Hydraulic Wind Power Transfer System is a promising alternative to conventional wind turbines, and an increased research in this field indicates the ability of this technology to replace conventional wind turbines. This technology not only provides an initial economical advantage by eliminating the need for a gearbox unit, but also provides further long term economical and reliability advantage by transferring the generation unit to the ground level, which provides a low-cost and easier maintenance and inspection. In addition, transferring the generation unit to the ground level and eliminating the gearbox decreases the weight of the wind turbine and thus reducing the size of the foundation. However, the unpredictability of renewable energies make them less reliable and prone to power shortage. To mitigate this issue, an energy storage system can be coupled to the generation system so that it would store the excess energy when the renewable source has more power than the demand, and return the excess energy back to the system when the renewable source is not able to meet the demand. Similar to conventional wind turbines, this technology also requires complex control algorithms to operate. These algorithms not only aim to improve the power quality and match the grid rules, but also aim to absorb the maximum amount of power from wind. To design proper control strategies, a detailed model of the system is needed for the simulation. During my research in this lab, my contributions included: (1) Introduction of a method to eliminate the need for proportional valve for frequency control and thus eliminating the valve pressure loss in the hydraulic wind power system (2) Integration of a energy storage system with the hydraulic system and design of a control system in order to maintain the generator frequency during power shortage (3) Proposal of an strategy and control method for Maximum Power Point Tracking (MPPT) and increasing the efficiency of the wind power absorption. Besides these mentioned points, I have also conducted simulations and have revised methods for wind speed estimation, and use of neural-networks to optimize the generators. However, due to time constraint or unfinished work, I have included them as recommendations for future research.

Indiana University-Purdue University Indianapolis (IUPUI)