Grant: Alliance Grants

A refrigerant leak is one of the major contributors to commercial refrigeration breakdowns in supermarkets or grocery stores. Slow leaks could cause frost buildup, temperature problems and failure to maintain temperature levels, resulting in a severe breakdown in normal operations in retail and food industries. On the other hand, industry-standard gas leak detector sensors are expensive to install and have limited capability of square footage coverage for large commercial refrigeration systems. Due to the high variation of liquid receiver levels, slow and periodic refrigerant leaks would remain undetected till they cause major refrigeration outages. Hence, developing an automated smart leak detection system using artificial intelligence (AI) models based on actual operating conditions could significantly reduce the overhead costs associated with system shutdowns and refrigerant fill-ups in grocery stores. In collaboration with the Neelands' Kalder team, the proposed research will develop a robust and automated refrigeration fault detection system, which will directly contribute to Canadian food industries and the broader economy, as well as the improvement of the critical infrastructure we live in. The goal of this project is to analyze data collected from various systems, driving meaningful insights and developing robust AI models for detecting refrigerant leaks in the form of anomalies. The outcome of this project would have a significant environmental impact through net-zero CO2 emission, substantial cost-saving, and, most importantly, reduce human efforts and erroneous leak maintenance and monitoring processes.