Publications by RAM Associates:
RAMWind Analysis for Turbine Performance (download)
RAM Associates report
Summary: The original use for RAMWind was for producing wind speed models. However, by experimentation it has been observed that turbine power can also be modeled directly, substituting average power at the turbine sites for wind speed at met tower sites as model inputs. This technique has been able to identify turbines that perform better than average, due to certain performance enhancements and also identify turbines that appear to perform below average.
Wind Farm Wake Losses – Summary of Past & Current Work (download)
AWEA Wind Resource Assessment Seminar, 2013, Las Vegas, NV
Presented by Jack Kline, RAM Associates
Summary: Since first working at US WindPower (later Kenetech), Jack Kline has conducted wake loss studies by varying methodologies. In the early days with USW and Howden Wind Parks, wake loss studies were performed by turning upwind groups of turbines off and on and the power output from downwind, waked, turbines was observed to determine what the energy losses were. These tests were sometimes accompanied by wind speed measurements.
Exposure-based Wind Flow Modeling with a Single Met Site (download)
AWEA Wind Resource Assessment Seminar, 2013, Las Vegas, NV
Presented by: Liz Walls and Jack Kline, RAM Associates
Summary: The RAMWind model performs best when there are four or more meteorological towers on a site with which to describe the variation in wind speed with terrain exposure. Modeling with RAMWind can be done with as few as two towers, although the accuracy is limited under those circumstances. However, at present, modeling a wind site with one met tower has heretofore not been possible with RAMWind. In this presentation, the results from an analysis of models produced at project sites over a broad range of terrain complexities (low to high levels of terrain exposure) shows a relationship between the sensitivity of wind speed to terrain exposure which is then used to develop a methodology by which a single met tower can be used to model wind speeds at a project. One of the key findings from this analysis is that the relative sensitivity of wind speed to exposure level decreases as terrain exposure increases.
Terrain Effects on Wind Speed Enhanced by Atmospheric Stability (download)
AWEA Wind Resource Assessment Workshop 2012, Pittsburgh, PA
Presented by: Jack Kline, RAM Associates
Summary: The thermal stability of the atmosphere has been observed to have significant and profound effects on various aspects of the wind and effects on turbine performance. Under stable conditions vertical wind shear tends to increase and wake losses within a wind farm are more severe, due to reduced mixing of the wake deficits. But beyond these effects, the wind’s response to the terrain is enhanced under stable conditions. In this presentation and analysis of the relationship between wind speed and terrain exposure is analyzed at a project in the Great Plains under various flow conditions. The analysis clearly shows that the response of the wind to downwind exposure is heightened, that is, the sensitivity of wind speed to terrain elevation variation in the downwind direction increases, as stability increases. This can produce either higher or lower wind speeds depending on the position of a given site (met tower or turbine) with respect to the wind direction and terrain formations.
Understanding and Quantifying Diurnal Wake Losses and The Impact on Overall Wind Farm Performance (download)
AWEA WINDPOWER 2012, Atlanta, GA
Presented by: Liz Walls, RAM Associates
Summary: When designing the layout of a wind farm, one of the main objectives is to minimize lost production due to turbine wake losses. While effort is made to keep wake losses as small as possible, there will inevitably be impact on the wind farm energy production and the total effect of the wake losses needs to be quantified. Currently, there are wake loss models used in the industry to assess the overall impact of these losses on a wind farm. While it has been shown that these models can capture the total loss due to wakes, these models do not consider differences in the wake losses due to changes in atmospheric stability (i.e. stable vs. neutral to unstable atmospheric conditions). This study demonstrates the large differences in wake losses during different atmospheric conditions.
Mapping of Geographical Trends of Wind Shear and Frequency of Extreme Wind Shear Events Across the Great Plains (download)
AWEA WindPower Conference 2011, Anaheim, CA
Presented by: Jack Kline, RAM Associates
Summary: In this presentation, wind shear measured at more than 85 tower and Sodar sites across the Great Plains are presented. The mean diurnal shear exponents are shown and geographical trends are observed. Frequency of extreme shear events, where the power law exponent exceeds 0.4, are presented and it was found that, on average, extreme shear occurs more than 20% of the time. Wind turbine simulations were conducted using NREL's FAST software and the results showed that high shear can have significant effects on cyclical loading experienced at the blade root.
Long-term Wind Speed Estimates from Short-term Data: So Many Ways to Get it Wrong! (download)
AWEA Wind Resource Assessment Workshop 2010, Oklahoma City, OK
Presented by: Liz Walls, RAM Associates
Summary:In this presentation, the accuracy of long-term wind speed estimates based on surface stations as reference was tested. The long-term estimates found from several MCP (Measure-Correlate-Predict) techniques and a Matrix/Lag1 method were compared. It was found that, since the relationship between the target and reference site is not constant, errors in this type of approach are inevitable.
Wind Issues Affecting Wind Projects from Mega Wind Farms to Instrumentation (download)
AWEA Windpower 2008
Presented by: Jack Kline, RAM Associates
Summary: This presentation discusses some issues that can affect energy estimations a wind farm sites, from large-scale wake effects produced from one wind farm impinging upon another to wind speed errors produced by anemometer instability of the NRG#40 anemometer. This was the first industry presentation related to the pervasive problem of Dry Friction Whip, which affected thousands of these anemometers. This problem was first reported to NRG Systems by RAM Associates.
A New and Objective Empirical Model of Wind Flow Over Terrain (download)
AWEA Wind Resource & Project Assessment Workshop 2007, Portland, OR
Presented by: Jack Kline, RAM Associates
Summary: This was the first introduction of the RAMWind model to the industry. This model is unique as it essentially self-calibrates itself to the local wind conditions, by using the observed wind data to dictate the relationship between terrain exposures and wind speed.
Wind Speed Measurement and Turbine Performance Projection Variance between NRG Max40 and Risoe P2546A Anemometers (download)
AWEA Windpower 2005, Denver, CO
Presented by: Jack Kline, RAM Associates
Summary: All anemometers are not created equal, nor do they necessarily measure the wind speed equally. This presentation discusses the differences in performance between two anemometers, the NRG#40, the most commonly used anemometer for wind resource assessment in the US, and the Riso P2546A, an anemometer often used for turbine power curve verification.
Effects of Tubular Anemometer Towers on Wind Speed Measurements (download)
AWEA Windpower 2004
Presented by: Jack Kline, RAM Associates
Summary: This paper was the first to show the effects of anemometer towers themselves on wind speeds for both side-mount booms and top-mount stub masts. The top-mount stub masts had been in wide use up to that time, but research by RAM, as presented at the conference, showed large errors in wind speed would result (over-speeding). Since this presentation this kind of anemometer mounting has been largely, if not entirely, abandoned.
Field Calibration of Maximum Cup, Climatronics and Met One Anemometers (download ppt or report)
AWEA Windpower 1999, Burlington, VT PPT
Presented by: Jack Kline, RAM Associates
Summary:
Recent Experience with the NOABL Model in Highly Complex Terrain
AWEA Windpower 1993, San Francisco, CA
Presented by: Jack Kline
Summary:
Turbulence Increases Due to Wind Turbines on an Operating Wind Park (download)
AWEA Windpower 1989, San Francisco, CA
Presented by: Jack Kline, RAM Associates
Summary: This paper presents results from wake tests performed at the Howden Wind Park in the Altamont Pass and investigates the changes in turbulence levels, both horizontal and vertical, wind speed and turbine power deficits due to turbine wakes under stable and unstable (neutral-unstable) atmospheric conditions.
Turbulence Characteristics at Howden Wind Park (download)
AWEA Windpower 1988, Honolulu, HI
Presented by: Jack Kline, Howden Wind Parks
Summary: This presentation includes results of detailed analysis of turbulence characteristics at the Howden Wind Park in the Altamont Pass of California. Details such as frequency and magnitude of 1-second gusts, and magnitudes of horizontal shear in the presence of turbine wakes.
Publications by others that reference RAM:
Deep Array Wake Loss in Large Onshore Wind Farms (A Model Validation) (download)
AWEA Wind Resource Workshop 2010, Oklahoma City, OK
Presented by: Justin Wolfe, E.ON Climate & Renewables
Summary: In this presentation, a number of wake models, including the RAM Wake model, were used at an operating wind farm to test their ability to accurately calculate wake losses in a "deep array' consisting of at least five rows of turbines. The results showed that all models tested produced results consistent with the observed wake losses. It is noted that the RAMWind model was used to normalize the turbine performance data for terrain effects, rather than any commercial wind flow model.
Wind Flow Modeling Software Comparison (download)
AWEA Wind Resource Assessment Workshop 2009, Minneapolis, MN
Presented by: John Vanden Bosche, Chinook Wind
Summary: In this presentation, wind flow model results are compared at two project sites with moderately complex terrain. The models include the industry standard, WaSP, some CFD (Computational Fluid Dynamics) models and the RAMWind model. At both sites the RAMWind model produced more accurate results than all other models.
Comparison of WAsP, MS-Micro/3, CFD, MWP, and Analytical Methods for Estimating Site-Wide Wind Speeds (download)
AWEA Wind Resource and Project Energy Assessment Workshop 2009, Minneapolis, MN
Presented by: David VanLuvanee, GEC-DNV
Summary: In this presentation, wind flow model results are compared at four project sites with moderately to highly complex terrain. The models include the industry standard, WaSP, some CFD (Computational Fluid Dynamics) models and the RAMWind model. At three of the four sites the RAMWind model produced more accurate results than the other models.
RAM Reference Material
Summary of RAM's Wake Loss Model (download)
Prepared by Jack Kline, RAM Associates, September 2009
Summary: This document discusses the methodology of the RAM Wake model for calculating energy deficits in wind farms due to turbine wake effects. This model was validated in a recent study, the results of which were presented by Eon Climate and Renewables at the Windpower 2010 Conference and the Wind Resource and Project Energy Assessment Workshop 2010.
RAM Matrix/Lag-1 Hybrid Wind Speed Data Estimation (download)
Prepared by Jack Kline, RAM Associates, November 2009
Summary: This document discusses the methodology of the RAM Matrix/Lag-1 technique for wind speed estimation. This technique is used to estimate missing data where necessary, and for long-term wind speed estimation at project met sites.
RAMWind Analysis for Turbine Performance (download)
RAM Associates report
Summary: The original use for RAMWind was for producing wind speed models. However, by experimentation it has been observed that turbine power can also be modeled directly, substituting average power at the turbine sites for wind speed at met tower sites as model inputs. This technique has been able to identify turbines that perform better than average, due to certain performance enhancements and also identify turbines that appear to perform below average.
Wind Farm Wake Losses – Summary of Past & Current Work (download)
AWEA Wind Resource Assessment Seminar, 2013, Las Vegas, NV
Presented by Jack Kline, RAM Associates
Summary: Since first working at US WindPower (later Kenetech), Jack Kline has conducted wake loss studies by varying methodologies. In the early days with USW and Howden Wind Parks, wake loss studies were performed by turning upwind groups of turbines off and on and the power output from downwind, waked, turbines was observed to determine what the energy losses were. These tests were sometimes accompanied by wind speed measurements.
Exposure-based Wind Flow Modeling with a Single Met Site (download)
AWEA Wind Resource Assessment Seminar, 2013, Las Vegas, NV
Presented by: Liz Walls and Jack Kline, RAM Associates
Summary: The RAMWind model performs best when there are four or more meteorological towers on a site with which to describe the variation in wind speed with terrain exposure. Modeling with RAMWind can be done with as few as two towers, although the accuracy is limited under those circumstances. However, at present, modeling a wind site with one met tower has heretofore not been possible with RAMWind. In this presentation, the results from an analysis of models produced at project sites over a broad range of terrain complexities (low to high levels of terrain exposure) shows a relationship between the sensitivity of wind speed to terrain exposure which is then used to develop a methodology by which a single met tower can be used to model wind speeds at a project. One of the key findings from this analysis is that the relative sensitivity of wind speed to exposure level decreases as terrain exposure increases.
Terrain Effects on Wind Speed Enhanced by Atmospheric Stability (download)
AWEA Wind Resource Assessment Workshop 2012, Pittsburgh, PA
Presented by: Jack Kline, RAM Associates
Summary: The thermal stability of the atmosphere has been observed to have significant and profound effects on various aspects of the wind and effects on turbine performance. Under stable conditions vertical wind shear tends to increase and wake losses within a wind farm are more severe, due to reduced mixing of the wake deficits. But beyond these effects, the wind’s response to the terrain is enhanced under stable conditions. In this presentation and analysis of the relationship between wind speed and terrain exposure is analyzed at a project in the Great Plains under various flow conditions. The analysis clearly shows that the response of the wind to downwind exposure is heightened, that is, the sensitivity of wind speed to terrain elevation variation in the downwind direction increases, as stability increases. This can produce either higher or lower wind speeds depending on the position of a given site (met tower or turbine) with respect to the wind direction and terrain formations.
Understanding and Quantifying Diurnal Wake Losses and The Impact on Overall Wind Farm Performance (download)
AWEA WINDPOWER 2012, Atlanta, GA
Presented by: Liz Walls, RAM Associates
Summary: When designing the layout of a wind farm, one of the main objectives is to minimize lost production due to turbine wake losses. While effort is made to keep wake losses as small as possible, there will inevitably be impact on the wind farm energy production and the total effect of the wake losses needs to be quantified. Currently, there are wake loss models used in the industry to assess the overall impact of these losses on a wind farm. While it has been shown that these models can capture the total loss due to wakes, these models do not consider differences in the wake losses due to changes in atmospheric stability (i.e. stable vs. neutral to unstable atmospheric conditions). This study demonstrates the large differences in wake losses during different atmospheric conditions.
Mapping of Geographical Trends of Wind Shear and Frequency of Extreme Wind Shear Events Across the Great Plains (download)
AWEA WindPower Conference 2011, Anaheim, CA
Presented by: Jack Kline, RAM Associates
Summary: In this presentation, wind shear measured at more than 85 tower and Sodar sites across the Great Plains are presented. The mean diurnal shear exponents are shown and geographical trends are observed. Frequency of extreme shear events, where the power law exponent exceeds 0.4, are presented and it was found that, on average, extreme shear occurs more than 20% of the time. Wind turbine simulations were conducted using NREL's FAST software and the results showed that high shear can have significant effects on cyclical loading experienced at the blade root.
Long-term Wind Speed Estimates from Short-term Data: So Many Ways to Get it Wrong! (download)
AWEA Wind Resource Assessment Workshop 2010, Oklahoma City, OK
Presented by: Liz Walls, RAM Associates
Summary:In this presentation, the accuracy of long-term wind speed estimates based on surface stations as reference was tested. The long-term estimates found from several MCP (Measure-Correlate-Predict) techniques and a Matrix/Lag1 method were compared. It was found that, since the relationship between the target and reference site is not constant, errors in this type of approach are inevitable.
Wind Issues Affecting Wind Projects from Mega Wind Farms to Instrumentation (download)
AWEA Windpower 2008
Presented by: Jack Kline, RAM Associates
Summary: This presentation discusses some issues that can affect energy estimations a wind farm sites, from large-scale wake effects produced from one wind farm impinging upon another to wind speed errors produced by anemometer instability of the NRG#40 anemometer. This was the first industry presentation related to the pervasive problem of Dry Friction Whip, which affected thousands of these anemometers. This problem was first reported to NRG Systems by RAM Associates.
A New and Objective Empirical Model of Wind Flow Over Terrain (download)
AWEA Wind Resource & Project Assessment Workshop 2007, Portland, OR
Presented by: Jack Kline, RAM Associates
Summary: This was the first introduction of the RAMWind model to the industry. This model is unique as it essentially self-calibrates itself to the local wind conditions, by using the observed wind data to dictate the relationship between terrain exposures and wind speed.
Wind Speed Measurement and Turbine Performance Projection Variance between NRG Max40 and Risoe P2546A Anemometers (download)
AWEA Windpower 2005, Denver, CO
Presented by: Jack Kline, RAM Associates
Summary: All anemometers are not created equal, nor do they necessarily measure the wind speed equally. This presentation discusses the differences in performance between two anemometers, the NRG#40, the most commonly used anemometer for wind resource assessment in the US, and the Riso P2546A, an anemometer often used for turbine power curve verification.
Effects of Tubular Anemometer Towers on Wind Speed Measurements (download)
AWEA Windpower 2004
Presented by: Jack Kline, RAM Associates
Summary: This paper was the first to show the effects of anemometer towers themselves on wind speeds for both side-mount booms and top-mount stub masts. The top-mount stub masts had been in wide use up to that time, but research by RAM, as presented at the conference, showed large errors in wind speed would result (over-speeding). Since this presentation this kind of anemometer mounting has been largely, if not entirely, abandoned.
Field Calibration of Maximum Cup, Climatronics and Met One Anemometers (download ppt or report)
AWEA Windpower 1999, Burlington, VT PPT
Presented by: Jack Kline, RAM Associates
Summary:
Recent Experience with the NOABL Model in Highly Complex Terrain
AWEA Windpower 1993, San Francisco, CA
Presented by: Jack Kline
Summary:
Turbulence Increases Due to Wind Turbines on an Operating Wind Park (download)
AWEA Windpower 1989, San Francisco, CA
Presented by: Jack Kline, RAM Associates
Summary: This paper presents results from wake tests performed at the Howden Wind Park in the Altamont Pass and investigates the changes in turbulence levels, both horizontal and vertical, wind speed and turbine power deficits due to turbine wakes under stable and unstable (neutral-unstable) atmospheric conditions.
Turbulence Characteristics at Howden Wind Park (download)
AWEA Windpower 1988, Honolulu, HI
Presented by: Jack Kline, Howden Wind Parks
Summary: This presentation includes results of detailed analysis of turbulence characteristics at the Howden Wind Park in the Altamont Pass of California. Details such as frequency and magnitude of 1-second gusts, and magnitudes of horizontal shear in the presence of turbine wakes.
Publications by others that reference RAM:
Deep Array Wake Loss in Large Onshore Wind Farms (A Model Validation) (download)
AWEA Wind Resource Workshop 2010, Oklahoma City, OK
Presented by: Justin Wolfe, E.ON Climate & Renewables
Summary: In this presentation, a number of wake models, including the RAM Wake model, were used at an operating wind farm to test their ability to accurately calculate wake losses in a "deep array' consisting of at least five rows of turbines. The results showed that all models tested produced results consistent with the observed wake losses. It is noted that the RAMWind model was used to normalize the turbine performance data for terrain effects, rather than any commercial wind flow model.
Wind Flow Modeling Software Comparison (download)
AWEA Wind Resource Assessment Workshop 2009, Minneapolis, MN
Presented by: John Vanden Bosche, Chinook Wind
Summary: In this presentation, wind flow model results are compared at two project sites with moderately complex terrain. The models include the industry standard, WaSP, some CFD (Computational Fluid Dynamics) models and the RAMWind model. At both sites the RAMWind model produced more accurate results than all other models.
Comparison of WAsP, MS-Micro/3, CFD, MWP, and Analytical Methods for Estimating Site-Wide Wind Speeds (download)
AWEA Wind Resource and Project Energy Assessment Workshop 2009, Minneapolis, MN
Presented by: David VanLuvanee, GEC-DNV
Summary: In this presentation, wind flow model results are compared at four project sites with moderately to highly complex terrain. The models include the industry standard, WaSP, some CFD (Computational Fluid Dynamics) models and the RAMWind model. At three of the four sites the RAMWind model produced more accurate results than the other models.
RAM Reference Material
Summary of RAM's Wake Loss Model (download)
Prepared by Jack Kline, RAM Associates, September 2009
Summary: This document discusses the methodology of the RAM Wake model for calculating energy deficits in wind farms due to turbine wake effects. This model was validated in a recent study, the results of which were presented by Eon Climate and Renewables at the Windpower 2010 Conference and the Wind Resource and Project Energy Assessment Workshop 2010.
RAM Matrix/Lag-1 Hybrid Wind Speed Data Estimation (download)
Prepared by Jack Kline, RAM Associates, November 2009
Summary: This document discusses the methodology of the RAM Matrix/Lag-1 technique for wind speed estimation. This technique is used to estimate missing data where necessary, and for long-term wind speed estimation at project met sites.