The increasing energy demand, the need to reduce greenhouse gas emissions and the shrinking reserves of fossil fuels have all enhanced the interest in sustainable and renewable energy sources, including wave energy. Many concepts for wave power conversion have been invented. In order to extract a considerable amount of wave power, single Wave Energy Converters (abbreviated as WECs) will have to be arranged in arrays or 'farms' using a particular geometrical layout, comprising large numbers of devices. As a result of the interaction between the WECs within a farm, the overall power absorption is affected. In general, the incident waves are partly reflected, transmitted and absorbed by a single WEC. Also, the wave height behind a large farm of WECs is reduced and this reduction may influence neighbouring farms, other users in the sea or even the coastline (wake effects of a WEC farm). The numerical wave propagation model MILDwave has been recently used to study wake effects and energy absorption of farms of WECs, though without taking into account wave regeneration by wind in the lee of the WEC-farm which can be significant in large distances downwave the WECs. In this paper, the implementation of wave growth due to wind in the hyperbolic mild-slope equations of the wave propagation model, MILDwave is described. Several formulations for the energy input from wind found in literature are considered and implemented. The performance of these formulations in MILDwave is investigated and validated. The modified model MILDwave is then applied for the investigation of the influence of the wind on the wakes in the lee of a farm of wave energy converters.