Robust control for maximum power point tracking of tidal current turbines

Abstract

In recent years, there has been a growing interest in tidal current energy as a potential source for predictable green electricity generation. Due to the harsh marine environment, the Tidal Current Turbine (TCT) system has to be designed to be robust and to work reliably to minimize the need for intervention. Faults may happen in the TCT system, particularly tidal current speed sensor faults, as this sensor is exposed directly to seawater and significant forces from the tidal current flow. Tidal current velocity sensor-less control strategies are needed to enable the TCT system to continue operating under the situation where the tidal current velocity sensor fails. This thesis presents two sensor-less control strategies: Perturb and Observe (P&O) control and Fuzzy Logic (FL) control. Models of a conventional grid- connected TCT and a stand-alone TCT generation system are presented. The P&O control is developed for both models and its performance analysed. FL control is then applied in the grid-connected TCT generation system model to compare the control performance. The simulation results show that variable step size P&O can track the maximum power point (MPP) with fluctuations of +/- 2% around the desired value for both grid-connected and stand-alone systems. The grid-connected model was tested under both step and ramp changes of tidal current velocity to test the response to both sudden and continuous variation of input. The P&O control has reliability issues caused by the algorithm itself, however the FL control was able to solve this problem and can also track the MPP with less than 2% variation in generator speed around the required value. Both P&O control and FL control are tested with realistic tidal current velocity input for a tidal cycle (low tide-high tide-low tide), and multiple variable step size P&O control was developed and applied to solve tracking issues. The performance comparison of the two control strategies in terms of efficiency, turbine and generator power and generator reference speed ware analysed. The results show that the sensor less control strategies developed in this thesis can allow the TCT generation system to operate without the tidal current velocity sensor, thus improving the system robustness whilst reducing system maintenance and power generation cost.