Integrating Machine Learning Into Process-Based Modeling to Predict Ammonia Losses From Stored Liquid Dairy Manure - Dissertation

Aug 4, 2026

Upload by Aurelia

Storing manure on dairy farms maximizes fertilizer value, controls costs, and reduces environmental pollution, yet ammonia volatilization during storage continues to cause substantial nitrogen loss. This work addresses limits of process-based models, including uncertain manure heat/mass transfer coefficients and the use of ambient air temperature surrogates that can underestimate emissions. Measurements from three farms with different storage practices supported machine-learning and physics-informed approaches to improve temperature prediction and ammonia-loss estimation. Results highlight stronger flux–temperature alignment, key roles of wind speed and crust thickness, and improved generalization via physics-informed models.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages362

Reading Time15 h 12 min

AccessFree to download

Abstract

Background and problem statement

Data collection and modeling workflow

Modeling methods and inverse coefficient estimation

Key findings and implications

Integrating Machine Learning Into Process-Based Modeling to Predict Ammonia Losses From Stored Liquid Dairy Manure Rana A. Genedy Dissertation submitted to the Faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biological Systems Engineering Jactone A. Ogejo , Chair Matthias Chung Ryan S. Senger Julie E. Shortridge April 25, 2023 Blacksburg, Virginia Keywords: Physics-informed neural networks, Machine and deep learning, Dairy manure, Ammonia emissions, Manure management. Copyright 2023, Rana A. Genedy Integrating Machine Learning Into Process-Based Modeling to Predict Ammonia Losses From Stored Liquid Dairy Manure Rana A. Genedy (ABSTRACT) Storing manure on dairy farms is essential for maximizing its fertilizer value, reducing management costs, and minimizing potential environmental pollution challenges. However, ammonia loss through volatilization during storage remains a challenge. Quantifying these losses is necessary to inform decision-making processes, improve manure management, and design ammonia mitigation strategies. In 2003, the National Research Council recommended using process-based models to estimate emissions of pollutants, such as ammonia, from animal feeding operations. While much progress has been made to meet this call, still, their accuracy is limited because of the inadequate values of manure properties such as heat and mass transfer coeﬀicients. Additionally, the process-based models lack realistic estimations for manure temperatures; they use ambient air temperature surrogates, which was found to underestimate the atmospheric emissions during storage. This study uses machine learning algorithms’ unique abilities to address some of the challenges of process-based modeling. Firstly, ammonia concentrations, manure temperature, and local meteorological factors were measured from three dairy farms with different manure management practices and storage types. This data was used to estimate the influence of manure characteristics and meteorological factors on the trend of ammonia emissions. Secondly, the data was subjected to four data-driven machine learning algorithms and a physics-informed neural network (PINN) to predict manure temperature. Finally, a deep-learning approach that combines processbased modeling and recurrent neural networks (LSTM) was introduced to estimate ammonia loss from dairy manure during storage. This method involves inverse problem-solving to estimate the heat and mass transfer coeﬀicients for ammonia transport and emission from stored manure using the hyperparameters optimization tool, Optuna. Results show that ammonia flux patterns mirrored manure temperature closely compared to ambient air temperature, with wind speed and crust thickness significantly influencing ammonia emissions. The data-driven machine learning models used to estimate the ammonia emissions had a high predictive ability; however, their generalization accuracy was poor. However, the PINN model had superior generalization accuracy with 􀔇 2 during the testing phase exceeded 0 .70, in contrast to-0.03 and 0 .66 for finite-elements heat transfer and data-driven neural network, respectively. In addition, optimizing the process-based model parameters has significantly improved performance. Finally, Physics-informed LSTM has the potential to replace conventional process-based models due to its computational eﬀiciency and does not require extensive data collection. The outcomes of this study contribute to precision agriculture, specifically designing suitable on-farm strategies to minimize nutrient loss and greenhouse gas emissions during the manure storage periods. Integrating Machine Learning Into Process-Based Modeling to Predict Ammonia Losses From Stored Liquid Dairy Manure Rana A. Genedy (GENERAL AUDIENCE ABSTRACT) Dairy farming is critical for meeting the global demand

FAQs

Why are ammonia losses during dairy manure storage difficult to predict with traditional process-based models?

Traditional models rely on manure property estimates like heat and mass transfer coefficients and often use ambient air temperature instead of manure temperature, which can underestimate volatilization during storage.

What data was collected in this study and how was it used?

Ammonia concentrations, manure temperature, and local meteorological factors were measured from three dairy farms. The data helped quantify how manure characteristics and weather conditions influence ammonia emission trends.

Which modeling approaches were used to estimate ammonia loss, and what were the results?

Four data-driven machine learning algorithms plus a physics-informed neural network (PINN) were used to predict manure temperature, and a physics-informed LSTM framework was introduced to estimate ammonia loss. The PINN showed superior generalization, and optimizing process-based parameters improved performance.

Aug 4, 2026

Upload by Aurelia