KOBRA
https://kobra.uni-kassel.de:443
The KOBRA digital repository system captures, stores, indexes, preserves, and distributes digital research material.2020-05-23T19:32:58ZUsing CFD Modelling to Relate Pig Lying Locations to Environmental Variability in Finishing Pens
https://kobra.uni-kassel.de:443/handle/123456789/11568
The purpose of this research was to determine which environmental factors within the pen space differ between the locations where pigs choose to lie and areas they avoid. Data on external weather engconditions and the construction parameters for an existing commercial finishing pig building were input into a Dynamic Thermal (DT) model generating heat flow and surface temperature patterns in the structure and these were then input into a Computational Fluid Dynamics (CFD) model to generate data on the theoretical spatial patterns of temperature and air velocity within one room of this building on a specified day. The exact location of each pig in six selected pens within this room was taken from images from ceiling-mounted video cameras at four representative time points across the day. Using extracted air velocity and temperature data at the height of 0.64 m above the floor and a grid of approximately 600 mm to create a series of ‘cells’, the effective draught temperature (TED) was calculated from the models for each cell. Using a sequential regression model, the extent to which the actual lying locations of the pigs could be reliably predicted from the environmental conditions generated by model outputs and other pen factors was explored. The results showed that air velocity, TED and proximity to a solid pen partition (all significant at P < 0.05) had significant predictive value and collectively explained 15.55% of the total explained deviation of 17.13%. When the presence of an adjacent pig was considered, results showed that lying next to an adjacent pig, TED, air velocity and temperature accounted for 53.9%, 1.3%, 1.5% and 0.5% of the deviation in lying patterns, respectively (all P < 0.001). Thus, CFD model outputs could potentially provide the industry with a better understanding of which environmental drivers affect pigs’ lying location choice, even before a building is built and stocked.
2020-03-03T00:00:00ZJackson, PaulNasirahmadi, AbozarGuy, Jonathan H.Bull, SteveAvery, Peter J.Edwards, Sandra A.Sturm, BarbaraThe purpose of this research was to determine which environmental factors within the pen space differ between the locations where pigs choose to lie and areas they avoid. Data on external weather engconditions and the construction parameters for an existing commercial finishing pig building were input into a Dynamic Thermal (DT) model generating heat flow and surface temperature patterns in the structure and these were then input into a Computational Fluid Dynamics (CFD) model to generate data on the theoretical spatial patterns of temperature and air velocity within one room of this building on a specified day. The exact location of each pig in six selected pens within this room was taken from images from ceiling-mounted video cameras at four representative time points across the day. Using extracted air velocity and temperature data at the height of 0.64 m above the floor and a grid of approximately 600 mm to create a series of ‘cells’, the effective draught temperature (TED) was calculated from the models for each cell. Using a sequential regression model, the extent to which the actual lying locations of the pigs could be reliably predicted from the environmental conditions generated by model outputs and other pen factors was explored. The results showed that air velocity, TED and proximity to a solid pen partition (all significant at P < 0.05) had significant predictive value and collectively explained 15.55% of the total explained deviation of 17.13%. When the presence of an adjacent pig was considered, results showed that lying next to an adjacent pig, TED, air velocity and temperature accounted for 53.9%, 1.3%, 1.5% and 0.5% of the deviation in lying patterns, respectively (all P < 0.001). Thus, CFD model outputs could potentially provide the industry with a better understanding of which environmental drivers affect pigs’ lying location choice, even before a building is built and stocked.Comparison Between Two Hydrodynamic Models in Simulating Physical Processes of a Reservoir with Complex Morphology: Maroon Reservoir
https://kobra.uni-kassel.de:443/handle/123456789/11567
Two 3D hydrodynamic models, AEM3D and MIKE3, are compared in simulating hydrodynamics of the Maroon Reservoir in southwest Iran. The reservoir has a complex bathymetry with steep walls, which makes it a good case for studying the performance of hydrodynamic models. The models were compared together and with measured water temperatures from different locations of the reservoir in a five-month period between December 2011 and April 2012. The results indicated that the AEM3D model, which uses a finite difference scheme with a purely z-level vertical discretization, showed better consistency with observations so that the AME and RMSE of the model remain below 1 °C. The MIKE3 model showed overall higher errors from 56% to 130% larger than AEM3D and the level of error strongly depends on its vertical discretization method and the turbulence model. The lowest errors by MIKE3 were seen by the k-ε turbulence model with a hybrid z-sigma discretization, while the highest errors were generated by using the sigma vertical discretization. The vertical mixing model in AEM3D model, used instead of the constant eddy viscosity or k-ε formulation, showed a better performance in modeling vertical mixing and wind mixed layer, which is another reason of observing better results by this model than MIKE3. Overall, this study shows AEM3D as a more appropriate model for simulating deep and complex reservoirs with steep slopes and walls.
2020-03-14T00:00:00ZZamani, BehnamKoch, ManfredTwo 3D hydrodynamic models, AEM3D and MIKE3, are compared in simulating hydrodynamics of the Maroon Reservoir in southwest Iran. The reservoir has a complex bathymetry with steep walls, which makes it a good case for studying the performance of hydrodynamic models. The models were compared together and with measured water temperatures from different locations of the reservoir in a five-month period between December 2011 and April 2012. The results indicated that the AEM3D model, which uses a finite difference scheme with a purely z-level vertical discretization, showed better consistency with observations so that the AME and RMSE of the model remain below 1 °C. The MIKE3 model showed overall higher errors from 56% to 130% larger than AEM3D and the level of error strongly depends on its vertical discretization method and the turbulence model. The lowest errors by MIKE3 were seen by the k-ε turbulence model with a hybrid z-sigma discretization, while the highest errors were generated by using the sigma vertical discretization. The vertical mixing model in AEM3D model, used instead of the constant eddy viscosity or k-ε formulation, showed a better performance in modeling vertical mixing and wind mixed layer, which is another reason of observing better results by this model than MIKE3. Overall, this study shows AEM3D as a more appropriate model for simulating deep and complex reservoirs with steep slopes and walls.Probabilistic upscaling and aggregation of wind power forecasts
https://kobra.uni-kassel.de:443/handle/123456789/11566
Background
Wind power forecasts of the expected wind feed-in for the next hours or days are necessary to integrate the generated volatile wind energy into power systems. Most forecasting models predict in some sense the best value, but they ignore the other possible outcomes which may arise because of forecasting uncertainties. Probabilistic forecasts, on the other hand, also predict a distribution of possible outcomes with their respective probabilities that specific power values will occur and therefore have higher information content. In this work, we address two problems that hinder the introduction of probabilistic forecasts in practice: (1) no measurement data are available for some wind farms and (2) the flexible aggregation of probabilistic forecasts for changing wind farm portfolios.
Methods
We present an approach based on copulas that can solve both problems by modeling the spatial correlation structure between reference wind farms. By sampling from the resulting joint probability distribution, probability forecasts can be upscaled from reference wind farms to wind farms without power measurements. Furthermore, the results can be aggregated to probability forecasts of portfolios of arbitrary and changing size.
Results
We perform experiments by applying our procedure to three use cases. The results are quantitatively evaluated with different probabilistic scores. For single target wind farms, our approach is as good as a state-of-the-art reference even if no data are available for the wind farm under consideration. For portfolios, our approach also allows forecasts to be made if no data is available for some wind farms and also to aggregate flexible portfolios of changing sizes, which was not possible before.
Conclusion
Our work solves two problems that hindered the introduction of quantile probabilistic forecasts in the application. This work opens a pathway for many different applications, e.g., predictive grid securities with stochastic optimization, better marketing of renewable energies, or allow to compensate for forecast errors in various applications.
2020-03-16T00:00:00ZHenze, JanoschSiefert, MalteBremicker-Trübelhorn, SaschaAsanalieva, NazgulSick, BernhardBackground
Wind power forecasts of the expected wind feed-in for the next hours or days are necessary to integrate the generated volatile wind energy into power systems. Most forecasting models predict in some sense the best value, but they ignore the other possible outcomes which may arise because of forecasting uncertainties. Probabilistic forecasts, on the other hand, also predict a distribution of possible outcomes with their respective probabilities that specific power values will occur and therefore have higher information content. In this work, we address two problems that hinder the introduction of probabilistic forecasts in practice: (1) no measurement data are available for some wind farms and (2) the flexible aggregation of probabilistic forecasts for changing wind farm portfolios.
Methods
We present an approach based on copulas that can solve both problems by modeling the spatial correlation structure between reference wind farms. By sampling from the resulting joint probability distribution, probability forecasts can be upscaled from reference wind farms to wind farms without power measurements. Furthermore, the results can be aggregated to probability forecasts of portfolios of arbitrary and changing size.
Results
We perform experiments by applying our procedure to three use cases. The results are quantitatively evaluated with different probabilistic scores. For single target wind farms, our approach is as good as a state-of-the-art reference even if no data are available for the wind farm under consideration. For portfolios, our approach also allows forecasts to be made if no data is available for some wind farms and also to aggregate flexible portfolios of changing sizes, which was not possible before.
Conclusion
Our work solves two problems that hindered the introduction of quantile probabilistic forecasts in the application. This work opens a pathway for many different applications, e.g., predictive grid securities with stochastic optimization, better marketing of renewable energies, or allow to compensate for forecast errors in various applications.Comparing the effects of generating questions, testing, and restudying on students' long‐term recall in university learning
https://kobra.uni-kassel.de:443/handle/123456789/11565
We compared the long‐term effects of generating questions by learners with answering questions (i.e., testing) and restudying in the context of a university lecture. In contrast to previous studies, students were not prepared for the learning strategies, learning content was experimentally controlled, and effects on factual and transfer knowledge were examined. Students' overall recall performance after one week profited from generating questions and testing but not from restudying. When analyzing the effects on both knowledge types separately, traditional analyses revealed that only factual knowledge appeared to benefit from testing. However, additional Bayesian analyses suggested that generating questions and testing similarly benefit factual and transfer knowledge compared with restudying. The generation of questions thus seems to be another powerful learning strategy, yielding similar effects as testing on long‐term retention of coherent learning content in educational contexts, and these effects emerge for factual and transfer knowledge.
2020-01-21T00:00:00ZEbersbach, MirjamFeierabend, MaikeNazari, Katharina Barzagar B.We compared the long‐term effects of generating questions by learners with answering questions (i.e., testing) and restudying in the context of a university lecture. In contrast to previous studies, students were not prepared for the learning strategies, learning content was experimentally controlled, and effects on factual and transfer knowledge were examined. Students' overall recall performance after one week profited from generating questions and testing but not from restudying. When analyzing the effects on both knowledge types separately, traditional analyses revealed that only factual knowledge appeared to benefit from testing. However, additional Bayesian analyses suggested that generating questions and testing similarly benefit factual and transfer knowledge compared with restudying. The generation of questions thus seems to be another powerful learning strategy, yielding similar effects as testing on long‐term retention of coherent learning content in educational contexts, and these effects emerge for factual and transfer knowledge.