More quality in Wind Energy Yield Assessments with loss calculation based on time-series

Due to the development in the wind energy market, the demands on wind energy yield assessments are rising. Very exact yield prognoses of planned wind farms are more important than ever. The loss calculation based on time-series enables the calculation of the daily and seasonal distribution of the losses. This leads to an improved accuracy concerning the yield analysis. But only the consistent use of the loss calculation's advantages leads to a real added value.

Tuulivoimatuotanto - Julkaistu: 2.10.2018 - Teksti: Dominik Fremgen, Carmen Bartelmai

New challenges arise on project planner, wind consultants and operators due to recent developments in the wind energy market. Wind farm operators need wind energy yield assessments, which model precisely and realistically their wind farm's energy yield. Prospectively, greater attention will have to be paid to the calculation of yield losses, which are caused by restrictions, to approach this truth in wind energy yield assessments.

The loss calculation's quality depends essentially on the data basis as well as on the models

The loss calculation's accuracy is thereby ascertained by two factors; firstly by the data basis. Knowledge of the temperature and humidity, the exact distribution of the wind directions as well as the daily variation of the wind speeds in the planned hub heights is decisive for the amount of the calculated losses of operating restrictions.

However, if appropriate measurements are not available at the planned locations, only data based on reanalyses or nearby monitoring stations will be applicable. Since these data do not consider small-scale variations, they represent inevitably only an approximation with increased uncertainties.

Furthermore, the accuracy of the losses depends as well on the models used during the calculation.

 

 

 

 

 

 

 

 

 

 

Figure 1: Loss calculation based on time-series

Since restrictions are often limited to annual and daily times, we attain a improved accuracy by calculating the losses on the basis of time-series. Since wind energy yield assessments are usually realised with a statistical distribution of wind speed without time dependency such as by applying the WAsP-model, there is no data basis by which a time-series based calculation would be directly possible. In addition to the usual yield calculations on the basis of statistical wind velocity distributions, time series from reanalysis or measurements are adjusted to the level of the wind conditions determined in the wind energy yield assessment.

Afterwards the related yield based on each valid power curve (performance-optimized, noise-optimized or complete shutdown) is assigned to each time step. Which mode of operation should be considered at each time steps, is determined with the aid of the relevant parameters such as daytime and season, wind speed, wind direction or temperature.

The influence of the performance restrictions on the wake inside a wind farm is considered during these calculations by modelling the wind farm model in accordance to N. O. Jensen and calculating the impact for each time step. This is necessary, because the wind farm shadowing, caused by the WTGs, is reduced by shutting down the WTG and by power reduction.

Yield increase of several percent in wind farm is possible

These yield profits can represent a yield increase of several percent, especially in the event of turbulence shutdown of upstream wind turbines, but also in the event of shutdowns due to other operating restrictions. Previous methods for calculating the yield losses often neglect this impact with the result, that a overcharge of losses is calculated for the whole wind farm.

The loss calculation on the basis of time-series, developed by Ramboll, considers the coincidence of several restrictions as well and leads to a profound calculation of the wind farm's losses.

The nowadays increasing number of restrictions at individual sites leads to the question, whether and how the restrictions interact and how this knowledge is included in the calculation method afterwards. For example, when switching off to protect birds and simultaneous noise reduction, only the "stricter" condition must be considered.

These interactions are often neglected by other calculation methods. In some circumstances this can lead to relevant mistakes during the calculation of the total losses, as the following case shows.

The wrong handling of single losses leads to relevant mistakes

The pure addition of the losses, e. g. of a shutdown because of turbulences and a noise reduction, leads to an overestimation of the yield losses (see bar in the middle in figure 2), because both are effective at increased wind speeds. Therefore, they are not distributed evenly to each other.

The same holds true when the multiplication of the efficiencies [(1- loss1)*(1-loss­2)*…*(1-lossn)] is applied (see the right bar in the figure 2). This method leads to a wrong calculation of the total losses as well.

The Ramboll method, on the other hand, examines the interactions of several restrictions at each time step. This is the only way to accurately determine the mutual influence and calculate the actual total loss (see the left bar in the figure 2). Differences of 2 % can arise according to how the individual losses are combined.

 

 

 

 

 

 

 

 

 

 

Figure 2: overestimation of the total loss by wrong handling of single losses

Another benefit of the Ramboll method is, that it enables a detailed consideration of the losses' distribution concerning daytime and season. As a consequence, results can be made plausible easily.

Reduced uncertainties in the wind energy yield assessment by using the time-series based loss calculation

In the creation of a wind energy yield assessment, the method leads to an increased accuracy of the loss calculation and therefore to a reduction of the overall uncertainties.

The method constitutes a further step in improving the methods used to provide wind farm planners a profound basis for profitability calculations.

Read Finnish abstract.

Contact:

Ramboll Finland Oy

Veli-Pekka Alkula

Kiviharjunlenkki 1A

90220 Oulu

Finland

T: +358 50 3389 379

wind@ramboll.fi

Ramboll CUBE GmbH

Carmen Bartelmai

Breitscheidstr. 6

34119 Kassel

Germany

T. +495561 288573-0

Onshore-wind@ramboll.com

Background:

Ramboll is a leading and independent engineering, design and consultancy company founded in Denmark in 1945. The company employs 13 000 globally and has especially strong representation in the Nordics, UK, North America, Continental Europe, Middle East and Asia-Pacific. With more than 300 offices in 35 countries, Ramboll combines local experience with a global knowledgebase constantly striving to achieve inspiring and exacting solutions that make a genuine difference to its clients, the end-users, and society at large.

Ramboll provides comprehensive expert services for the various life cycle stages of wind power projects. Our wind power expertise combines innovative thinking and solid experience. The expertise is based on long and versatile work in the field of wind power from feasibility studies to dismantling of wind turbines. Our consultancy services are spanning a number of sectors, including production assessment, land use planning, the environment, impact assessments, transport, construction design and supervision as well as due diligence processes.

Read more: www.ramboll.com

Mainoskuva