Automatic Modelling of Supermarket Refrigeration Cabinets

New paper shows method for parameterization of the physical properties of refrigeration system cabinets. This enables deeper insight in the systems and can potentially help automating fault diagnostics. The paper is conducted by DTU Compute and TI using refrigeration data supplied by AK-Centralen A/S.

A new paper (Leerbeck, K, Bacher, P & Heerup, C 2021) presenting a method for parameterization of supermarket refrigeration cabinet models has recently been published in The International Conference on Electrical, Computer and Energy Technologies.

The authors show that it is possible to estimate physical properties of a supermarket refrigeration room, e.g., heat capacity of current goods and the heat transfer coefficient of the walls and doors. These parameters can add a deeper layer of information in applications e.g., fault detection and control strategies. The paper focus on one specific evaporator in a refrigeration room, however, for refrigeration cabinets the same method can be applied directly.

The paper focus on the importance of regular re-estimation of the parameters due to changes in the physical system. The most important change is the amount of goods stored in the room, which affects the thermodynamics a lot. For instance, a Saturday night, the amount of goods is very low due to a busy and profitable day. In contrary, a Sunday night, the room is full of goods as new supplies have arrived for the upcoming week.

One problem, however, often occurs when re-estimating the parameters. We can experience the heat transfer coefficient through the doors and walls approaching zero – meaning the room is packed with near infinite insulation. The model still estimates the room temperature correctly, but the parameters identified will not help to represent the real physical system.

Nevertheless, it is shown that parameterization of the physical properties, is achievable, even if a real-life application might need a simpler and more robust model. That is a subject for future research to be decided.

Leerbeck, K, Bacher, P & Heerup, C 2021, Grey Box Modelling of supermarket Refrigeration Room. In Proceedings of the 2021 International Conference on Electrical, Computer and Energy Technologies. European Union, 2021 International Conference on Electrical, Computer and Energy Technologies, Cape Town, South Africa

Grey Box Modelling of supermarket Refrigeration Room - an interview with Kenneth Leerbeck, DTU