Filters

2016 - 20202

More filters

2019 - 201912020 - 20211
Reset filters

Settings

Subject: Regression × WGL Institute: ATB ×

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Supervised Machine Learning to Assess Methane Emissions of a Dairy Building with Natural Ventilation
    (Basel : MDPI, 2020) Hempel, Sabrina; Adolphs, Julian; Landwehr, Niels; Willink, Dilya; Janke, David; Amon, Thomas
    A reliable quantification of greenhouse gas emissions is a basis for the development of adequate mitigation measures. Protocols for emission measurements and data analysis approaches to extrapolate to accurate annual emission values are a substantial prerequisite in this context. We systematically analyzed the benefit of supervised machine learning methods to project methane emissions from a naturally ventilated cattle building with a concrete solid floor and manure scraper located in Northern Germany. We took into account approximately 40 weeks of hourly emission measurements and compared model predictions using eight regression approaches, 27 different sampling scenarios and four measures of model accuracy. Data normalization was applied based on median and quartile range. A correlation analysis was performed to evaluate the influence of individual features. This indicated only a very weak linear relation between the methane emission and features that are typically used to predict methane emission values of naturally ventilated barns. It further highlighted the added value of including day-time and squared ambient temperature as features. The error of the predicted emission values was in general below 10%. The results from Gaussian processes, ordinary multilinear regression and neural networks were least robust. More robust results were obtained with multilinear regression with regularization, support vector machines and particularly the ensemble methods gradient boosting and random forest. The latter had the added value to be rather insensitive against the normalization procedure. In the case of multilinear regression, also the removal of not significantly linearly related variables (i.e., keeping only the day-time component) led to robust modeling results. We concluded that measurement protocols with 7 days and six measurement periods can be considered sufficient to model methane emissions from the dairy barn with solid floor with manure scraper, particularly when periods are distributed over the year with a preference for transition periods. Features should be normalized according to median and quartile range and must be carefully selected depending on the modeling approach.
  • Thumbnail Image
    Item
    Non-linear temperature dependency of ammonia and methane emissions from a naturally ventilated dairy barn
    (Amsterdam : Elsevier, 2016) Hempel, Sabrina; Saha, Chayan Kumer; Fiedler, Merike; Berg, Werner; Hansen, Christiane; Amon, Barbara; Amon, Thomas
    Ammonia (NH3) and methane (CH4) emissions from naturally ventilated dairy barns affect the environment and the wellbeing of humans and animals. Our study improves the understanding of the dependency of emission rates on climatic conditions with a particular focus on temperature. Previous investigations of the relation between gas emission and temperature mainly rely on linear regression or correlation analysis. We take up a preceding study presenting a multilinear regressionmodel based onNH3 and CH4 concentration and temperaturemeasurements between 2010 and 2012 in a dairy barn for 360 cows inNorthern Germany.We study scatter plots and non-linear regressionmodels for a subset of these data and show that the linear approximation comes to its limits when large temperature ranges are considered. The functional dependency of the emission rates on temperature differs among the gases. For NH3, the exponential dependency assumed in previous studies was proven. For methane, a parabolic relation was found. The emissions show large daily and annual variations and environmental impact factors like wind and humidity superimpose the temperature dependency but the functional shape in general persists. Complementary to the former insight that high temperature increases emissions, we found that in the case of CH4, also temperatures below 10 C lead to an increase in emissions from ruminal fermentation which is likely to be due to a change in animal activity. The improved prediction of emissions by the novel non-linear model may support more accurate economic and ecological assessments of smart barn concepts.