Current wind turbines in large wind farms operate using “greedy” control strategies where the turbines aim to maximise their own power output (up to rated power) as the sole goal. As the grid penetration of wind turbines increases, the greedy wind turbine control methodology is no longer suitable as it hampers the network operator greatly in balancing the network.

In future scenarios of greatly increased wind penetration it is likely that wind farms will bid to provide power to a set power output in short time scale markets in order to aid the grid operator. The operator will want to output the maximum power possible in order to sweat their asset to the maximum degree. To provide the maximum amount of energy (not under-producing) whilst not exceeding the limit on power output (not over-producing) requires control of the wind farm's power output to a fairly thin band of power outputs, regardless of the variation in the wind. For this purpose turbines may sometimes be required to over-produce for limited periods, extracting energy temporarily from the rotor rotation. It is also desirable to minimise the structural loads on the turbines within the farm and so the distribution of the changes in power is an important consideration. 

This research uses wind farm modelling software developed at the University of Strathclyde to investigate maximum power output and variability in power with and without wind farm curtailment. Different control strategies are evaluated to find the optimum strategy for 'tight' curtailment. The potential of energy storage to improve this further is also investigated.