Wind speed is undoubtedly the single most important measurement in a wind-resource assessment campaign. As the primary predictor of energy production at a prospective wind farm, pre-construction estimates of long-term wind speeds require constant scrutiny. The process of improving the measurement tools used to obtain that information plays a continually important role in the global wind-energy industry.
While much attention has recently been focused on the exciting possibilities of remote sensors and their ability to accurately measure wind speed, cup anemometers remain the workhorse of wind measurement and their data the foundation of utility-scale wind projects. In recent years, equipment solicitations have referenced class 1 anemometry with increasing frequency. Given the expanded use of this term, one has to wonder, why is class 1 anemometry so sought after.
Levels of accuracy
Contrary to what is sometimes portrayed, the quality range of commonly used anemometers, from economical to high-end, is actually quite small. All of these sensors provide accurate wind-speed measurements, especially when considering long-term averages. What’s more, there remains an important role for the more economical sensors, which let developers stretch limited early-stage budgets. This results in deploying a greater number of measurement devices, thus improving the odds of financing projects in an increasingly competitive landscape.
That said, the top-performing anemometer models also deliver unique advantages, for example in complex terrain, which can result in lower measurement uncertainty. While each of these sensor types has a place, there is room for improvement in terms of how to compare cup anemometers and their relative performance.
Defining class 1
The term, class 1, comes from the IEC standard for power performance measurements (IEC 61400-12-1: First Edition, 2005-12), whereby a process is defined for the classification of different anemometer models (Annexes I, J). The classification process requires running two serialized sensors through a battery of influence parameter tests, which are comprised of four categories: calibration, bearing friction over a specified temperature range, torque, and off-axis or inclined flow. Once these tests are complete, the results are input into a computer model that simulates different wind conditions. The model outputs a set of classification indices, or two alphanumeric scores, which are averaged for the two tested sensors and henceforth apply to all like-model sensors. For example, anemometer classification indices resulting from a test look like this for the WindSensor P2546-OPR:
The classification indices are differentiated by the letters A and B, which are meant to symbolize a sensor’s abilities in flat and complex terrain, respectively. The IEC standard has an individual threshold for both the A and B numbers that essentially qualify a sensor as class 1 and the lower the number, the lower the uncertainty (Type A: 1.7, Type B: 2.5). Finally, a sensor’s classification indices can be used to calculate its operational standard uncertainty – a quantitative measure of its accuracy in real-world operating conditions.
Assessing classification results
In exploring this class 1 topic, an important paradox appears which is not well known throughout the wind industry. The simplicity of the classification’s summary scores belies the complexity of the process by which they are obtained. In reality, the subtleties of this complex evaluation process can significantly alter the resulting scores. That is not to say the process is overly complex, just that attention should be paid beyond the results. Simulating an anemometer’s real-world performance is a daunting task so much credit is deserved by those who have developed and contributed to the process over the years.
Many classifications, however, have been carried out by different entities using different processes. In some cases, influence parameter tests are not backed by the full measurement data required in the IEC standard. Instead, the classification scores rely on extrapolated data.
For these reasons, summary results alone may be considered an incomplete basis for comparison. While many of the classifications provide reasonable indications of a sensor’s performance, the variation in methodology makes it difficult to compare the results of two different sensors without closer scrutiny. Ultimately, the wind industry as a whole will benefit from more transparency and standardization.
One should always keep in mind that the classification process, while broad in its approach to performance characterization, is not all-encompassing. Other practical considerations when evaluating cup anemometers include durability (such as cups breaking, resistance to electrical discharges from nearby lightning), repeatability (IEC requires only two anemometers for classification), stability over time, power consumption, ease of installation, price, availability, and so on – all of which can have a real and material effect on a project’s wind-resource assessment campaign.
Cup anemometers continue to perform a critical function in the wind industry, and when the highest level of performance is required, it’s important to think beyond the term class 1. Buying any anemometer that checks the class 1 box or choosing a sensor with the best advertised classification indices may provide an easy basis for selection. These performance scores, however, should only be one consideration among many when making this important decision. wpe