Zephyr North Ltd. is Canada’s oldest and most experienced wind resource assesment consultancy, with over 25 years experience in the industry. We are an independent company providing a full start-to-finish wind resource assesment service for the wind energy industry. Among our capabilities are:
  • Acquisition, analysis and evaluation of historical wind resource data
  • Design, installation and maintenance of on-site wind monitoring systems, including data collection and quality assurance
  • Analysis of on-site data to determine the local wind resource characteristics
  • Correlation of historical and on-site wind data to estimate future wind resource
  • Sophisticated, state-of-the-art numerical modelling of regional and site-specific wind resources accounting for anemometer sheltering, topography and vegetation/land-use
  • Wind farm design and optimization accounting for complex wind flow, noise, setback, turbine sheltering, and other constraints
  • Mesoscale wind mapping using Environment Canada’s EnSim-WE modelling system
Acquisition, Analysis and Evaluation of Historical Wind Resource Data
Zephyr North Ltd. has developed a complete suite of proprietary, in-house computer programs to process data into tables, graphs and reports describing the historical wind resource in detail. Additional programs analyze the historical data for the consistency required for meaningful correlation analysis with on-site wind monitoring data. In addition, Zephyr North staff are intimately familiar with the “Station Inspector Report” (SIR) archive (both hardcopy and electronic versions) held at Environment Canada. Acquisition and use of these metadata are essential for a successful correlation analysis.
Design, Installation and Maintenance of Wind Monitoring Systems
Zephyr North Ltd. has extensive experience across Canada and internationally in making wind and other meteorological measurements. Some features of our wind measurement capabilities are:
  • Site selection: including map investigations, field surveys, and numerical modelling
  • Instrumentation : robust, high quality sensors and data loggers with calibrations traceable to MeasNet or NIST standards
  • Towers : from a custom-designed 10 m tower to 60 m tilt-up towers, or using existing towers
  • Zephyr North maintains its own fleet of wind monitoring stations, which are available for rent
  • Data acquisition : on-site, land-line or cellular telephone, radio and satellite data collection; Internet data transfer
  • Data recovery : typically greater than 99% using high quality sensors carefully installed along with rigorous data collection and inspection procedures
  • Data quality assurance : all data collected by Zephyr North Ltd. are subjected to stringent, automated quality-control procedures followed by careful visual examination
Analysis of On-Site Data
Zephyr North’s suite of purpose-written computer programs are also used to process on-site field data into summary statistics describing the local wind resource. Further capabilities include surrogate calculations to estimate long-term turbine generation capabilities (after correlation analysis), generation of statistics required by turbine suppliers for turbine suitability analysis such as the site temperature/wind joint frequency table, turbulence intensity distribution, 50-year return wind estimates, and so forth.
Correlation Analysis of Historical Data
Zephyr North Ltd. has also developed a suite of in-house programs to carry out correlation comparisons of historical (reference) and on-site wind data. These programs carry out three types of correlations – a simplistic wind speed ratio analysis, a standard sector-wise regression MCP (Measure-Correlate-Predict) analysis, and a matrix correlation analysis. The literature suggests that the matrix analysis is generally superior. Results of all three are compared.
Microscale Numerical Modelling
Refinement of the accuracy of on-site and regional historical data requires analysis with proven numerical models. At Zephyr North Ltd., we routinely use:
  • MS-Micro/3 : a suite of programs for predicting the effects of complex terrain on wind flow – used for correction of historical data sets, prediction of the wind regime as well as to optimize turbine locations. We have created a custom version of MS-Micro with many enhancements for exclusive use at Zephyr North Ltd
  • ShelCorr : a program for correction of the effects of upwind obstacles on wind data sets
  • TurbPower : an interactive program for quick estimation of wind turbine energy generation
  • WAsP and PARK : programs for wind resource assessment developed in Denmark and the European Community
  • ReSoft WindFarm: a wind farm design and analysis program developed in the UK and incorporating using MS-Micro code
  • These models are supplement by in-house, purpose-written programs

Wind Resource Assessment
There are four main activities required for the technical assessment of the wind energy resource in a region. The Canadian wind energy community demonstrates expertise in all of these activities and has led the world in the development of efficient, accurate and field-tested computer models.:
  • Assessment of Historical Data. As an initial step in wind resource assessment, the investigator must search out the historical wind data for the region. These data must be carefully treated to mitigate a number of problems that might exist – missing data, equipment malfunctions, icing effects (if possible), observation and recording biases, changes in instrumentation, location or exposure, and effects due to local terrain and obstacle sheltering, among others. The Canadian-developed models, MS-Micro/3 and ShelCorr can be used to adjust for the effects of the latter two.
  • On-Site Monitoring. For a major wind energy project, on-site monitoring of the wind regime is essential to confirm the wind resource estimates that might have been made initially from the historical data and numerical modelling. MS-Micro/3 is an excellent tool for deciding where to place monitoring stations since a single run of the program can provide a regional contour map of estimated turbine output. Selection of prime locations from this map is easy. Also, computer modelling is usually important in the assessment of terrain and sheltering influences on the monitoring data as described above for historical data.
  • Computer Modelling. Computer models of wind flow in complex terrain require gridded input fields of topography and aerodynamic surface roughness (land-use is used as a surrogate). The model can then generate the wind speed-up (or slow-down) factors. In the case of MS-Micro/3, these factors are calculated on a complete grid covering the model domain. With the addition of the wind speed/direction joint frequency distribution (which can be derived either from historical data or on-site monitoring data) and the power curve from a candidate wind turbine, the estimated power output from the turbine can be calculated at every grid point in the model domain. Note that this differs from WAsP (Denmark) which calculates the wind and power values at a single point – re-calculation is required for additional points and can be slow.Another Canadian-developed model, ShelCorr, can be used to determine the effect of obstacles on wind speed measurements. Often, anemometers are placed at airports where, for certain wind directions, they can be sheltered by hangars or terminal buildings. In other instances, anemometers may be downwind of lines of trees or fences. It is important to be able to estimate the magnitude of these effects in order to adjust the measured wind data and, hence, more accurately assess the wind regime at the anemometer location.
  • Correlation. On-site wind monitoring should be carried out for at least one year. However, this period is still insufficient to characterize the 20- to 30-year long-term characteristics of the wind regime required for accurate prediction of long-term energy output by a wind generation project. One way to increase the value of the (short-term) on-site monitoring data is to correlate them with long-term data from a historical measuring station located nearby (at an airport, for example). Until now, there has been no standard method for carrying out these correlations. Recently, Canada has developed, tested and published a methodology and will soon make it available as software for correlating long-term historical and short-term monitoring station data.

As can be gathered from above, the four activities are not necessarily carried out in a linear fashion. There are a number of interdependencies among historical data assessment, on-site monitoring, computer modelling and long/short-term correlation. Understanding these allows the investigator to optimize the extraction of accurate and reliable data from whatever sources he or she has available.

Wind Farm Design
Wind farm design is based on solid wind resource assessment.

Design activities include placement of the wind turbines in the terrain while respecting a number of objectives and constraints.

The primary objective, generally, is to place the turbines optimally so that the maximum amount of energy (Annual Energy Production, AEP) is generated. This objective is served, obviously, by placing as many turbines as possible within the wind farm. Unfortunately, though, when the turbines are placed too densely, “array losses” occur because the turbines mutually interfere with the wind flow. Wind farm design, then, concerns finding the balance that results in the maximum AEP.

There are also a number of constraints that can be brought to bear in the design of a wind farm that will modify the turbine layout. These include:

  • Noise: generally, the sound generated by the turbines (including those, possibly, from adjacent wind farms) cannot rise above a level set by local regulations
  • Road, building, community, lake, sensitive area setbacks: most jurisdictions require that turbines be a mandated distance from roads, buildings, communities, lakes, environmentally sensitive areas, etc.
  • Wooded area avoidance: many jurisdictions require that turbines avoided wooded areas which are considered to be wildlife breeding areas (Note also that there are wind resource reasons, as well, to avoid wooded areas).

A final wind farm design, which meets the objectives and constraints, is generally achieved through numerous design iterations with input from the wind farm developer, civil engineer, electrical engineer, environmental assessment consultant and wind resource consultant among others.

As an aid to wind farm design, realistic photomontages based on photographs of the local terrain can generally be produced using the wind farm design software. These are very useful for presentations at public meetings where individuals from the community will wish to see an accurate visualization of the wind project. A further refinement is the creation of a photo-animation that shows the turbine blades rotating.