Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
The development of a national-level energy assessment tool for the dairy industry
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2022 ASABE Annual International Meeting 2200539.(doi:10.13031/aim.202200539)
Authors: Philip Shine, John Upton, Michael D. Murphy
Keywords: decision support tool, dairy, energy, machine learning, precision agriculture
Abstract. This body of work pertained to the development and utilization of the National Artificial Intelligent Dairy Energy Application (NAIDEA). NAIDEA was developed to integrate macro-level survey information collected on Ireland‘s population of dairy farms with artificial neural network models developed to simulate total, milk cooling, milk harvesting and water heating electricity use using easily attainable farm details. These models were trained using monitored consumption data, milk production, stock data and infrastructural data collected over six years on 74 pasture-based dairy farms, and validated using nested cross-validation. The methodology also employed hyperparameter tuning and multiple variable selection techniques to identify the farm details that maximized prediction accuracy, that could then be collected as part of nationwide farm surveys. NAIDEA provides dairy stakeholders with macro-level dairy energy statistics such as electricity consumption per liter of milk, that can be monitored over time to calculate the effectiveness of changes to government policy. A filtering mechanism was also incorporated to allow users to filter energy statistics according to farm size or the presence of energy technologies such as plate coolers or variable speed drives. This allows the user to calculate energy statistics for specific dairy farm demographics. NAIDEA also calculates a Dairy Energy Rating for each farm. This rating can be used by farmers to identify opportunities to become more energy efficient and identify those farms that could benefit from installing energy efficient technology. NAIDEA was developed using data from typical Irish grass-based milk producers and can only be utilized within the scope of the data used for model training. For example, NAIDEA is not suitable for data relating to rotary, or robotic milking systems, dairy farms with a herd size greater than 388 dairy cows, or farms that operate a confinement based dairy system. Dairy farmers may then utilize the existing Agricultural Energy Optimization Platform to evaluate the cost benefit of specific energy saving projects by quantifying the monetary, energy and environmental impact related to the installation of technologies such as plate coolers, variable speed drives and solar photovoltaic systems, thereby bridging the gap between Ireland‘s dairy farmers and access to decision support.
(Download PDF) (Export to EndNotes)