Conference Agenda

Transport refrigeration is in the midst of a transformation towards electrification. Electrification will not only reduce direct vehicle carbon emissions but also mitigate additional indirect carbon emissions by reducing food losses through more consistent regulation of cargo temperature. However, one drawback is that electrified systems run on batteries, which take significantly longer to recharge than fuels. Therefore, accurately predicting the total energy required for trips is becoming increasingly important in order to enable informed decision-making by various stakeholders related to the electrified fleet. The trip of an electric transport refrigeration system (eTRU) can vary based on factors such as weather conditions, controller setpoint and average vehicle speed. Physics-based predictive models of eTRU are indispensable for range forecasting and fleet planning.

The present article introduces a Modelica based simulation methodology for conducting simulations and facilitating informed decision-making related to eTRU. The details of the different assumptions and equations used in developing component models like the (cargo) box, refrigeration system and the controller are provided. A simple ON-OFF controller model with hysteresis is connected to the box and refrigeration system model to simulate instantaneous power consumption of the eTRU. By integrating the instantaneous power consumption over the weather and the route profile of the trip, total energy consumption of the eTRU for the trip is obtained. The total energy consumption of the trip is divided into the precooling portion and the trip portion. It can be observed that using the shore power (grid electricity) for precooling can increase the vehicle trip range by up to 33.1%. A parametric study is conducted on different trip-related parameters, such as door opening frequency and duration, to guide energy-efficient practices for the crew in order to maximize the vehicle trip range. Additionally, a parametric study of control parameters, such as the restart band width of the ON-OFF controller and the box thermal resistance from aging, is conducted to demonstrate applicability of the simulation framework.

The transportation sector is responsible for more than 70% of petroleum consumption in the United States (US). It is currently the largest source of US greenhouse gas (GHG) emissions, accounting for roughly one-third of all domestic emissions. The demand for refrigerated transport has been steadily growing over the past decade. The cold chain has depended on refrigerated transport units primarily running on fossil fuel-based vehicles for the reliable delivery of food, medicines, and perishables, which is critical to sustenance, health, and economic growth. As a result, fleet operators are in the process of evaluating different zero-emission transport refrigeration unit product offerings to ensure they meet their delivery and cost targets.

The current article introduces a methodology to estimate energy consumption by refrigeration systems with integrated power sources like diesel engine or battery. These systems are commonly used to control the temperature of cargo boxes used to transfer temperature sensitive products in trucks, trailers, ships and trains. A data-driven machine learning model is used to calculate the energy consumption of a trip in this tool, as it can meet both the speed and accuracy requirements. Training and test data for the machine learning models are generated by simulating the performance of a diesel-powered unit controlled by an ON-OFF controller with hysteresis, using a physics-based model in the Modelica language. The data is fitted using three different models: Ordinary Least Squares (OLS), Kriging (KRG), and Neural Network (NN). A trade-off study is conducted to compare the accuracy and complexity of the three models, revealing that the KRG model is the most suitable for the application. The R2 score of each of the models on the test data for fuel consumption are 0.96, 0.98, and 0.97, respectively. The symmetric mean absolute percentage error for these models is 18%, 11%, and 18%, respectively. A typical 8-hour delivery in different weather zones is simulated with the model to identify total fuel consumption and total power consumption. The article also demonstrates how the impact of electrification of these systems from diesel engine powered refrigeration systems (DP-RS) to battery powered refrigeration systems (BP-RS) in various regions of the US can be calculated. By using the electric grid information, the emissions required for charging the BP-RS in each zone are calculated. The article concludes with a discussion of the impact of electrification of the fleet on GHG emissions in different zones and provides recommendations for the coefficient of performance.

Electrification of future commercial aircraft is a promising option to reduce the climate impact of the aviation industry. The multitude of electrically driven components in these more electric aircraft generate a significant amount of heat at moderate temperature levels, which place high demands on the thermal management system. Particularly whilst “hot-day take-off”-operation the maximum permitted component temperature can be close to or even lower than the ambient temperature. As a result, heat exchangers in conventional liquid cooling loops become excessively large and deteriorate the overall aircraft performance due to increased mass and drag.
This paper discusses different methods to deal with these challenges and suggests three thermal two-phase cycles using evaporative cooling of the electric components. In a simple pumped multiphase cycle, an evaporator, a condenser, and a pump are used. Compared to a conventional liquid cooling loop, higher heat fluxes can be achieved to decrease the size of the heat exchangers, but the ambient temperature on a hot day take-off may still exceed the permissible component temperature. To tackle that problem, a vapor compression cycle is presented. With this second cycle component temperatures below ambient temperature can be achieved, but the power consumption of the cooling system is increased compared to system one. The third cooling cycle uses heat recovery during cruise conditions and therefore has an even higher efficiency, but the system complexity is highly increased. For the discussed cycles, pentane emerges as the most promising refrigerant due to its favorable thermophysical properties and low environmental impact. A numerical analysis of the drag and mass evoked by the ram air heat exchangers highlights the potential of a vapor compression cycle to cool electrically driven components at low temperature levels. Additionally, a combined cycle with heat recovery during cruise operation may achieve power savings of up to 10% of the evaporative heat flow rate.

During the last decade, the progressive phase-out of high GWP refrigerants has shifted the refrigerant options mainly towards natural or HFOs refrigerants [1].
Under this situation, is of interest to investigate and optimize refrigeration systems, simultaneously on HFO alternatives with feasible GWP values, and other natural refrigerant alternatives. The consolidation of the most mature designs and the investigation of alternative options both demand thermal system simulations in order to analyse the systems from thermodynamic (energy efficiency), and dynamic/control (dynamics, thermal responses) perspectives.
With this purpose, the authors have developed/adapted Modelica models to analyse the road transport refrigeration units. A first section of the paper will be devoted to the calibration of the models using steady-state available experimental information for R452A. After that, the model will be used to confirm its capability to replicate functional or transient scenarios of interest. Finally, the model, adapted for the use of alternative refrigerants (when possible also with some calibration/verification using open literature experimental data or comparative studies), will be used for comparison purposes.
REFERENCES
[1] Sovacool, B.K., Griffiths, S., Kim, J., Bazilian, M., Climate change and industrial F-gases: A critical and systematic review of developments, sociotechnical systems and policy options for reducing synthetic greenhouse gas emissions, Renewable and Sustainable Energy Reviews 141 (2021) 110759

It is important to maintain food at appropriate temperatures at each step of the cold chain to preserve freshness and reduce spoilage as it is transported to the consumer. Short distance Transport Refrigeration Units (TRUs) are key final legs in the transport of goods and produce to local grocery stores from centralized hubs. Traditional solutions for TRUs are internal combustion engine (ICE) vehicles coupled to mechanically driven compressors that operate vapor compression units to condition a trailer to appropriate temperatures. Alternative technologies to de-carbonization the food chain include the electrification of the HVAC equipment via battery power or the conditioning of the space utilizing phase change materials (PCMs). Solar photovoltaic (PV) panels are considered as a drop in addition to improve overall system performance and battery utilization for a total of 4 case studies: baseline electrified system, PCM, electrified with PV, and PCM with PV. Comprehensive vehicle and trailer models are programmed in Dymola with thermal system parameterization, control logic, and overall model structure outlined for each case. It is shown that over the urban dynamometer drive schedule (UDDS) drive cycle that each system appropriately controls the temperature in refrigerated container. Future work will consider the overall reduction in carbon emissions from the de-carbonization of short hop refrigeration.