Inside a wind turbine simulation

An electrical power system (above), including asynchronous generator and transformer, modeled in SimPowerSystems, a MathWorks tool which permits building power system models using click-anddrag procedures. Analysis of the circuit can include interactions with mechanical, thermal, control, and other disciplines, made possible because all electrical parts of the simulation interact with components in the modeling library of the Simulink modeling program. Because Simulink uses MatLab as its computational engine, designers can also use MatLab toolboxes and Simulink blocksets. SimPowerSystems and a similar program for mechanical systems called SimMechanics share a special physical modeling block and connection line interface.
Select figure to enlarge.

Consider the inside of a typical wind turbine nacelle. It will contain a gearbox, electrical controls, a generator, hub mounting, blade controls, and several other major systems. In the case of today's utility scale turbines, all of this equipment is typically designed by different sets of people.

The process of uniting designs created by different teams often leads to integration issues. Despite this challenge, wind turbine power capacities have risen from about 20 kW in the 1980s to 7.5 MW today by using tools that enable early detection of problems with subsystem integration. These tools include system-level simulation models.

Experts in mechanical, hydraulic, electronic, and controls departments not only use simulation individually but also combine their models to optimize turbines through multidomain system-level testing. More specifically, a technique called Model-Based Design has been shown to halve development time for wind turbines and other complex mechatronic systems.

Model-Based Design addresses problems associated with designing complex controls through a combination of mathematical and visual methods. Model-Based Design proceeds in four, general steps: modeling , refining the design through simulation, automatically generating code for the embedded controller and testing, and continuous testing and validation.

Designers typically first model intricate systems such as wind turbines using generic models for such subsystems as actuators, generators, and gearboxes. Later in the design, they substitute more detailed models for the generic versions. The detailed models typically reflect actual performance of real components that are under evaluation for use in the turbine. A Simulink model with mechanical, electrical, and hydraulic systems is often used as a means to test for integration issues as the more detailed models are introduced into the design.
Select figure to enlarge.

The Model-Based Design approach generally involves having developers define mathematical models using continuous-time and discrete-time building blocks. This has an advantage, particularly in portions of the turbine design that involve writing software, by making it possible to employ hardware-in-the-loop simulation to test dynamic effects quickly and efficiently.

When it comes to the plant-modeling of a wind turbine, the process usually involves creating a block diagram-level description that incorporating differential-algebraic equations governing plant dynamics. This approach is combined with physical modeling where physical networks are assembled from components that represent the physical elements that make up the turbine.

Model-Based Design involves offline simulation and real-time simulation. Engineers typically investigate the time response of the dynamic system model to complex, time-varying inputs. The point is to find gross specifications, requirements, and modeling errors early rather than late in the design effort.

Applying Model-Based Design at this early point in the development process helps engineers build a virtual model of the wind turbine that incorporates its mechanical, electrical, hydraulic, and other physical aspects. Linking the model to the system requirements lets it serve as an executable specification for both subsystems and the system as a whole. This brings an unambiguous understanding of the requirements, which reduces the risk of design errors.

There are several advantages to using Model-Based Design in wind turbine design efforts. As we've noted, simulation identifies multidomain design and integration problems early, when they are easiest to correct. Additionally, it creates a behavioral simulation model through which the design can be tested against requirements concurrently as the design goes through successive iterations. Models can also serve as test benches for implementing components and thereby eliminate the need to create tests by hand. For similar reasons, they reduce interpretation errors. Finally, designers can use simulations to evaluate trade-offs, component interactions, and system-level metrics quickly before committing to expensive hardware prototypes.

Continue on next page