Wind Energy Update speaks to Brad Hutchinson,VP Global Marketing and Turbomachinery, and Ahmad Haidari, Global Industry Director for Process, Energy and Power, at engineering simulation software provider, ANSYS, to learn how simulations can address issues such as noise, and the engineering simulation challenges presented by scaled-up components.
Interview by Rikki Stancich
Wind Energy Update: How are engineering simulation models being used to tackle the issue of noise from wind turbines?
Brad Hutchinson: Noise is a big issue for people living nearby wind farms or turbines and people need to have more certainty on the noise levels they can expect from wind farm developments.
You need high-level technology to predict noise, which can be broken up into two key types: mechanical and aerodynamic.
To predict aerodynamic noise you need to understand the noise source, which involves complex computational fluid dynamics. You need to have high performance computing to handle the fine meshes required in order to predict details of the aerodynamics which is the source of noise.
Once you have predicted the source, you need to then analyse how it propagates or decays.
We are in the process of developing better solutions for propagation of aerodynamic and mechanical noise.
Wind Energy Update: How close are you to resolving the noise issue?
Ahmad Haidari: If you wave a stick in the air, the noise created is caused by fluid dynamics. We need to get a better understanding of the propagation of noise.
Then there is mechanical noise, for example the sound created when you rub your hands together. In this respect we are looking at the source of the noise.
While the ability to predict noise is improving, there is still some way to go. Understanding the transmission of noise is a challenge.
At this stage, we could make it easier to predict noise, but the solution as it stands is not necessarily packaged and delivered as it ultimately will be. But some of the core components are good.
With time, people gain experience using our tools, which is fed back to us – we don’t develop the tools in a vacuum. But where noise prediction is concerned, we haven’t yet gone through that cycle.
Wind Energy Update: In what way, or to what extent are turbine designs likely to change as noise inhibition is worked into the design?
Brad Hutchinson: When designing, there are always trade-offs, for example, performance for noise. But if you can’t understand the impact of design on noise, you can’t change it. Obviously, once you understand what impact design has on noise, you would pick the design that performs better on noise, or better on aerodynamic design.
Most likely you’d see changes in the blade and tower design or perhaps in the proximity of the blade to the tower. The general configuration depends on the emphasis of noise, on costs and on other factors.
Wind Energy Update: What efficiency gains have been achieved in recent years due to increasingly detailed engineering simulations?
Brad Hutchinson: It really depends on what you are looking at. If you look at the aerodynamics of the turbine blade, the flow at the root of the blade and the tip is very different. A big challenge in aerodynamics is to get this right. Developing the turbulence models that go into our flow solvers and enable us to accurately predict these difficult flows has been a major area of research for us for the last 15 -20 years.
A lot of work and significant improvements have been achieved in turbine blades – we’ve seen something like a 5% improvement in efficiency.
Operating range is a key factor. Wind speed direction and gust is never constant, so you need to analyse wind speed, direction, and how the aerodynamics react to these variations – the better the blade is designed, the better it can handle variations.
Wind Energy Update: As components increase in size, what impact does this have on designing engineering simulations?
Brad Hutchinson: As components increase in size you still need the same small size of the computational grid on and near the blade to resolve the important boundary layer flow immediately adjacent to the blade, whether you are talking about a 20 metre blade or a 60 metre blade length. Resolution of the near-blade (boundary layer) flow is critical to accuracy.
As blade size increases, you just need a bigger mesh because the computational domain is that much bigger, which means bigger models and high performance computers; to get good CFD analysis you need to model a considerable region to get adequate information.
Ahmad Haidari: Ten years ago, people had a smaller blade. As blades got longer and thinner, and as the engineering demands have increased, there has been a need for a higher fidelity model.
Brad Hutchinson: Composite materials are used in the construction of wind turbine blades due to their high strength and low weight. Careful structural analysis is required to achieve a satisfactory design, without using excess materials.
Finite-element modeling for regular materials is very well developed.
Composite materials are more complex, so we have done a lot of work in terms of incorporating more advanced physical models and developing custom user interfaces to facilitate advanced simulation of wind turbine blades and other composite material components.
Wind Energy Update: ANSYS engineering simulation software covers the full range of engineering challenges in wind energy systems. How often are these applied holistically across a given project (i.e. from component design to array layout)?
Brad Hutchinson: We are suppliers of software – we provide the tools that engineers either working on advanced concepts at the advanced R&D stage or at the detailed design stage can use. Ultimately our tools become part of the design system for a given product.
Ahmad Haidari: Our models can be used from blade design through to offshore wind turbine installation, looking at all aspects of engineering, from wave loading to siting.
Customers will buy the software licence and develop it internally according to their specifications.
To respond to this article, please write to the editor:
Rikki Stancich: email@example.com