New Zealand Wind Energy Site Modelling - Beyond the Wind

A spatial data model using fuzzy logic modelling techniques has been developed to identify potential sites for wind farms in New Zealand. Wind farm development is increasing in NZ with the focus of the energy industry shifting towards renewable energy sources. NZ has an excellent wind resource making it ideal for wind power generation. Using spatial data modelling techniques this study takes the variables that influence wind farm site selection (e.g. windiness, slope, and land use) and combines them into one map that can be used to identify areas that are suitable locations for wind farms. This model could be an excellent resource for any company wanting to find land for new wind farm projects in New Zealand.  For our modelling Kenex have collaborated with Aurecon, a leading global group created by the recent coming together of three world-class companies, namely, Africon, Connell Wagner and Ninham Shand.  Aurecon provide Kenex with technical expertise on wind farm terrain, design, and mesoscale wind speed modelling.  Kenex expertise in spatial data analysis and modelling is used to apply this knowledge to predictive targeting for ideal wind farm sites.

Fuzzy logic is a modelling method used to combine spatial data using subjective judgment. With this method themes are weighted and statistically combined so that every theme has a relevant influence on the model output. This provides a much more integrated result than just adding or subtracting themes from the initial data, i.e. the wind speed grid. The fuzzy logic modelling technique requires the creation of classified grids (predictive maps) for each variable. Each grid class is weighted using a fuzzy membership function (a value between 0 and 1), which expresses the degree of importance of the various map layers as predictors for wind farm locations. The spatial variables considered in this model are windiness, terrain, elevation, proximity to roads, proximity to power lines, population density, distance from built up areas, distance from waterways, and consideration for land use and conservation areas. All spatial data used in this model is held by Kenex and came from a number of different organisations including DOC, GNS Science, NIWA, and Terralink. The predictive maps are combined using a combination of the fuzzy functions (AND, OR, SUM, product and gamma), to produce a final map that ranks potential wind farm locations and takes into account all of the input themes. This modelling technique has not been applied within the wind farm industry in New Zealand before.  Quality wind speed data is a requirement in any wind modelling project and advanced analysis of satellite data by Aurecon that predicts the wind speed over a study area is utilised in our projects. This meso-scale modelling of atmospheric data provides us with continuous detailed coverage of wind speed and wind direction.

Our current model has successfully identified existing and planned wind farm locations in New Zealand and also identified areas known to be unsuitable for wind farm development (e.g., National Parks, waterways and towns). For example in the Wellington region the sites for Meridian’s West Wind and Mill Creek projects have come up as highly suitable whereas urban areas and zones of conservation interest come up as not very suitable (see inset figure). This match to existing knowledge has verified the modelling technique and also highlighted areas where the modelling could be improved with higher quality data. The modelling has targeted specific areas of NZ that would be suitable for wind farm development and has already located several potential targets for follow-up investigation. Our modelling over all of New Zealand to find new targets for development is enhanced further in individual regions of interest.  In these models we use more detailed mesoscale wind speed modelling from Aurecon and advanced terrain modelling to predict areas where there might be turbulence from terrain features.  This regional modelling can be used to map the possible extent of a wind farm, plan how many turbines you might be able to support at the site, and even target areas to investigate for turbine placements.

Please feel free to download our capability statement on wind farm site modelling (click for PDF) or our recent presentation at the annual New Zealand Wind Energy Conference (click for PDF).

Please contact Katie or Matt here at Kenex if you would like more information about this project.