CFD for Urban Wind Analysis | Turbine Placement | Wind Siting

CAEbridge has developed a solution for the rapid evaluation of wind resources at building roof tops. The company offers its fast turn around analysis services for determining whether there is any economic sense in purchasing wind turbines for a specific building. If there is, the analysis is extended to deduce the optimal installation spots for the units on top of the building.

Roof top wind generators under 10kW power rating are emerging rapidly as an alternative to offsite wind farms. The relative advantages of onsite solutions include electric grid independence and ability to sell back unused energy back to the grid.

Extreme care has to be taken in locating the optimal installation spot for turbines. The power generated by a wind turbine is proportional to the cube of the incoming wind speed. Hence even small deviations from the optimal spot can result in significant loss of power, hence dollar savings. Thus, significant wind resource assessment effort goes into the siting of rural wind farms using weather maps, long term data acquisition and empirical models.

One hurdle with urban wind resource evaluations is the significant dependence of the wind on building architecture, its neighborhood and the terrain. Depending on the building arrangements and shapes, parameters like the wind direction, speed and turbulence levels can change significantly and non-intuitively. This complexity inhibits the use of empirical wind models which are quite common in the rural space.

One way to remedy the problem can be collecting wind data on the roof of interest using anemometers. However, this process requires months to years of data acquisition with several meters. This approach is hardly practical given transaction decisions for the purchase of turbines need to be made within days. Slightly faster, but also significantly more costly approach may be wind tunnel testing which requires constructing an accurate physical model of the neighborhood of buildings and collecting wind speed distribution data on the model. Set aside the timing issues, the cost of this approach is simply prohibitive.

The most practical and effective alternative to the above approaches is performing an accurate analysis using computational fluid dynamics (CFD). With a detailed computer model for the neighborhood of buildings, and archives of already collected meteorological wind data, CFD can easily predict the complex wind patterns around buildings and the roof top of interest. CFD simulation costs and delivery times are a fraction of the above alternative methods, while the amount of information obtained from it is multi-fold larger. Exploiting CFD, one can build complex target functions based on flow speeds and turbulence levels and pin point to an absolute optimal spot for the turbine installation.

To test the merit of CFD, we set up a controlled experiment, in which two computer models for a 120m high "antenna tower" were constructed. In both, the objective was set as the identification of the ideal turbine installation points on the tower roof. In the first model, only the antenna tower and an immediate neighbor were simulated. In the second model, a more generous 3 block neighborhood of the tower that includes 21 buildings was tested.

For both simulations the following parameters were set:

  
In the below figures, the top view of the incoming streamlines illustrates how the wind patterns change near the antenna tower:

For most typical wind turbine designs and installation attitudes, horizontal component of wind is what accounts for bulk of energy generated. Hence we visualized the wind utilizing arrows that illustrate the direction and speed in the horizontal plane. The roof surface was colored by the turbulence level which has an adverse effect on the turbine efficiency. The differences between the results for the two simulations clearly illustrate different optimal spots for wind turbine installations. Even more critical to the turbine purchase decision process, there is significant difference in the energy production potential between the two cases. The neighborhood case has about 50% less wind speed, hence the time to recoup costs can be 8 times as long as the simplified case!

Assuming the purchase decision is already made, let's focus on the economical impact of taking the wrong decision on points for the installation of 3 turbines rated at 3kW each with an overall cost of $20,000. If we based our decision purely on the results of the first simulation, which is closer to an intuitive one based on meteorological data, the predicted monthly electricity production would be 4500kWh (based on power curves of a reputable turbine brand in the market). In reality, these wrongfully installed units would produce only 216 kWhr based on the more realistic wind speed data of simulation 2. If we based our installation point decisions on the more appropriate simulation 2 data, then the production level would be 648 kWh. Clearly one loses 432 kWh per month due to installing the turbines at the wrong spot. At 20c/kWh electricity cost, this equates to losses of $1036/year. As another data point on the economic impact, it will take 20 additional years to recoup the cost of the turbines purely due to installing them at the wrong spots. Given the results of the correct CFD analysis, investing on these turbines may not have been a good idea to begin with.

In the above example, several levels of detail in the analysis were omitted. For instance, the wind was assumed to have one dominant direction, although a typical wind flower would indicate 2-4 directions that need to be tested. The impact of the installed units on the wind changes the picture for optimal spot choices on the roof. Clearly a turbine should not be in the wake of another one or if an array of turbines is being installed, several spacing and staggering patterns need to be evaluated. In addition, one can play with the elevation of the turbines from the roof to perform a cost/benefit analysis on the burden of additional hardware versus the increased electricity production rate.

The only known tool that will allow for the quick and low cost testing of the above list of adjustable parameters is CFD. CAEbridge is offering its services in performing similar wind assessments and turbine placement analyses. As the demand increases, CAEbridge will refine its processes further to push out a user friendly, urban wind analysis and turbine placement optimization software product in the market.