Heat stress costs the US dairy industry an estimated $1.5 billion annually due to decreases in milk production and reproductive efficiency and, in periods of extreme heat stress, an increase in fatalities. Heat stress can also threaten animal welfare, putting at risk the long-term social acceptability of dairy farming. Despite the efficiencies that accompany high milk yield, high-producing cows in particular present challenges for the dairy farmer in terms of their vulnerability to heat stress. Heat stress is expected to remain a critical issue for the dairy industry in the coming decades as climate change models predict further increases in average temperatures and the frequency of heat waves. Unfortunately, in many of the currently operating dairy barns, cooling systems are designed and controlled without proper sensing and decision-making capabilities, and thus often fail to deliver effective cooling.
This project seeks to solve this inefficiency and address the imminent challenge by developing a control system that measures and predicts the level of heat stress and maintains an optimal microclimate inside a dairy barn to minimize the impact of heat stress. The developed system will largely automate the decisions typically associated with heat-stress management and provide farmers with actionable information while relieving them of the dilemmas created by a surfeit of often contradictory options. More specifically, we propose to combat heat stress in dairy cattle by using a novel engineering system that relies on continuous physiological and micro-environmental sensing, real-time thermal-induced behavior analysis, and computational fluid dynamics (CFD)-based microclimate control.
- Develop sensors that can continuously monitor cows’ physiological and behavioral responses in real time.
- Measure the level of heat stress in real time and adjust the barn’s cooling system and prevent the animals’ heat-induced stress from worsening.
- Perform CFD-based analysis and optimization to find the optimal design and most efficient runtime control.
- Implement and deploy a prototype in UW-Madison dairy barns with different housing and ventilation systems in order to evaluate performance and cost effectiveness.
- Initiate a new direction in the drive to develop a cyber-physical system (CPS) for closed-loop control of dairy barns and advance the use of real-time data analytics and control in other precision-agriculture domains.
- Improve modern dairy barns’ ability to cope with heat stress while also reducing energy and water resource consumption and enhancing the welfare of dairy cattle.
- Raise public awareness of sustainable farming and animal welfare.
This research is sponsored by National Science Foundation (NSF) and USDA National Institute of Food and Agriculture (NIFA).
Younghyun Kim, Lead PI
Department of Electrical and Computer Engineering
College of Engineering
Role: Sensing and machine learning