Electricity generation on offshore wind parks has an increasing economic importance - the European Commission foresees that 12% of the wind energy will be produced on offshore installations by 2020, and this share is likely to increase further in the following years. However, the continuously varying offshore environment and the interaction among turbines at large park sites impose significant challenges to the efficient and economically feasible power production. In this work, we present a distributed control system that optimizes the operation of the wind park and adapts to different weather conditions and park configurations. The proposed optimization scheme uses a model for the interactions among the wind turbines in the wind park and the estimated energy production on turbine level to guide the optimization process. The solution can be extended to joint optimization of operation and maintenance on wind park level by incorporating health and wear models in the performance criteria and constraints. Extended simulations on large-scale wind parks show that the proposed scheme improves energy production in low-wind situation and is able to balance production to reduce wear.