A Machine Learning Based Analysis of Double Effect Vapor Absorption Refrigeration Cycle based Cold Storage

Abstract

The demand for sustainable solutions in cold storage is being propelled by the increasing need for renewable energy. To meet this demand, solar energy can be seamlessly integrated into cold storage systems. However, in order to ensure efficient utilization of solar energy, accurate predictions of solar data are crucial to address uncertainties and overcome challenges. Various statistical techniques, including regression, SVM regression, and neural network models, can be employed to forecast solar information by leveraging past solar data. These models rely on data from a selected projection model to generate solar energy. The utilization of a parabolic trough collector is an effective method to convert solar energy into heat, resulting in significant fuel savings of approximately 28 units per year, equivalent to 11% of the overall fuel demand. This reduction in fuel consumption not only leads to cost savings but also reduces dependence on unsustainable energy sources. A comparative study evaluating the performance of cold storage systems compared a double effect vapor absorption refrigeration system to a single-effect system. The double-effect system was preferred due to its lower input heat requirement (1815.55 kW < 3023.87 kW) and higher coefficient of performance (COP) value (1.3 > 0.76) in the multi-effect system. These factors indicate enhanced efficiency and optimal functionality, making the double effect system the superior choice for cold storage applications. Additionally, the study also considered cost and emissions implications. The findings demonstrate the feasibility and long-term sustainability of integrating solar energy into cold storage systems. The analysis reveals that renewable energy integration can lead to a reduction of 30% in greenhouse gas emissions. Embracing sustainable energy sources is not only environmentally wise but also crucial in transitioning away from fossil fuel-dependent systems. The study proposes a viable solution by incorporating a solar system into a fuel generator-based cold storage system, offering the potential to address energy-related challenges. By utilizing solar data and efficient technologies like the parabolic trough collector, solar energy can significantly enhance the sustainability of cold storage systems. After a thorough evaluation, the N BEATS model was identified as the most suitable predictive model for the system.