 
 
OPERATION – TR10 
1  Introduction, definitions and workflow overview ............................................................................. 2 
1.1  Basic concept ........................................................................................................................ 2 
1.2  Workflow overview ................................................................................................................. 2 
2  Data format and project preparation .................................................................................................. 3 
2.1  Format of the SCADA data .................................................................................................... 3 
2.2  Format of external data ......................................................................................................... 4 
2.3  Create Existing WTG objects ................................................................................................ 5 
2.4  Create a new PERFORMANCE CHECK session ................................................................. 6 
3  Data: Import SCADA ............................................................................................................................ 7 
3.1  Import 10-minute data............................................................................................................ 7 
3.2  Pair and load ........................................................................................................................ 12 
3.3  Setup categories .................................................................................................................. 15 
4  Data: Time series SCADA ................................................................................................................. 23 
5  Data: Import SOLD ............................................................................................................................. 24 
6  Data: Plausibility ................................................................................................................................ 26 
7  Site yield: Wind speed correlation ................................................................................................... 28 
8  Site yield: Wind speed regression ................................................................................................... 31 
9  Site yield: Moving power curves ...................................................................................................... 32 
10  Result .............................................................................................................................................. 34 
11  Quality factor .................................................................................................................................. 35 
12  Report ............................................................................................................................................. 35 
 
   

Introduction, definitions and workflow overview 2

13.1 Introduction, definitions and workflow overview

Based on the “Erneuerbare-Energien-Gesetz 2017” (“EEG 2017”), the owners of German wind farms

commissioned after 1

of January 2018 are obliged to report the produced electricity after 5, 10 and 15 years of

operation. The complete calculation of the Site yield and Quality factor are described in the Technical Guideline

No. 10 (hereinafter TR10), which is the detailed methodology description introduced in the EEG 2017, Annex 2.

The “Quality factor calculation (TR10)” sub-module includes the full implementation of the TR10, rev. 3. In

order to use it, a separate license of this tool as well as a license for PERFORMANCE CHECK are required.

Besides these, the license for the module BASIS is required. Additionally, the module METEO is recommended,

as the TR10 stipulates the use of mesoscale and / or reanalysis data and such data (e.g. EmdWrf Europe+,

ERA5, MERRA2, etc.) can be downloaded with this module.

13.1.1 Basic concept

The concept of the TR10 tool fully follows the TR10 guideline, rev. 3. The analyzed WTG(s) are represented in

windPRO by “Existing WTG objects” . The data import and the data analysis to Existing WTG objects is done

in the TR10 tool, which is a sub-module of the module PERFORMANCE CHECK. The basic concept of the TR10

guideline, as well as of the TR10 tool, is as follows:

1) Assignment of status codes to categories (sorted from the highest to the lowest priority)

a. Category 2: Constraint or not available due to other matters

b. Category 4: Constraint due to optimized selling

c. Category 3: Constraint due to feed-in management

d. Category 1: Constraint due to permit matters

e. Category 0: Normal operation

2) Assignment of the 10-minute time stamps to the categories using the status codes and 1)

3) Calculation of electricity not produced due to categories 2, 3 and 4.

a. Calculation of correlations between wind speed time series

b. Creating consistent wind time series

c. Creating moving power curves

4) Generating results

13.1.2 Workflow overview

The evaluation described in this manual requires the licensed modules BASIS, PERFORMANCE CHECK, TR10

and optionally METEO.

The workflow is as follows:

• Create new “Existing WTG” objects.

• Download meso- and / or reanalysis data using METEO object(s) (optional).

• Start PERFORMANCE CHECK or QUALITY FACTOR CALCULATION from the ribbon Loads &

Operation.

• Select “Quality factor calculation (TR10)” on the tab “Concept choice”.

• Import 10-minute SCADA data with production, wind speed, wind direction, accumulated power

production and other required signals and load this data into the “Existing WTG” objects using the “Paid

and load” button.

• Import and merge status codes from turbine log files to 10-minute production data using the “Setup

categories” button.

• Import the “Sold electricity”.

Technically not available

Einspeisemanagement. Measure used by the TSOs in Germany to stabilize the electrical grid by reducing the

actual power generation in case of overproduction in the grid.

Data format and project preparation 3

• Evaluate the plausibility.

• Create “Consistent wind speed time series” according to the TR10 chapter 6.1 using the tabs “Wind

speed correlation” and “Wind speed regression”.

• Create applicable power curves according to the TR10 chapter 6.2 using the tab “Moving power curves”.

• Go to “Site yield → Results” and use the buttons “Calculate production” and “Calculate site yield” to get

the results.

• Go to the tab “Quality factor”. Entering the Referenzertrag

to the WTG(s) finishes the calculation of the

“Quality factor”, which is the final result of the whole calculation process.

• Go to the Tab “Report” and generate the report(s) for the grid operator.

13.2 Data format and project preparation

13.2.1 Format of the SCADA data

The TR10 guideline stipulates that the raw SCADA data is used for the analysis. This assumes that the format

exported by the SCADA software is in a format that windPRO is capable to read. Standard export features of a

SCADA system typically contain an option to export the data in e.g. *.csv or *.txt files. These formats are

supported by windPRO.

There are two main data sources from the SCADA systems – the 10-minute production data and the status logs.

Both use separate import mechanisms within the TR10 tool, as the structure of these sources is different.

13.2.1.1 10-minute SCADA data

The imported 10-minute production data must be in a “matrix” form with header in the first line, with the time

stamp in the first column and with the data in the following columns. Besides this, the data importer is quite

flexible, and can handle data in single or multiple files, e.g. one turbine per file; all turbines in one file; one turbine

per day, per month etc. in one file.

Figure 1 Example of SCADA data opened in MS Excel

It is important to have an identifier (ID) of the turbine in the file headers, in a column or in the file names (if one

file per turbine). Later this ID will be used to pair the SCADA data with an Existing WTG object. It is useful to

name the existing WTG(s) with the identical names that the SCADA system uses, as windPRO can then pair the

existing WTG objects with the corresponding SCADA data automatically. For the purpose of the TR10 it can be

Reference production published by FGW for each WTG type and hub height according to the EEG.

Data format and project preparation 4

also practical to use the EEG-Anlagenschlüssel

as WTG ID. More details regarding the WTG IDs can be found

in chapters 13.3.1 and 13.3.2.

13.2.1.2 Status logs

The preferred formats of the status logs are *.csv or *.txt, too. The status logs are typically exported in a format

with the beginning and the end (or duration) of an event. The events are then described with status codes,

potentially followed by sub-codes. These are often followed by a description. The irregular structure in the status

logs is going to be transformed into a regular 10-minute structure by windPRO including the assignment of the

10-minute time stamp to exactly one TR10 category according to TR10 requirements.

Figure 2 Status logs example

Some of the SCADA systems can deliver the 10-minute production time series including the already assigned

status code. The signals name can be named e.g. “First alarm in the 10-minute interval” or similar. The use of

such signal is not in compliance with the TR10. The assignment of the status codes to 10-minute time stamps

was done by the SCADA system itself, but it is required that this is done by an independent tool. Although the

use of such signal is not compliant with the TR10 text, the TR10 tool allows its use for testing purposes and

internal calculations. This triggers a deviation-warning and is reported in the pdf report “Deviations from TR10”.

The import of the status codes is described in chapter 13.3.3.

13.2.2 Format of external data

On the top of the SCADA data, some external data is needed to finalize the analysis according to the TR10.

13.2.2.1 EinsMan and optimized marketing start & end times

It can be the case that EinsMan events (category 3) or optimized selling events (category 4) do not have explicit

status codes in the WTG SCADA. If so, the beginnings and the ends of these events must be imported from an

external source. The format of the imported data can be again either a *.csv or *.txt file.

WTG specific ID used e.g. by invoicing. Consists of 33 symbols and begins with an “E”.

Data format and project preparation 5

Figure 3 EinsMan and optimized selling example

The EinsMan Nr. in the screenshot above is just for instance; the only two necessary columns are the ones with

the start and end (resp. with the duration). The import of the EinsMan & Optimized selling data is described in

chapter 13.3.3.3.

13.2.2.2 Status codes library

According to the TR10, the events logged in the status log shall be assigned to TR10 categories and then to the

regular 10-minute time series. The assignment of the WTG-type specific status codes to the TR10 categories

shall be done by an independent body that has proven to the FGW the necessary prerequisites and qualifications

of the authorised persons. The approval of the authorised person is granted by the responsible FGW advisory

board "EEG-Kategorisierung".

The current implementation in the TR10 module assumes the use of a list in a form of *.csv or *.txt file with a

header in the first line and the list of the status codes with a main code, optionally secondary code, text

description and corresponding TR10 category for each of the status codes in the lines below the header.

Figure 4 Status list example

A detailed description can be found in chapter 13.3.3.5.

13.2.3 Create Existing WTG objects

Either manually by inserting an Existing WTG object , copy-paste the coordinates from a spread sheet or

download turbine positions directly through the Online WTG Data tool . Alternatively, you can convert New

WTG objects into Existing WTG objects by copy-pasting and using the paste option “Edit object(s) before

pasting”. Remember to assign an ID to the WTG object to either Description or User label. It is practical to use

the WTG ID from the SCADA system as “Description” and the Anlagenschlüssel as “User label”:

Data format and project preparation 6

Figure 5 Existing WTG object

13.2.4 Create a new PERFORMANCE CHECK session

Start PERFORMANCE CHECK module using “Quality factor calculation” from the tab “Loads & Operation“:

Figure 6 Starting the module from Loads & Operation tab

Create a new session and name it. As soon as done, confirm with Ok and enter the session by double clicking

it in the list of the sessions.

Figure 7 Create a new TR10 session

Currently, the PERFORMANCE CHECK module supports three different concept choices:

- Model validation / calibration / performance check

- Post construction evaluation following IEC 61400-26-1 and 61400-26-2

- Quality factor calculation (TR10)

In order to proceed with the TR10 analysis, please select the TR10 concept and confirm with Ok. As always,

windPRO is recommending the next move by coloring with green color. In order to proceed with another kind of

analysis, please follow the dedicated manuals.

Data: Import SCADA 7

Figure 8 PERFORMANCE CHECK - Concept choice

13.3 Data: Import SCADA

To run the TR10 analysis (and any PERFORMANCE CHECK analysis in general), you need to import SCADA

data, through the following process:

• Prepare / export the SCADA data as *.csv or *.txt files from the SCADA system.

• Create Existing WTG objects (including IDs).

• Start PERFORMANCE CHECK and load the data.

• Setup the import filter (Auto detect).

• Pair and load.

• Merge the status logs with the 10-minute-based time series.

13.3.1 Import 10-minute data

The structure of the import filter in the module PERFORMANCE CHECK (no matter on the concept choice) is

similar to the METEO object.

Data: Import SCADA 8

Figure 9 Add files

The first step is to import the 10-minute SCADA data from the WTG(s). windPRO is advising the next logical

step of yours and proposes it by green coloring. Use the button “Add file(s)” (or “Add folder”) and select the files

(or folder with files).

Figure 10 Define signals (1)

During the very first import of 10-minute data, the autodetect function (1) is typically very helpful. With this,

windPRO tries to identify the signals in the headers of the 10-minute files. The user has the option to modify the

suggested structure and specify the line with the header, the first line with data and field separators. At any step,

the user can preview the file structure (using the button “View file”) and adjust file formats if required.

When the setup of the import filter is finished, it can be saved with “Save” (2) in *.pci format and re-used, e.g.

when the user does an analysis of the same project once again from scratch or when preparing an analysis of

the WTG type with the same or similar structure of the SCADA data. In such case, the *.pci file can be simply

loaded with the “Load” button (3). The small “+” button in the left bottom corner (4) allows the user to add lines

(signals) manually.

Data: Import SCADA 9

Figure 11 Define signals (2)

Following signals are necessary / recommended for a TR10 analysis:

• Date/Time (mandatory)

• Wind speed (mandatory)

• Wind direction, resp. nacelle position (mandatory)

• Production (mandatory)

• Cumulative production (not-presence triggers a deviation to TR10)

• Round per minute (not-presence triggers a deviation to TR10)

• Pitch angle (not-presence triggers a deviation to TR10)

• Ambient temperature (not-presence triggers a deviation to TR10)

• Air pressure (recommended)

• Error code (if available in the 10-minute time series).

In case of missing signals that are relevant for the plausibility check according to the TR10 chapter 4.5, following

warning appears after moving to “Pair and load”:

The user can decide to ignore it, which causes that the plausibility check will be performed using only the

available signals. Such incompliance with the TR10 is documented in the report.

Data: Import SCADA 10

For some signals it is necessary to define the sub-type: Is the wind speed signal representing the mean, max,

min or std? Is the production signal accumulated? Once the sub-type is defined, and the units are set, the data

is converted and appears in the rightmost column. Please note that you also must specify the unit for the time

stamp. Specific care must be taken in setting the unit of the power signal correctly. It is, however, at any time

possible to return and change the unit. In such case, the data must be re-loaded using “Pair and load”.

Figure 12 Signals multi-edit

Multi-editing of the signal is possible by marking the signals you want, dragging the mouse or by “Shift-click” (1),

specify the type, sub-type and the unit (2) and finally pressing “Apply” (3).

Sometimes, it is necessary to change text to numbers. E.g. if your date stamp contains text like “DEC” you can

translate this into “12”. Or if invalid data is marked “NAN” you can change it to “-999”.

A very important step is the setup of the time stamp, resp. whether the SCADA data represents the beginning of

the period or the end of the period. This selection is in default setup empty and the user is not allowed to continue

in the analysis until either the end of the beginning of a period is selected.

For example, does the time stamp “02.10.2018 10:00” represent the interval “09:50 – 10:00” or “10:00 – 10:10”?

Typically, the manufacturers are using the format “End of a period”, but there can be exceptions. If you are not

sure, please clarify with the WTG manufacturer.

In the next step of the process, it is necessary to assign the production data to a specific WTG. There are several

options how to define the WTG ID setup. The WTG ID can be located either in a file(s) with the SCADA data or

in the name(s) of the file(s). In case you know where to find the WTG-ID, you can directly set it up (1). If you do

not know, where to find it, a “WTG-ID Guide” will help you with the necessary specifications (2).

Figure 13 WTG ID setup using the guide

Data: Import SCADA 11

For example, the advanced setup can be helpful when importing data from more files, where the WTG ID is a

part of the file name. You can define the symbol before and after the WTG ID string. E.g., if the names of the

IDs of the WTG(s) are the same as the names of the files, the determiner behind the WTG ID string is the symbol

“.”:

→

Figure 14 Advanced WTG ID setup

Another option is to have the WTG ID in the header directly next to the names of each of the signals. If this is

your case, mark the position (click on it with the mouse…), where the ID of the first WTG can be found (red

marked), then do the same for the second WTG (green marked) and optionally for the last WTG (violet).

Data: Import SCADA 12

Figure 15 WTG ID setup using ID in Header

If the setup of the WTG ID has been successful, they will now show up:

Figure 16 WTG ID check

13.3.2 Pair and load

When the Import filter is done, you can proceed with the loading of the data using the “Pair and load” button.

Should windPRO detect that some signals essential for the TR10 analysis are missing, you might be warned.

Use the green “Add” button to add the “Existing WTG objects” representing the WTGs that are going to be the

subject of the assessment.

Then the “Select WTG(s)” window appears. Activate the layer with the existing WTG objects that are the subject

of the analysis. This is done by checking the checkbox by one (or more) layers in the upper part of the window.

Data: Import SCADA 13

Figure 17 Select existing WTG objects

When done, select the existing WTG object(s) by checking their checkboxes in the lower part of the window.

When all WTGs are selected, confirm with Ok.

In case the naming convention of the existing WTG objects is identical to the WTG ID used in the files with the

SCADA data, windPRO will automatically pair the objects with the production data using the button “Auto pair”.

If the automatic pairing is not possible, the user can pair them manually using the drop-down selection menu

that is available in each line, resp. by each of the WTGs – see column “WTG ID from import”.

Figure 18 Load SCADA data into the existing WTG objects

Before loading the data with the green button “Load data” and leaving the “Pair and load” window, double check

and potentially correct the nominal wind speed(s) and nominal power output(s) by the day and night

modes. These are used for the calculation of the moving power curves according to chapter 6.2.

“Use night mode” is activated in default to fulfill the requirement to calculate the moving power curves for day

and night separately. The correct nominal wind speed and nominal power output of the day / night mode should

Data: Import SCADA 14

be defined here manually by the user. The default time for night in Germany is 10PM – 6AM. Any change or

deactivation of this feature is reported in the pdf report “Deviations from TR10”, which the user is notified about.

Besides that, please pay attention to the checkbox “Time series in local time”.

This checkbox is activated in the default setup. It means it is assumed that the imported time series includes

already the switching between summer and winter time. If the imported time series does not have switching

between summer and winter time, the checkbox needs to be deactivated. This causes that the nighttime in winter

is still considered as 22-6h, but the nighttime in summer is shifted to 21-5h.

When the setup is finished, click “Load data”, so the time series are imported into the “Existing WTG objects”.

During the import of the time series, windPRO checks their consistency. Should there be a time stamp that is

not in the regular 10-minute format, windPRO will notify the user and request a corrective measure:

When confirming with Ok, windPRO will assign this irregular time stamp to the nearest regular one. This means

that windPRO firstly checks, whether the closest regular time stamp is missing. If so, windPRO assigns the

irregular one to the nearest missing one. Is the nearest time stamp not missing, windPRO checks whether the

second nearest time stamp is missing. If so, windPRO will use the irregular one on the position of the second

closes time stamp. If neither the closest nor the second closest regular time stamps are missing, windPRO marks

the irregular time stamp as duplicate. “Abort” will abort the import and no import of the data is going to be done.

The reason is that the inconsistent time stamps must be handled before the analysis proceeds to the import of

the status codes, as the presence of the irregular time stamps in the time series could fully spoil the conversion

of the status logs into the 10-minute time series. If needed, the user can therefore abort the import here and

double check the SCADA data, potentially export the SCADA data once again from the SCADA tool.

It can also happen that there are some duplicates found in the time series, e.g., due to the assignment of the

irregular time stamps as per above or due to the switching between summertime and wintertime, which are not

clearly specified by the TR10. When there are some duplicates found, the user is notified:

All duplicates are deleted automatically and documented in the pdf report. Please select Ok, leave the “Pair and

load” window using “Close” and move directly to the assignment of the status codes to the TR10 categories

Data: Import SCADA 15

using the button “Setup categories”. It is marked with the green color again and you can find it just below the

“Pair and load” button.

13.3.3 Setup categories

13.3.3.1 Basic description

The “Setup categories”-feature (tab “Import SCADA”) helps the user to import the status logs into the TR10 tool

and to merge the status codes in the form “beginning – end”, resp. “beginning + duration” with the production

data in a regular 10-minute format.

13.3.3.2 Concept choice

The user can select from two concept choices when entering the “Setup categories”:

Figure 19 Setup categories - Concept choice

Status signals and/or categories in time series

As already mentioned in chapter 13.2.1, some SCADA systems are capable to export the 10-minute time series

incl. the status codes. Typically, the code represents the first code activated in the 10-minute time series. This

approach is not in compliance with the TR10, as the SCADA system cannot be considered as independent tool.

But on the other hand, it might be the case that the user decides to make the assignment of the status codes

and TR10 categories into the 10-minute time series outside of windPRO in an inhouse tool. In such case this

feature will help. As windPRO cannot guarantee that the assignment of the TR10 categories into the 10-minute

time series was done correctly, the selection of this option will trigger a deviation to the TR10, and it will be

documented in the “Deviations from TR10” pdf report.

In case this option was selected, the only relevant tab from the “Manage categories” setup is the tab “Manage

status signals” – skip to chapter 13.3.3.5.

Separate status signals

This is the foreseen way of a TR10 compliant evaluation and therefore this option is always pre-selected. The

user will import step by step the necessary data. By this, the status logs in the format “beginning – end” and the

status log library can be merged as well as overlapping states can be evaluated and are prioritized. This results

in a regular time series in 10-minute resolution where each of the time stamp is assigned to exactly one TR10

category.

The status logs are separate files logging all operational states of a WTG. The main difference in comparison to

the production data is the format. The production data are in a regular 10-minute format; the status logs are

saved in the format “from – to” or “from + duration”.

The way how to import the status logs in windPRO is similar to the process of the import of the production data

described in previous steps. The first step is to import the file with the status logs (1). When imported, the import

filter must be defined. windPRO needs to understand, in which column to search each of the signals (2). The

next step is to define the format of the date and time. This is not recognized from windPRO automatically but

must be typed manually (3). The symbols for “day – month – year” are “d.m.y”, the symbols for “hour – minute –

second” are “h:m:s”. The WTG ID must be defined here too (4). This is identical to the definition of the WTG ID

during the import of 10-minute SCADA data described in chapter 13.3.1. The option to save / load the import

filter in the *.pci format that was introduced in chapter 13.3.1. is available here, too (5).

Data: Import SCADA 16

Figure 20 Import status logs

As soon as you are done with the import of the status codes, it might be the case that you need to import some

external data that are not part of the SCADA status logs mentioned already in chapter 13.2.2, point 1.

Some of the SCADA systems have an explicite status code for EinsMan and Optimized marketing, some don’t.

If there are such signals available in your analysis, you can fully ignore the tabs EinsMan (6) and Optimized

selling (7); they are deactivated in the default setup. If such explicite status codes are not available, you can

import them from an external source in the format described in chapter 13.2.2.

13.3.3.3 „Einspeisemanagement and Optimized selling“

The default setup is “EinsMan (resp. Optimized selling) included in the status logs”. This means that the events

that belong to these categories have an explicit status code directly in the SCADA system. Shouldn’t this be the

case, you can deactivate the checkbox and upload the data manually from an external source.

Data: Import SCADA 17

Figure 21 Import Einspeisemanagement and / or Optimized selling

The expected format of the data is: “Start date and time → End date and time” and the import filter logics are

identical to the ones of the status logs. The only difference is that the feature must be activated by unchecking

the checkbox “EinsMan (resp. Optimized selling) included in the status logs”.

Some more information about the formats of imported external (= non-SCADA) data are in chapter 13.2.2.

The data imported here are valid for all WTGs and therefore there is no setup of the WTG ID here. As the result,

windPRO will combine the times from the status logs and the times from external sources representing the

optimized selling. All situations imported using the “EinsMan” tab will be asigned to the TR10 category 3 for all

WTGs. And similar for the optimized selling – these situations will be asigned to category 4 for all WTGs.

13.3.3.4 Pairing to the WTG objects

The way how to use the “Pair and load” feature here is identical to the way of use during the import of the 10-

minute SCADA data described in chapter 13.3.2. If the IDs of the existing WTG objects in windPRO and the IDs

used within the status logs are identical, the “Auto pair” will automatically assign them. If this is not the case, you

can assign manually by using the drop-down menus in the rightmost column by each of the WTG.

Figure 22 Merge Status logs with existing WTG objects

Data: Import SCADA 18

13.3.3.5 Manage status signals

The idea of TR10 is that there is going to be a list of all available status codes for each particular WTG and that

each of the status codes is going to be assigned to exactly one TR10 category by an independent body that has

proven to the FGW the necessary prerequisites and qualifications of the authorised persons.

Figure 23 Manage status signals

The aim of the step in this tab is to merge the status logs uploaded in the previous steps with the library / list

(ideally certified), where all of the existing status codes (for the particular WTG type) are assigned to exactly one

TR10 category. The data can be either uploaded from the clipboard (using copy-paste e.g. from a spread sheet)

or imported from an external file (in *.csv or *.txt format). Another option is to use the status signals directly from

the status logs, if they were recorded and saved directly by the SCADA system.

The imports from a file and from the clipboard are very similar to each other and very similar to the import of the

status codes. The first steps are the import of the data itself (1) and the setup of the data types (2).

Figure 24 Import status list from file (or from clipboard)

Data: Import SCADA 19

It might be necessary to apply the “Multiplying factor” (3):

→

Some manufacturers use the so called “Primary and Secondary status signals”. Unfortunately, windPRO cannot

handle such data structure in its native form, as it only can work with single and unique status codes. Therefore,

the “Multiplying factor” was introduced. The idea is to multiply the primary error code with some big value, e.g.

1000. windPRO multiplies then each primary status code by 1000 and adds the secondary status code to this

number. E.g. for a primary status code 12 and secondary status code 2 the result will be 12002, which represents

a unique status code for the further process of the analysis. As soon as you defined explicitly a primary and

secondary code (2), the import setup will not allow you to leave until the multiplier is defined:

When the setup is finished, you can save it in the *.pci format, so you can use it in the future (4).

As soon as everything is settled, confirm the setup with Ok and leave the window. windPRO will inform you about

the amount of the identified status codes in the library:

Confirming this notification with Ok will close the import of the status codes library and will show you the list of

imported status codes:

Data: Import SCADA 20

Figure 25 Imported status list

This list contains all status codes found in the library, where all of them are assigned to exactly one category.

You have the option to change the category of any of the status codes here, but be aware that if you do so, the

category update explanation will be requested, and the change will be documented in the pdf report “Deviations

from TR10”.

It is necessary that all status codes that were recorded during the operation via the status logs find their

equivalent (assigned to the TR10 category) in the overall list that you imported in this step. If this is not the case,

which means if there was an event during the operation with a code that is not present on the list, windPRO will

notify you and will create the respective status code in the list automatically. Such a code will be marked with

“[AUTO]” and will be automatically assigned to the category 2, which means “WTG technically not available”.

The issue will be documented in the pdf report “Deviations from TR10”.

Using the Import / Export buttons you can always save or load saved setups. Similar to import filters for the 10-

minute SCADA data or for the status logs.

13.3.3.6 Load/review

Figure 26 Use “Load” to merge Status logs with time series

windPRO is again suggesting the next step with green, which is also the only possible step at this point – to load

the imported data into the time series with the button “Load”. When done, all status signals from the status log

in the format “from – to” or “from + duration” format are assigned to the regular 10-minute time series and each

of the 10-minute time stamp is assigned to one of the categories according to the TR10.

Data: Import SCADA 21

As soon as the upload is done, you can review all 10-minute time stamps and the status signals assigned to

these time stamps. In the lower part of the window, the overlapping status signals are displayed (if any):

Figure 27 Final time series with status codes and TR10 categories assigned to each 10-min. time stamp

The time frames with potentially overlapping events are handled in compliance with the TR10. The longer one

(in each 10-min time stamp) has a priority and at identical length of two or more events, the event belonging to

category 2 has the priority, followed by 4, 3, 1 and 0.

Manual assignment

The button Manual assignment allows two options. Via “Reset”:

an already executed import of the status signals into the 10-minute time series can be deleted from the tab Time

series SCADA. The second option (by simply clicking at the button, not the arrow on its right side) is the manual

assignment of one or several status signals relevant for one 10-minute time stamp. This can be relevant for

faulty assigned status signals. A manual assignment needs to be documented in the appearing text field and is

documented in the pdf report “Deviations from TR10”.

Data: Import SCADA 22

Figure 28 Manual re-assignment

Status signal frequency

This tab gives the user an overview of the frequencies of the particular status codes. Such overview / analysis

is not required by the TR10 and has therefore just informational value here.

The buttons “Preview” on the right side of the recorded status signal allows you to display the relationship of

each of the particular status signals to the power curve:

Figure 29 Example of a particular status code vs. contractual power curve

Go back to the Load/review tab and when ready, click Ok and all data in the changed structure, which means in

the 10-minute format, will be saved into the “Existing WTG objects”.

Data: Time series SCADA 23

13.4 Data: Time series SCADA

There is no setup option available at this tab at this stage. The imported time series can be checked on this tab,

before proceeding further in the analysis. Typical indicator, that something could have gone wrong during the

import, is a high number of “Out of range” values. Sometimes the reason is an incorrect setup of the import filter

for the import of 10-minute data. E.g. an incorrect unit or decimal separator.

The data can be marked, copied to the clipboard (the use of the button “Copy” is recommended) and pasted into

an external tool, e.g. MS Excel.

Figure 30 Time series with assigned codes and TR10 categories after plausibility checks and processing

of gaps

The button “Visualisation” in the right bottom corner leads the user to the time series in a graphical form:

Data: Import SOLD 24

Figure 31 All signals can be visualized like in a METEO object

13.5 Data: Import SOLD

The sold data are imported as monthly production data. The import of a time series (e.g. Lastgänge

) is not

supported, as this would not be in compliance with TR10.

The first step is to create a new grid connection point:

As soon as created, a new line and the “Load data” button appears in the right window.

Figure 32 Mask for import of sold electricity

The “Load data” button brings the user to a table that is expecting the import of monthly produced electricity in

MWh. You can either manually type the values into the table or use the button “Load data from file or clipboard”.

A time series with production measured at the grid connection point. Normally in 15-minute resolution.

Data: Import SOLD 25

For importing the data from the clipboard or from a file, there is an import mask similar to one used for the import

of the status codes:

Figure 33 Import sold electricity (and eventually EinsMan and Optimized selling)

The assumption is that the source is a table with the relevant columns and monthly resolution. Month & year,

Produced electricity, EinsMan, Optimized selling. See more details about the expected data format in chapter

13.2.2.

It is possible to create more grid connection points when there is more than one grid connection point. The

assignment is then done by the drop-down menus of the WTGs:

As soon as the first grid connection point is selected for one of the WTGs, windPRO offers the option to assign

all WTGs to the same grid connection point automatically:

Data: Plausibility 26

13.6 Data: Plausibility

Before any calculation is available here, all checkboxes need to be checked and necessary descriptions need

to be filled. These declarations are formal requirements defined in the TR10.

Figure 34 Plausibility checks

When this step is finished, please proceed with the button “Calculate time slices”. This calculates the total

portions of each of the EEG 2017 categories during the analyzed time frame. Based on this result, the calculation

of the TR10 availability can be done.

Figure 35 Calculate time slices

When done, the buttons “Missing data”, “Monthly availability” and “Plausibility” become available. The “Missing

data” and “Monthly availability” is not necessary to use. It gives an overview of missing data and about the

availability (which was relevant till revision 2 but has no importance nowadays) respectively.

Data: Plausibility 27

Figure 36 Calculate missing data

The TR10 monthly availability:

Figure 37 Monthly availability

This has just informational value, as the TR10 does not require calculation of availability anymore.

After this step is done, a plausibility check needs to be done. The plausibility check in the TR10 tool fulfills the

check defined in chapter 5.2. It defines, how the so-called SF

Park

shall be calculated, so the electrical losses

can be removed from the 10-minute time series, as described in chapter 5.3. The SF

Park

needs to be calculated

on a monthly basis and is considered as plausible, when being in the range 87% ≤ SF

Park

≤ 109%.

Scaling factor

Site yield: Wind speed correlation 28

Figure 38 Plausibility check of production time series from SCADA vs. sold electricity

Each SF

Park

in this range is marked as “Passed” in the tool and no further action is needed. The SF

Park

-factors

outside the specified range are marked as “failed” and need to be handled.

The user has three options how to handle the failed plausibility checks month by month. No matter which of the

options has been selected, a description describing the reasons for the failed check is required before the user

is allowed to leave the plausibility check window. A failed plausibility check and the respective correction are

documented in the report.

13.7 Site yield: Wind speed correlation

Figure 39 Correlation calculation

The TR10 presumes that as soon as the WTG is not producing or underperforms due to any reason, the wind

measured by the nacelle anemometer is not reliable anymore, as the nacelle transfer function (NTF) is only valid

for normal operation. Therefore, it is necessary to correct wind speeds in non-normal conditions and fill gaps in

the wind speed time series in order to generate a consistent wind speed time series. This is done by the

replacement of all wind speeds of a turbine during events of categories 2, 3 and 4 with a source with the best

correlation under normal operation (category 0). The button “Define normal operation” can be used in order to

Site yield: Wind speed correlation 29

consider the category 1 as normal production too. This can be helpful if the amount of available time stamps for

the calculation of correlations is too low.

Use the button “Calculate” to calculate the correlation of the available sources of wind data among each other

under normal operation (Category 0). The correlation calculation is performed on a monthly basis and sector

wise (with north sector defined as ≥345° - ˂15°) and are always done using 3-months windows. This means that

the month, that the correlation is calculated for, uses the data from that month + the data from the months before

and after this month. E.g. the correlations for July are based on June, July and August. The minimal amount of

samples in a 3-month slot is 8640. If this cannot be achieved in a correlation calculation in a month, the month

selection must be extended by one month before and one month after the currently selected time frame. The

named example time frame would be then extended by May and September. In case the total amount of the

8640 samples is achieved, but the amount of the samples in one of the sectors does not reach 12, then the time

frame is to be extended in the similar logic (= plus one month before and after) just for that particular sector.

Before proceeding with the calculation of the correlation, please note the two setup-buttons on the right side.

Reference database

The TR10 allows to use external sources to create the so-called consistent wind time series. Besides the nacelle

wind speeds from the nearby WTGs, it is allowed to use e.g. data from nearby met masts as well as reanalysis

and / or mesoscale models, too. In windPRO, these are represented via METEO objects. They can be activated

by simply activating the corresponding checkbox of the selected source, resp. by the selected heights:

Figure 40 External meteorological time series

Using the button “Import online data”, it is possible to create new METEO objects from online sources available

at the EMD server.

Setup wind direction master

The WTGs within a wind farm typically show some deviations in the wind direction measurement, resp. in the

nacelle positions. Although the nacelle is correctly positioned towards the wind, the recorded wind direction might

be incorrect. In case the differences in wind directions within the WTGs are too high, it is recommended not to

Site yield: Wind speed correlation 30

use them directly and to define a wind direction master manually. It is possible to select either one of the WTGs

or a METEO object.

Figure 41 Selection of wind direction master

After the reference METEO objects were added (if necessary) and as soon as the wind direction master was

defined (if necessary), the only way how to proceed is to use the button “Calculate”.

When the calculation is finished, you will see a graphical overview in the left side of the window (1). This part

represents a matrix of turbine and reference data, where correlations of all inputs for selected months and sectors

are compared. When you move with the mouse pointer within this area, you will see some details in (2). This

represents the correlation calculated from a source to another source.

Figure 42 Calculated correlations overview

The monthly correlations are displayed in the right part of the window (3). The numbers in the directional sector

columns represent the amount of the months that were used for the calculation of the correlation. 3 months is

the minimum period which has to be applied. This overview might be a bit extensive and therefore there is the

checkbox “Show only expanded” (4). This checkbox is in the default setup activated and assures that all months

Site yield: Wind speed regression 31

with enough data within the standard time-range (8640 samples within 3 months) are not visible. The whole table

can be exported into the clipboard using the button “Export”.

13.8 Site yield: Wind speed regression

After the wind speed correlations calculated for the normal operation (normally only category 0 but can be

expanded by allowing category 1 using the button “Define normal operation” on tab “Wind speed correlation”)

were calculated on the 3-month basis and sector wise, the next step is to use the best correlating sources to

replace the wind speeds of all time stamps belonging to categories 2, 3 and 4 by all WTGs The whole method

is described in the TR10, chapter 6. 1.

After clicking at “Wind speed correction”, windPRO will start replacing wind speeds of time stamps belonging to

categories 2, 3 and 4. As soon as the process is done, the results in form of a small table (1) and a “DNA graph”

(2) are shown. The table refers to the average wind speed of the original wind speed time series, to the average

wind speed of the new (consistent) wind speed time series, difference (in %) between the original and the

consistent average wind speed and in the last column about the amount of replaced samples (in %). There is a

small legend in the right corner (3), explaining the different colors in the “DNA graph”.

Figure 43 Calculation of consistent wind time series

Notice the orange color in the graph, which is used for missing data. The presence of the orange color in the

graph means that not all wind speeds of time stamps belonging to category 2, 3 or 4 could have been replaced,

as there was no source found to replace from. E.g. during an overall data outage, when all analyzed WTGs have

no data and there is no external source to get the data from (reanalysis, mesoscale model). Such situation is not

in compliance with the TR10 and although the user is allowed to continue and to finish the analysis, an

incompliance warning is triggered, and the incompliance is documented in the pdf report “Deviations from TR10”.

Site yield: Moving power curves 32

In order to solve this problem, the simplest solution is to go one step back to the “Wind speed correlations” tab

and to download some online data (reanalysis, mesoscale) from the EMD server or allow there the category 1

as normal operation (if justifiable). The correlations with these sources are typically not perfect but is still better

than an incompliance due to missing data.

There are some visualizing and export features available on the tab “Wind speed regression”:

- Show time series – Visualization of the time series in a graphical form similar to the point 13.4.

- Create METEO object(s) – Creates new METEO object(s) with the consistent wind speed time series.

- Export time series (CSV) – Exports the consistent wind speed time series into a *.csv file.

Export correction information (CSV) – Exports the details about the replacements that have been done

in the time series.

13.9 Site yield: Moving power curves

As the real-life power curve typically differs from the expected / guaranteed and can change in time, the TR10

defines that 3-month moving power curves need to be derived from the SCADA data in order to determine the

not produced electricity. Moving power curves base on all time stamps belonging to the categories 0 and 1 and

are to be used for the calculation of the not produced electricity due to events assigned to the categories 2, 3 or

4. The tab “Moving power curves” contains two functionalities – the button “Calculate” and the check “Air density

correct”. As soon as the button is used, the moving power curves for each of the 3-month slots are calculated.

“Air density correct” is checked in default which means the air density correction is done according to the TR10

chapter 6.2. Unchecking will trigger a deviation from TR10 which will be documented in the pdf report “Deviations

from TR10”.

The power curves are calculated separately for day and night mode, as required in the TR10. The day and night

setup is used from the data entered during the “Pair and load” step – see chapter 13.3.2.

The power curves are calculated for 1m/s wind speeds bins: 0,5m/s ≤ v < 1,5m/s; 1,5 m/s ≤ v < 2,5m/s; etc.

There are some criteria that need to be fulfilled in terms of the amount of available samples. The minimal total

amount of available samples (= categories 0 and 1) is 8.640 (60 days). In case one of the wind speed bins has

less than 6 samples, the power curve shall be interpolated for that particular bin using the bins below and above

the missing one.

Site yield: Moving power curves 33

Figure 44 Calculated power curve(s)

There are some view-setup options on the right side of the tab. None of the changes here has any effect to the

calculation, but only on what is being displayed. The user can select, whether the day or night power curve

should be displayed, etc.

Result 34

13.10 Result

This tab gives an overview of the results so far. Click “Calculate production”. This step will sum up all 10-minute

production values that has been imported from the SCADA system (E

Prod

After E

Prod

has been calculated and the plausibility check has been done, the button “Calculate site yield”

becomes available. As soon as this one is used, you will see further results required by the TR10.

Figure 45 Results (not scaled to 5 years)

- Availability is present in the results as additional information and has no relevance.

- E

Prod_skal.

is the scaled production, resp. the production cleaned by electrical losses (TR10 chapter 5.3.).

- Minderungsfaktor is defined in TR10 in chapter 6.5. It is used to adjust the E_Ausf by defined allowed

outage-hours e.g. due to maintenance (60 hours by 5 years time series, accordingly scaled if the time

series is shorter).

- E

EinsMan

is the electricity not produced due to the constrains by the TSO (category 3).

- E

is the electricity not produced due to the optimized marketing (category 4).

- E

Ausf

is then the fictitious electricity not produced due to the WTG´s unavailability (category 2).

- SE

is the Site Yield and it is calculated as:

o 𝑺𝑬

𝒚,𝑾𝑻𝑮𝒊

= (𝑬

𝑨𝒖𝒔𝒇,𝑾𝑻𝑮

𝒊

+ 𝑬

𝑷𝒓𝒐𝒅_𝒔𝒌𝒂𝒍.,𝑾𝑻𝑮

𝒊

+ 𝑬

𝑬𝒊𝒏𝒔𝑴𝒂𝒏,𝑾𝑻𝑮

𝒊

+ 𝑬

𝑶𝑽,𝑾𝑻𝑮

𝒊

) ∗ 𝟗𝟖%

Quality factor 35

13.11 Quality factor

For the calculation of the final result of the TR10, the quality factor, the WTG-specific Reference yield needs to

be inserted. This can be done either step by step for each of the WTGs separately or in one step, if the Reference

yield of all WTGs is identical.

Figure 46 Results (scaled to 5 years)

The Reference yields can be found on the FGW

homepage, which can be accessed directly from windPRO

using the button “Open URL for reference yields”.

13.12 Report

The final results and formal TR10 requirements, can be exported into pdf reports at the end of the analysis from

the tab “Report”. Besides the TR10 requirements the report additionally aims at documenting the data basis,

data processing, partial results resulting from the analysis procedure and deviations from the required procedure.

Figure 47 Reports tab

Fördergesellschaft Windenergie

Report 36

The format of the report, resp. the formal requirements of the report, are defined in the TR10 in chapter 8. The

button “Report setup” leads to the list of available reports.

Figure 48 Reports setup

The TR10 tool in windPRO follows these requirements and extends it by some extra reports. For example, all

time stamps that were marked as duplicates and / or irregular time stamps and removed are listed in the report

“Duplicates, irregular time stamps and manually disabled data list”. In case of incompliance with the TR10 during

the analysis, this will be reported in the report “Deviations from TR10”.

Figure 49 Main report example