Welcome Guest 

Show/Hide Header

Welcome Guest, posting in this forum requires registration.

Please note that we - the Docear team - will not answer support questions. If you need help, please read the manual, read the FAQ, or discuss here in the forum with other Docear users.

When reporting bugs, or asking questions, please be as specific as possible. Name your operating system (e.g. Windows 7, 64 Bit), your JAVA version (e.g. Java 6, 32 Bit), whether you are using the latest Docear version, and provide step-by-step instructions that allow to reproduce the bug or problem you are reporting. Feel free to upload screenshots.

Due to spammers and the new European GDPR initiative, we decided to deactivate user registrations. This means, only users with an existing account will be able to post in the forum. We are sorry for this but see no other way.
Pages: [1]
Author Topic: InProceedings: extra period and blank spot
Posts: 4
Post InProceedings: extra period and blank spot
on: May 5, 2017, 11:14

The reference produced in word by docear4word has an extra period and blank spot. Here is the produced reference:

Vogt, S., J. Dobschinski, T. Kanefendt, S. Otterson, and Y.-M. Saint-Drenan. 2015. A Hybrid Physical and Machine Learning Based Forecast of Regional Wind Power. 14th Wind Integration Workshop. . Brussels.

And here is the bibtex:

author = {Stephan Vogt and Jan Dobschinski and Thomas Kanefendt and Scott Otterson and Yves-Marie Saint-Drenan},
title = {A Hybrid Physical and Machine Learning Based Forecast of Regional Wind Power},
booktitle = {14th Wind Integration Workshop},
year = {2015},
month = oct,
abstract = {This Paper deals with the forecast of several wind
power plants in a spatial domain such as Germany. Machine
learning algorithms, in this case artificial neural networks, can
be trained given historical power data and numerical weather
predictions. These models are used with current weather
forecasts to predict the future power feed-in of single wind
power plants.
Instead of machine learning models, a physical model can
predict future feed-ins. This is often inferior because it misses
information about the site and the power plant implicitly
contained in the power measurements. On the other hand,
physical models can forecast when no historical time series
is available.
If power forecasts are required for areas, there are often some
plants for which measurements are not available. Therefore,
it is meaningful to combine machine learning and physical
modeling. Two methods are described that combine the advan-
tages of both approaches. The first method is a very simple
averaging of the time series from each model. The other is
a new algorithm, that averages both models depending upon
the spatial arrangement of the wind turbines in the forecasting
area. The procedures are tested for accuracy of prediction of
the feed-in into German wind power production in the year
Index Terms?regional wind power forecast, hybrid physical
and machine learning model, artificial neural network},
file = {Vogt15HybridPhysMLrgnFrcst.pdf:Vogt15HybridPhysMLrgnFrcst.pdf:PDF},
location = {Brussels},
url = {},

Thank you for your feedback. Please note that we - the Docear team - currently do not have the time to answer support questions. However, we encourage all Docear users to help each other.

Pages: [1]
Mingle Forum by cartpauj
Version: ; Page loaded in: 0.047 seconds.