model was conducted using data from small-scale models which were tested in a wave tank at Queen’s University in Belfast, UK. The present document builds on this preliminary validation exercise using real sea measurement data from the WaveRoller demonstration unit deployed at the Peniche site (Portugal). The following high-level conclusions can be drawn from the loads model validation exercise and sensitivity studies presented in this report: The PTO torque is relatively insensitive to the viscous drag formulation, PTO force cap and consideration of non-linear effects. This is likely due to the low wave heights simulated and therefore lower forces and velocities. The inclusion of wave spreading generally reduces the simulated PTO torque. In some sea states, the WEC-Sim results change from over-predicting the measured data to under-predicting it, although the overall effect on the overall percentage difference is relatively small. The inclusion of a database of hydrodynamic coefficients generally increases the predicted PTO torque. This is consistent with the findings of the initial MegaRoller loads modelling [D1.2, 2019]. The inclusion of a database of hydrodynamic coefficients allows the prediction of non-zero mean flap angles, although the predicted mean angles are still significantly lower than the measured data provided, with the simulation generally under-estimating the mean by an average of 67%. The updated model, including spread waves and a database of hydrodynamic coefficients, results in a slightly more accurate representation of the measured data, although the difference in overall percentage difference between WEC-Sim and the measured data is small. More generally, this work suggests that, even for normal operating conditions and relatively benign sea states, WEC-Sim is capable of predicting RMS values of key metrics to within approximately 20%. Differences for extreme events and when considering maximum values are likely to increase, potentially significantly so. For concept design studies, such as MegaRoller, this level of accuracy may be acceptable, especially considering the load factors typically applied which are intended in part to compensate for this uncertainty. However, for detailed design and in-depth optimisation studies, it is likely that further development of these tools would be required (e.g. by calibrating against tank test data or CFD simulations). In a second exercise, the structural model developed under WP1 was verified by comparing outputs from Code_Aster with a separate finite-element tool. A similar FE model of the MegaRoller flap was built in the FE solver SAP2000, and two static case studies were considered to compare the flap deflections output from both models. In a FE analysis, finding the correct mesh density is, among other variables, one of the significant parameters to control the accuracy of the results. Typically, a high-density mesh means more accurate results. However, this must be balanced against the available computational resources. In a preliminary step to the verification exercise, the effects of different mesh sizes on the simulation results were investigated. Maximum deflections of the flap were shown to converge as element size was reduced, with similar convergence patterns featured by both FE solvers. A mesh size of 0.5m x 0.5m was chosen as an initial balance between accuracy and computational requirements The deformation of the flap estimated by the two FE models was then compared for two static load cases, varying the wave angle. Deformation time-series and deformation maxima at three example points across the width of the flap showed close agreement between the two FE models. Overall, these results provide confidence in the FE model used as part of the WP1.