Parametric Study on the Effects of Wind for Large Scale Wind Turbines

Research will begin to study the effects of wind on large scale wind turbines. A Simulink model will be used to measure a baseline of how effective a PID controller is at controlling a wind turbine. Knowledge of future wind speeds will be added to see how much more effective a controller is if it 1) knows what wind speeds will be present 2) knows what wind speeds will be present with error 3) knows what wind speeds will be present with a time delay.

It is desired to know which parameter will lead to the biggest gain in the advancement of wind turbine control.

Electric Drive Dead Zone & Compensation

The 'dead zone' is a characteristic of electric drives that puts a limit on the minimum amount of torque needed for rotational speed of the driveshaft. It is, on a torque/speed curve, a region where no rotational speed exists for a torque not equal to zero. The dead zone is modeled graphically as shown below.

When an electric drive is modeled exhibiting dead zone characteristics, the torque from the electric drive is delayed in time from an electric drive not exhibiting these same characteristics.

Dead zone non-linearity affects systems which use electric drives and this discontinuity must be compensated for in some way. One such way of compensating for dead zone works best if the system physics are well-known and the system's model does not inherently change with respect to time. A system with nonzero velocity that enters the deadzone region will coast to a stop at some distance. Knowing the desired position of the system and the velocity at which the system entered the dead zone region, the distance that the system coasts can be measured and applied as a correction to the desired input. Over multiple iterations at varying velocities, a compensation table can be created that will compensate for the coasting caused by the dead zone nonlinearity. This table can be constructed by measuring the required states and programming a fuzzy logic controller to correct the input.

A system model is being constructed in Simulink and results will be posted as they are achieved.

Batteries and Ultracapacitors

Research begins into how to optimize the energy storage system of an electric vehicle (EV). It has been discussed that ultracapacitors and batteries need to work in conjunction to create an attractive solution. Just as important as the energy storage elements themselves is how to interconnect and control these two elements. Efficient charging/discharging of each element requires a system which knows the state of charge (SOC) of each element, current power requirements, and a driving pattern history. This controller will be able to properly handle power flow and extend battery life, thus leading to lower operating costs for an EV. This hybrid controller will be known as a Hybrid Energy Storage System (HESS).

Energy Storage / Power Systems Control for PHEVs

Plug-in Hybrid Electric Vehicles (PHEV)s are a form of electric vehicle which include an electric drive as well as gasoline-powered generator to recharge the energy storage elements which supply power to the electric drive. The two major elements that comprise Energy Storage systems are ultacapacitors and batteries. Batteries have a higher energy density than ultacapacitors but are unable to supply large, quick bursts of energy and also unable to properly store the power generated during regenerative braking. Ultracapacitors are able to charge/discharge very quickly, making them ideal for power delivery when the driver requires large forward-motion acceleration and also for storing the electrical energy generated during regenerative braking.

In an effort to maximize the efficiency of the Energy Storage system, an optimal control strategy should be implemented to control the charging/discharging of the system depending upon the vehicles current state and driver inputs. This optimal control strategy would direct the flow of energy from the vehicle's regenerative braking system to the ultracapacitor bank and should discharge the battery and ultracapacitors depending upon current driving habits.

This control system could be implemented with Fuzzy Logic controllers, Neural Network controllers, or a dynamic gain controller.

TO RESEARCH: Regenerative Braking, Fuzzy Logic, Neural Network / Adaptive Control.